9/30/2021: Math is complete!
My brain has been marinating in mathematics all month long. I really love mathematics. This project requires quite a bit of signal processing math, and a good chunk of this math is not anything you’ll find in existing math software libraries. To get this done, I returned to my older system that still runs Windows XP and is not connected to the internet. My life’s work in mathematics is all on this machine. It was good to return to it, and I must say it was good to work on a computer that isn’t connected to the net. Even though having instant access to all of human knowledge can be very useful, it’s also very distracting. I may detach more often.
Anyway, here is the current status: Starting this week I’ll be revising the EMpick circuit schematic with the few changes I found were needed. I’ll also be revising the PCB to match the schematic. This is the necessary step before ordering a few boards that should represent the final, shippable version of the EMpick. While the boards are being fabricated, I’ll update the plastic enclosure – just a bit to do there, and then I’ll order a 3D printed enclosure to make sure everything fits and that the battery compartment access works as it should.
What I’m quietly saying is that we are right now entering the finalization stage of this project.
Here are a few pics from math land:
8/30/2021: Much progress..
Well, I’m firing on all 5 computer screens now! At the moment I’m deeply into revising the math to adjust for the new control system. Everything is similar but different. For math I use MathCad, an old version from before they were bought by Parametric Technologies. Amazingly, math remains valid even without the latest updates! The math is on those two screens that are bright and washed out in the picture.
Here are the items that I must complete before we can have the first working EMpicks:
- Once the math is revised, the math code needs to be revised. This is made much easier by getting the math right before coding it.
- Demonstrate that I have control over individual harmonics and groups of harmonics. (Like the Vo-96 but better.)
- Connect (2) to the buttons on the EMpick to make basic features user-accessible.
- Revise the electronics and the PCB. There are a few small changes that need to be made before I can make boards in any quantity.
- Update the Solidworks mechanical enclosure design. Not much to do here, but the small loudspeaker must be moved over a bit.
- Order small quantity of components and assemble a few ‘real’ EMpicks. Make videos. Test, test, test. Revise if necessary.
- Make a larger quantity of EMpicks and ship them to supporters who have been waiting the longest.
- Make more, get into advertising it again, ship EMpicks to everyone who has pre-ordered, and then to new customers.
- Take a vacation.
This is the basic outline. Everything needs to be tweaked a bit, but all of these items are already 95% ready to go. Firmware remains the biggest task.
This is the list of what needs to be done. How long will all this take? I don’t know. I never do, but so far I always do finish everything and ship my products.
Please stay well everyone. I’m fully vaccinated with the Pfizer and the experience has given me no reason to hesitate in recommending these vaccinations for everyone. One more data point for your consideration.
7/21/2021: The Breakthrough!
I grabbed my iPhone to capture this moment. This is the control system that solves everything:
It has been a long while since I posted anything to this channel. And an even longer while that I’ve been working on the EMpick. There were external delays, new rules around lithium ion batteries that caused me to radically change the power structure design, and then the tariff on electronic parts. But it’s fair to say the longest delay has been due to the fact that I’m a very stubborn inventor. I could not “just go with” a lesser solution just to get the product finished, not when I could see the right solution in my imagination. I must say, dragging that solution out into reality has been a real struggle. But it’s here, real, working, and I wanted to mark this moment with a short video demo “proof of concept.”
I know some of you will point out that the Ebow will do this, so wtf am I talking about? The difference is that this is not a sustainer, it is a controller. This difference can be experienced with the Moog Guitar, the Vo-96 and the Wond. But those past controllers had flaws, restrictions that limited their acceptance, such as the need to use special all-steel strings instead of the much favored nickel wound or nickel plated “softer” strings. And such strings are objectively better, with a nicer feel, better pitch definition and a sweeter tone.
This new controller will work perfectly with nickel or any other type of string that has magnetic response. It is a coherent, laser-like device. It also has at least four times the power and “authority” over the string than my other controllers, which are themselves much more powerful than any sustainer. With this controller I can dial in any combination of harmonics of string vibration, any tone, with the flexibility of a keyboard synthesizer. This goes beyond just sustaining a note, but the greater difference is that all this is happening on the strings themselves. The strings make the sound. Everything can be heard acoustically as well as electrically. This controller opens the door to full, real-time synthesis of actual string vibration. There are no restrictions. This means I can pick up where the Vo-96 left off and bring the full power of acoustic synthesis into reality.
New things take time, and it’s also true that new things have to wait sometimes for the right time, a time when people are ready to move to something new. Thank you, supporters, for being patient while I complete this obsession and give you the EMpick. It’s not here yet, but this moment is where I “got unstuck!”
– Paul Vo
7/11/2021: Not now!
Hello everyone, I can’t post right now. I’ll post the update sometime between now and the end of this month. Please come back then.
6/10/2021: Place holder...
I’m living and breathing the EMpick, but here it is the 10th of June and I’m not yet able to show the core technology working. I’m close – I just have to write a little bit of very fast running code in ARM assembly language and this EMpick should open its eyes and sing an actual note or two.
I mean, look, most product developers have one or two computer screens in front of them. I have 5, and each one is showing me a different aspect of the EMpick. I’m stitching this together into a whole. Whoever has the most screens gets it done, right?
5/10/2021: A good couple of weeks!
Taking that break showed me that I need to modify how I’m using time. I’ve always had this ‘nose to the grindstone’ thing and it’s helped me get a lot done in the past. However, this worked because just about every day there was something else that needed my attention. I’d be outside a few times a week. I’d be talking with friends, or attending someone’s gig — something. Now I see that Covid took away these “distractions” and left me with just get up in the morning, eat something, walk ten feet and sit down to work on the EMpick, and that was it. And in retrospect I think I worked less effectively this way than I did when I had more different things that needed doing.
Anyway, as I said the last time, once I’d booked my trip I was able to break through the coding impasse and make some progress. And when I returned, I made even more progress. Good work on the code right up to and including today. In a few days I’m going to visit my family for a bit — a weekend at least, perhaps a bit longer. Everyone has had their immunizations now. It has been almost two years since I’ve seen them all, so I’m looking forward to this. And in the meantime, I’ll keep coding.
That’s it for now, if you are still reading.
– Paul Vo
4/10/2021: I need a break!
Well, so one day after booking a few days away, it was like the clouds cleared and I was able to get past this difficulty I was having with the firmware. Tells me something.
I’m in a rut. I’ve been trapped here for a year and I’ve been doing exactly the same thing every day. Yes, I’m making progress. However, the truth is, my output is slowing and I’m procrastinating. I’ve had both my vax shots. I really need a break. I need to do something totally different for a few days, and I need to be somewhere other than here for a few days. I’m thinking I’d really like to drive to the beach and walk alongside the ocean. I’d like to eat some food I haven’t cooked myself. All that.
For a few days now I’ve been stalled on a very difficult bit of the design. I keep going around in circles with it. I think if I can get away for a bit and try not to think about it, I’ll be able to come back to it with my brain freshly reset and solve this puzzle.
I know this isn’t what everyone wants to hear, but it’s the way things actually are. At this moment I have to admit it’s not productive to keep being here and doing just this.
Here is what has been in my brain:
I need some of this for a few days:
3/10/2021: Can I spend a solid month writing firmware?
Yes I can! And incredibly, when I do this I end up with more lines of code than I started with!
I’m getting closer to making it all work, but until it does there is not much to say and not much to show. It’s just the way it is right now.
Tomorrow I get my first shot of one of the two shot vaccines. Hopefully I won’t notice a thing.
in the meantime… more coding. I’m coding to two processers right now. The EMpick has a small fast one that is perfect for driving all the timing pulses and monitoring the user input switches, etc. The bigger one is fast too, but it’s best for doing the math and telling the first processor what to do. It also has a Bluetooth transceiver, which will be for many things but initially will be for updating the firmware.
Here is an infra-red shot of the EMpick circuit. It runs very cool; the color makes it look toasty but it’s only a few degrees C.
2/10/2021: Empick Universe
I’m trying not to come up for air. Here are a few pics of the EMpick universe. It’s going reasonably well, though it is very deep and consuming. Still, it’s moving forward, slowly…
1/4/2021: Happy New Year...
I know, that was 4 days ago, but it’s still new, right? I don’t want to stop working on the EMpick right now to put a post together. Things are moving with some momentum. I’ll post as soon as I hit a plateau. Or a mesa. Butte not later than a week or so from today. This is just to let everyone know things are going well in EMpick land.
– Paul Vo
11/20/2020: It's here! And it fits!
Yesterday, (Friday 20-11-20), I finally received these prototype boards. They look good. I’m about to connect to one of these and methodically test all the functions. It’s not just this assembly order that has taken a lot longer than it was supposed to. Some of my clients are having the same experience, and these are not all with the same supplier.
The top image is the EMpick PCB by itself, and the bottom image is the same PCB within one half of the enclosure I designed last year. It does fit, so that’s one thing that is correct. In a few days, basically as of December, I’ll be able to work on this and nothing but this looking forward.
11/11/2020: New circuit board STILL OUT FOR FABRICATION and ASSEMBLY.
I’ve been delaying this update in the hope that I’d receive the prototype boards and be able to share pics and optimism. But I feel I need to update everyone even though the update is kind of crazy. It has been 9 weeks and I learned today that the assembly house has not yet received the raw PCBs. The assembler is in the USA but the PCB service is in China. I will post pics just as soon as I get the boards and I’ll update this post as soon as I get definite info about what the trouble is. They told me almost a month ago they’d be shipping in a few days and then there was an error and a delay. I’ve used this company before and they were solid.
10/3/2020: New circuit board STILL out for fabrication and assembly.
What a frustrating month! It’s like the general insanity is seeping in where it really doesn’t belong. Board houses always offer relatively good sounding turn around times, but the reality is that you’ll have to have a back and forth with them about this and that, and each such volley can burn through a couple of days. But this month, well, just various annoying problems popping up in whack a mole fashion, mostly to do with other mundane aspects of life. Nevertheless, I’ve mostly been attending to the firmware in prep for the arrival of these few EMpick circuits. I’ve nothing to show for these efforts visually, but I do have the core framework set up. It’s all straight C and very ‘real time’, no cycles lost to the convenience of an RTOS.
RTOS? In a general purpose computer such as is in your phone or Mac or PC, there is an operating system such as Windows or whatever Apple is calling their OS these days that schedules and activates myriad services that all share the various cores out between possibly hundreds of ‘threads’ within a modern CPU. The particular thing you think you are doing with your computer may not even be the activity that is using the majority of your computer’s processing power. In systems that are more dedicated to a specific use such as in some appliances, medical equipment like an MRI or X-ray machine, the various chips in your automobile and the robots in factories you will find an RTOS, or Real-Time Operating System. This is a different kind of operating system that is optimized to guarantee prioritized ‘on-time’ execution of tasks. The response time of an RTOS may be a hundred times faster than say Windows 10, but the CPU or more likely MPU or SOC (system on a chip) is still jumping around doing a bit of this task, setting it aside and doing a bit of a different task, then a third task and then back to the first task just in time to deliver the result on schedule, etc. But there is ‘real-time’ and there is REAL-TIME…
The EMpick has two processors. One is very simple and very fast. The essential core of the vibration controller runs within this chip, which also generates all the timing signals that go out to the analog hardware to perform the motion control function. This is all about the precise control of magnetic fields and this control must be continuous. You can’t “pause” a field and get back to it later because this isn’t virtual, it’s physical. So this chip needs no RTOS. It does basically one thing and it does it with nanosecond speed. Then there is another chip that will handle user input and less time-sensitive things such as waveform analysis and the generation of control commands – sort of like the prefrontal cortex in our brains. In fact, this chip is an ARM variant called a Cortex. It is intimately coupled with a Bluetooth transceiver. Initially BT will be used only for firmware updates, but hey, it’s a transceiver, a wireless wire; an invisible USB connection. This means that the capabilities of the EMpick are open-ended, even to the eventual possibility of interfacing directly with your DAW. But that’s in the possible future. I’m planning on opening this all up so it’s not just me adding such functions but anyone else who has a need or an interest. It may not be that useful in the context of the EMpick, but I’m designing the architecture so that it can be expanded to any number of strings. Piano, for example? Sure. It would be expensive, but amazing.
At this moment though, the new EMpick hardware is not yet real. I’m hoping the boards arrive next week. They ought to. Please, October, no more obstacles!
Images from last month:
The final circuit board ready for prototyping:
The way it was a few weeks ago:
At the start:
The PCB will sit within the EMpick enclosure; you can probably see how it fits
I hope everyone is well. There is so much drama in the world right now, but musicians are finding ways to get back to performing. Outdoor concerts with a well-distanced audience, lots of online music. I keep hearing new music that blows me away. Currently I’m enjoying “Larkin Poe”. And yes, the EMpick is EMerging…
- Paul Vo
9/3/2020: New circuit board out for fabrication and assembly.
EMpick hardware is ready! It has a string driver/transducer, a microphone, a small loudspeaker, lots of digital signal processing power, and a Bluetooth radio for data and firmware updates.
And the EMpick has one more thing. A trick thing. I don’t want to describe this yet because I need to prove it works. It works great in my brain. Possibly it will work in reality. If it doesn’t it will be because I haven’t yet understood a deep technical point quite correctly. But I think I have.
These layouts are weird. You zoom in and work all day on a section. Then you zoom out, and the whole day’s work shrinks down almost to nothing. Getting this done has always been grueling this way. At least it ends up as PCB ‘artwork’. And I revised the enclosure design at the same time as I worked on this, so the revisions to that are almost complete now too.
I’m going to not look at a computer screen this Saturday.
The EMpick now has two brains. One is sparse but very fast and tireless. It handles all the doing. The other is dense and deep. It will be doing the thinking. Both these chips come from the same company. I’m pondering opening up this project, making it open-source except for the essential core that embodies the collocated vibration controller. This core is the invention, it is the new thing that has not previously existed in any form. By itself the core controller does nothing, but governed by firmware it can make a guitar string do just about anything. Or a fat bass string. Or anything else that responds to magnetism and generates sound. So if I open up firmware development to all who code, this could exceed what I can imagine.
As a practical matter, the tools for writing such firmware are inexpensive, much less than the cost of the EMpick itself, so anyone capable of coding and interested in making the EMpick do something as yet unimagined could dive in and do it. I think I’ll do this, but first things first…
The EMpick PCB artwork is going out Monday morning. The PCB will be fabricated, the components will be placed and soldered, all to be done by a service that specializes in such prototype assemblies. Meanwhile, I’ll be returning to writing firmware. A certain basic set of firmware is required to test the hardware. All the fun stuff arises from the firmware.
The final circuit board ready for prototyping:
The way it was a few weeks ago:
At the start:
The PCB will sit within the EMpick enclosure; you can probably see how it fits
I hope everyone is well. There is so much drama in the world right now, but musicians are finding ways to get back to performing. Outdoor concerts with a well-distanced audience, lots of online music. I keep hearing new music that blows me away. Currently I’m enjoying “Larkin Poe”. And yes, the EMpick is EMerging…
- Paul Vo
8/1/2020: Finishing the Revisions.
If you’ve been following this blog you know this is the third version of the EMpick that I’m working on. Each one has been a different and incrementally better design. This version represents entirely new work and I think overcomes all of the difficulties of the previous designs. The battery thing is taken care of using the solution from version 2, and this version ought to be less expensive to produce, because with the tariffs I would have had to charge over $300 for the version 2 design to have the project be viable and sustainable from a business point of view.
Anyway, I thought I’d try to convey the amount of work involved. The following pictures may or may not convey this, I don’t know. The most time consuming thing about doing something like this is the simple but overwhelmingly significant fact that when I get stuck, there is no cheat sheet or existing reference to turn to. I have to invent my way along as I go and this often means going deeply through several potential solutions before finding one that works.
Once the schematic is finished it’s time to translate it into a physical circuit board layout that expresses the abstract schematic in physical terms. Design tools help with this by transferring the connections of the schematic onto something often called a “rats nest” as follows:
Next, I find some good music to listen to and methodically move all these components around into a logical positioning that fits within the area of the EMpick circuit board. This isn’t just a matter of untangling this mess. It is also necessary to think about field interactions; some things need to be distant from other things while some things must be together. Everything also has to comport with the design rules supplied by circuit board manufacturers. Components can only be so close to one another and the same goes for the copper traces. I’m close to finishing this version 3 PCB. Here is what it looks like today:
Once I’m finished, (about another week from now), it will look something like the version 2 EMpick PCB, which is displayed below for comparison:
Once I get the current design to this completed state I’ll be sending out the database to get a few PCBs manufactured and populated by components. These will be shipped back to me and then I can begin testing.
I’ve never before had to go through three full and different versions of a product before completing it. Perhaps I’ve just been lucky for a long time, but over the years I’ve nailed it on the first attempt several times and always by the second attempt. This project was supposed to be relatively “quick” but it has turned into one of the most difficult things I’ve ever worked on.
I’m very excited about this, even though it’s been difficult. This design has an entirely new collocated controller, also my third such thing. The first version powered the Moog Guitar and a fundamentally similar but more sophisticated version became the Vo-96. This type of controller works very well as long as the strings are not too thick, (it would not work for bass strings), and provided the strings were made of steel and not any other magnetic material such as nickel. Of course, most guitarists prefer nickel plated or pure nickel wound strings so having to use special strings and not being able to drive bass strings were significant drawbacks. The Wond was based on a second type of collocated vibration controller which is a very strange thing. This controller fires exponential impulses at the string that quickly become increasingly coherent but are always rather jagged and fierce sounding when they impinge on a pickup. Still, this tech has promise and the Wond does deliver some very unusual sounds. This controller handled fat strings and different materials better, but not with perfection. When I started with the EMpick I had ideas about how to use digital signal processing to improve this stochastic wavelet controller but this proved to be an elusive goal. The current EMpick features a third kind of collocated controller, one I’m sure I would not have come up with without having worked on a product for the medical electronics industry. I became involved with this because it also employs precise control of magnetic fields, (basically my thing), but it made me thing about everything from an entirely different direction. I guess all this bounced around in my subconscious for a while and combined with my musical work to emerge as the current collocated vibration controller for the EMpick. Tests and simulation indicate that this controller finally overcomes all the obstacles. Any type of electric guitar or bass string should work, as well as any other vibrating tines or blocks such as in the marimba, etc. Any size, any geometry, any magnetically responsive material. It will also reach through the phosphor bronze wrap of acoustic strings to couple to the inner steel core without distortion of the control force.
This is now taking shape. It will go into the same plastic housing I developed for the previous EMpick design, though a few small changes must be made to this too:
So, what can I tell you about when this will be done? I know I’m working as fast as I can and as fast as is consistent with doing a good job. I know it’s getting closer. I hope nothing else happens in the external world to hold me back. And I know I’ll finish this project. And I’m pretty sure it will have been worth it. However, I know many of you have been waiting an unreasonable length of time for this to be done. I am trying to make sure it is worth the wait.
Things in the world remain crazy. It’s like the world fell off a cliff in January. I guess we won’t return to the way things were, but I’m sure things will get better. People are finding new ways to enjoy music, and musicians, well nothing is going to stop musicians from making music. These days will be prominent in history, so there is a certain sense of awe and vigilant awareness that seems appropriate. And there is just plain surviving, not getting the virus and being able to buy food and pay rents and mortgages and such, all made much more difficult and challenging than it was just a few months ago. Nevertheless, we go forward. Stay well everyone, and we’ll get through this!
- Paul Vo
6/30/2020: Making the Revisions.
I’ve revised the EMpick schematic and now I’m making the revisions to the PCB. This won’t be the final revision, but it will test two critical innovations that work in simulation. I need to see these work in reality.
I can’t believe June has flown by so quickly. I’m trying to not be overcome by the impact of the events that began on May 25th with the death of George Floyd. This certainly looked like a slow and methodical murder to me. I don’t want to get into this here, I’m sure everyone is well aware of it and of all that has followed. It has affected me. Sometimes I wonder if what I’m doing is still meaningful, or even sensible. Sometimes it seems what I’m doing is the only sensible thing I can be doing. It has certainly been an unhappy and worrying month.
I’ve received some very encouraging messages from supporters and some heart-warming feedback from a long-time owner and player of one of the Vo-96 acoustic guitars I put together back in 2014. Thank you! This pandemic has been incredibly hard on musicians, who were already struggling with poorly paying gigs and next-to-zero payments from streaming services. So much has changed so quickly. Will it all come back? This is a difficult time for so many reasons, but I can still get lost in my work, at least for a while each day. It’s true that the project is once again moving forward, and I’m looking forward to doing physical tests and to having something to show, tell and listen to. And then final revisions, and then on to making many and finally, shipping EMpicks to you. (Rest assured, the EMpick will work for online performances too.)
I must finish these circuit revisions and send out for a couple of test boards. The software, ( it’s firmware, I’m just going to assume everyone knows that code that goes into some kind of “thing that does stuff” is generally called ‘firmware’, whereas code that is coming and going out of computer memory many times a day depending on which program you are using is called ‘software’), the firmware to test these innovations is ready. I just need the hardware. It’s all just ware. 🙂
So this is where things are at this moment. I hope the world gets better. I hope the music returns. And I hope everyone stays safe and well!
- Paul Vo
5/27/2020: Made it through!
I just want to let everyone know that the frustrating technical challenge I referred to in the previous blog entry is now behind me, and behind us. Forward motion has resumed!
5/18/2020: Working on it!
The current project status:
It would be so nice if time would just stop for a while. I’m working on the EMpick, and if I stand back and look, I am making progress. However, for the past two weeks I’ve been stuck in a frustrating hall of mirrors otherwise known as the peripheral configurations possible in the Cortex M33-based MPU that is the new heart of the EMpick. This is a great chip, nice price, not tied completely to China and therefore less of a worry geopolitically, (what a thing to have to consider!), it has an FPU with DSP extensions, and it has wonderfully capable peripheral subsystems — all the stuff the EMpick needs from a bluetooth radio to PWM drivers for delivering power to the EMpick coils. I need various complex things to be going on in the background and this chip is definitely capable of meeting all these requirements. But configuring it to do so has been a nightmare. There is a whack-a-mole effect. You do one thing, and this other thing changes slightly, or it wants to happen at any instant other than the instant you need it to happen, or must use a pin that you need for something else. The manufacturer, (Silicon Labs), is working on a tool called the “Hardware Configurator”. On a project for a client I have used a version they have that supports a simpler chip they make. It’s a great tool, very helpful in configuring the chip. They have been working on the Hardware Configurator for this more advanced and complex chip for longer than I’ve been working on the EMpick, and they aren’t yet giving a date when it will be available. It’s just software BTW. It’s a software tool that helps one to configure their hardware peripherals. They’ll probably finish it long after I’ve done it all by hand.
Once I’ve sorted this out it will just be there in the background automatically doing its magic tasks, and I can resume the work of bringing over all the code previously written for the TI chip into this new Cortex beast. This also is tedious, but it’s far more straightforward than the peripheral configuration has been. Everything else, (PCB design, the plastic, the battery-based power supply, the transducers), is in a nearly complete state except for this MPU, but everything needs a few small revisions to be completely ready. The PCB needs the most revision as this chip is replacing the older one. But it’s not a bad place to be, overall. I just need to keep at it.
Back at the end of last year I gave a forecast for May, but added something like, “unless there is some catastrophe”. And here we are. My response to it was to do all the consulting work I had lined up on a full time basis to get it done while my clients were still able to pay me. That worked out, and since April 15th or so I’ve been 100% back on the EMpick. So I’m determined to keep people informed, but reluctant to give a solid date, and I’m not saying anything anymore about aliens landing or other unexpected world-changing obstacles.
The fact is, there is nothing within the EMpick that has been done before. It’s all new, patentable territory. Even the control system isn’t the one I’ve used in the Vo-96 and the Wond. It’s much better, but it’s also completely new. This makes for slow going even without the world falling apart. Frankly though, the biggest factor of all is that I’ve need to earn money as a consultant to buy the time I need to work on the EMpick.
I am feeling a bit stir-crazy. The siege is being lifted, but not realistically for me as I’m in the vulnerable age group. I hope everyone is staying safe. I don’t know what the long term answer is here. I really don’t want to ever get this virus. I’m all for herd immunity, but I can’t be part of the herd. I guess everyone has to measure their own risk and act accordingly. Fortunately, I don’t have to be out and about to make a living. If this were not so, I’d be in a much worse state of mind.
Okay, until the next blog entry, please be well, thank you for your patience and support, and please be safe.
– Paul Vo
4/18/2020: Not Going Viral!
I’m finally finished my ‘keep the lights on’ work. This weekend I’m going to put a bunch of that stuff away and restore the lab setup I need for working on the EMpick.
From this weekend on, I have no distractions or consulting jobs to do. A long unbroken stretch of time lies ahead, and I’ll be isolated here in the lab and working on the EMpick. This is the first time things have lined up this way since this project began, and a perfect opportunity to bring it to completion.
The lab should look like this. Currently it’s a mess.
…and yes, this is indeed “The Masked Inventor”, saying please stay safe and well, and thank you.
- Paul Vo
4/04/2020: Hey, where did the world go?
I ended the previous blog entry as follows:
“Okay, I’m publishing this and going back to the code. At some point I need to update the web site and attend to other such matters. But not now, not while I’m in zombie programmer mode.”
Right, well I did emerge from zombie coder trance. I looked around.
I’m sure you don’t need me to reprise what has happened and how much the world has changed in a very short span of time.
You are here to read about the status of EMpick development. Some of you have signed on very recently, and I am deeply appreciative of this because at a time like this, it’s not just that you’d like to have an EMpick, it’s that you believe, and are willing to put money on it, that the world will re-normalize and allow me to finally bring this device fully out of my imagination and into reality. Some of you signed on a long time ago. It’s into 3 years ago. I thank you so much for your patience. There have been external events that have caused me to radically change aspects of the design. But it’s also my fault in that I’m an inventor, and it is very difficult for me to go forward with an idea when I’ve come up with a much better idea. And there is an additional reason: I’m not in the least bit wealthy. I work for a living. In 2017 I managed to be frugal enough to save enough to let me work on the EMpick, (called the Wond II back then), for a straight run of about 9 months. It wasn’t long enough, and other things happened. You can go back in the blog and see, though I didn’t discuss my personal situation back then of course. I wouldn’t normally do so, but I’m not normally 2 years behind in completing something. A few months yes, but not 2 years. This brings us to today, to this moment:
Once I recovered from the shock of realizing the magnitude of this crisis, I also realized that thankfully I wasn’t in the worst possible situation to weather it. I’ve had two client projects on the go. I’m finished with one for now. (This is the medical device project, which I feel quite good about.) The newer and current project is something I can’t say anything about, except that it is closer to the music industry. I’m just days away from completing the first phase of this, and I have been told that the next phase is on indefinite hold, as I expected and as is happening everywhere in the economy at every level. My work is solitary anyway, has been for years. I’m well-stocked with food. My modest but rewarding “social life” has gone entirely virtual. (The Internet better stay operational, or there really will be chaos.) Anyway, once I complete this phase of the “paying” project, I’ll be free to concentrate 100% of my work time on the EMpick, and secure in knowing I can keep on this way for some months.
I haven’t done any work on the EMpick since the last blog post. I was going to be splitting my time up between the EMpick and this consulting engineering work, but prudent self-protection took over and I’ve been working to finish and be paid for the consulting work while this is still possible — because who knows what comes next?
So this is the status: I’m just days away from completing the first phase of the newer project. It is all code at this point, with some simple hardware I built long ago. I’m polishing up the code, (refactoring – sort of.) It all works. This low-level coding has really helped my software chops. I’m sticking with C. I love C, I think I know it thoroughly again now, and today’s C compilers turn out assembly code that is very close to optimal. The EMpick code is running on an arm Cortex processor with FPU and DSP extensions, (and integrated Bluetooth among other essential peripherals). It’s an amazing chip at an amazing value and the C compiler is superb. In this development environment I can wrap the C code and move up to C++ any time I want, if I need a higher level of abstraction. Possibly I’ll do this for the user interface and other higher level but slower stuff, some of which is ported from earlier EMpick work in C++. The EMpick today is quite different under the hood, different and significantly better.
Just writing about it here makes me want to get back to it right away.
Regarding May and beta: As I wrote to a customer/friend the other day: Nothing will stop me so long as I’m healthy and able to do it. Caveat: My crystal ball is darker than it has ever been. I don’t know what is ahead. And even if the EMpick were ready for beta at this moment, I could not do it because beta depends on services that are not currently dependable, (I’m not in position to send money and components to a company that could close down the next day), and it also depends on local friends getting together with me part time to help, which isn’t currently safe or practical.
It’s not like I’m going to run out of things to do on the EMpick, that’s for sure. My plan going forward has been to (1) make sure firmware updates operate easily and reliably, and (2) deliver the EMpick with solid hardware and a basic set of behaviors, (the beta EMpicks), and then continue to have it manufactured and continue to add to the features and fix bugs and change things according to user feedback. Where the line is drawn on the beta features is therefore flexible. Beta will happen when it becomes practical and possible.
I really do have a one-track mind when it comes to projects. It’s not that multitasking is more difficult for me than it is for others. It’s that being responsible for everything on a product from idea to shipping is already quite enough multitasking. The EMpick will be better with no other projects to juggle.
Meanwhile, I’m also growing my hair long.
Please stay well, everyone. Really, be careful out there.
- Paul Vo
0/07/2020: In the flow writing the software.
I’ve had analog hardware connected to the DSP micro-controller for over 3 weeks now. I’ve been in a kind of trance for the last 3 weeks, zero time on Facebook or anywhere else. I really need to stay in this not-unpleasant hyper-focused state, because there is a lot of code to write and port from past work and sometimes rewrite when the ported code doesn’t make sense in this new architectural context.
It’s going well. When I hit some kind of natural place to pause, I’ll write a more complete update. Meanwhile, if I can stay in this zone, I’ll get a lot done quickly.
I guess a byproduct of working like this is it’s like being in quarantine, which helps me avoid the Covid-19 virus. I hope everyone stays free of this virus. Wash hands, etc., you know the drill. Covid-19 is certainly being taken seriously, I don’t ever remember SXSW ever being cancelled before, not to mention Musikmesse, also cancelled for 2020. Panic is never called for, but cancelling huge events such as this costs the organizers and participants millions — not something that’s done if it’s not truly necessary.
Okay, I’m publishing this and going back to the code. At some point I need to update the web site and attend to other such matters. But not now, not while I’m in zombie programmer mode.
- Paul Vo
02/02/2020: A great start to the year!
Alright, I’ve taken this Sunday evening to do all the tasks that are not core development tasks. It’s not that these tasks aren’t important. Keeping you aware and updated on how things are going with the EMpick is very important. It’s that when things are going well I’m “in the flow” as they say, and it requires an effort of will to step out of that flow to do anything else. I’m sure you must know the feeling. I would not be surprised to learn that nothing tangible ever gets done without being and staying in the flow.
02/02/2020 — one of those numerically unusual dates, anyone notice?
I’m quite comfortable with this new software development tool set now. This coming month I’ll be connecting actual hardware, (some of it hand constructed but still hardware), to this software running on a development platform. Strictly speaking some of this is not new software, it’s ported over from the DSP microcontroller I was using in rev 2 of the design before the tariffs came in. But most of the core code that implements the harmonic control system is brand new because this current code supports my newest collocated controller and it is quite different than before.
So this will be a significant month. On the one hand, I’m retracing steps I’ve mastered before. On the other hand, this time I’m doing it with much more powerful and capable controller, one that opens up some new possibilities. I can’t wait to get to those new possibilities, but first I have to get the project back to where it was before events vaporized my financial model. Then it’s onward toward a working physical model I can hold in my hand and demonstrate, and it is on to building features (which is all software), and refining the product.
It would perhaps be nice to not have to worry at all about cost, but this would be irresponsible and ineffective. The EMpick simply must be affordable and its intended price of $249.oo is already on the high side. It’s up to me to make it good enough to justify this price, especially in light of the other hand-held devices out there such as the venerable Ebow. This is a challenge.
And also, even though you, (and over 400 others), have preordered your EMpick, if all I did at this point was deliver these orders, I’d be well below zero into the nasty negative bucks overall. So it simply must be appealing enough to guitarists, (and to anyone playing anything made of magnetically responsive metal from the kalimba to steel drums), to generate a lot more sales. I’m hoping to find 10,000 musicians to buy it. We’ll see. No pressure. 🙂
Can you tell I’ve also balanced my books, paid my bills and sent out a couple of invoices this evening?
After I’ve climbed back to were I was BT*, things will get interesting.
|||||| May 2020 ||||||
Thank you all for your patience and support.
- Paul Vo.
* [ “BT” – Before Tariffs, or Before Trump. ]
December 15, 2019: Preparing for the push, year end stuff.
I just realized I haven’t posted anything in 6 weeks or so. It’s been about this long I’ve been fighting a virus, some kind of common cold. This hasn’t prevented me from working but has slowed me down. I’m grateful that I very seldom get anything like this. It’s here in Asheville though. I hope it’s not everywhere. Here I’ve met a few people with the same complaint. Seems to last about 6 weeks, a really long time for a chest cold that isn’t flu. But I’m not the only one here with it, so that is what it is.
Ah well, such is life. Despite this nasty thing, I’ve set everything up for the push to beta on the EMpick and climbed the ramp on these new development tools. I must say, I am really enjoying getting back to writing my own software. I should have done this earlier. I kept talking myself out of it. It is very consuming of mental energy. But it gives me all the tools I need to finish the EMpick.
I’ve cleared the decks this month of all but one consulting project. I must do some hours of consulting work to “keep the lights on”, but now I can give a lot of time to the EMpick. Hopefully nothing will get in the way between now and May.
I’m taking about 4 days over the holidays to spend with family. I hope everyone can have a break. I guess whatever else this time of year means for you, it seems universal that some time with family is part of it.
So: Happy Holidays! (Not quite that time yet, but I probably won’t post again until 2020.) Happy new year too!
- Paul Vo.
November 3, 2019: The new Empick schedule, and its background.
First, the current progress:
Since the last post mid-October, I’ve sorted out being able to use a 1.5 volt AA battery to power the EMpick. I’ll probably ship EMpicks with an ordinary alkaline 1.5 volt AA battery just so it will work right out of the box, but I will strongly recommend everyone purchase a li-ion 1.5 volt rechargeable battery. I’ve tested some of these now and they are really amazingly good. Getting up to the voltages needed by the EMpick is a challenge, but I have this worked out and ready to rev into the design. If you can get a 3.7 volt li-ion AA cell, (not available everywhere), it will last even longer in the EMpick.
It really looks like this kind of battery is here to stay. More and more brands are appearing. I have yet to see such a battery offered by one of the very largest and well-known brands. These brands’ business model depends on batteries being disposable. I imagine there is much hand-wringing in those boardrooms about a 1.5 volt AA battery that is inexpensive, lasts pretty much as long on a charge as a regular alkaline AA, puts out a very stable voltage that doesn’t sag until the battery is completely discharged, and will probably last for several years.
As for data and firmware updates: I’m designing in something I can hardly believe exists: A relatively inexpensive Bluetooth AND IEEE 802-11 wireless module. It has great development tools too. So: no wires!
So, when can this all be done?
I’ve written a schedule, up to the point of the BETA run. Drum roll:
This is realistic. If I could work on the EMpick full-time it could be faster but I must take on a certain amount of consulting work to keep this ship afloat. I’ve factored this into the schedule.
How many BETA EMpicks will be made? Quite a few supporters have expressed an interest in being BETA testers. Some don’t want to, on the theory I guess that the real EMpicks will be the ones from the production run, and maybe the BETA units will have flaws that will be corrected in production EMpicks. This is possible though it is much more likely that it is the software, (firmware), that will change between BETA and production due to feedback I receive from BETA testers. Everyone will receive the same firmware updates. We’ll see how it goes. Once I have working BETA EMpicks I’ll begin advertising again. Production will follow BETA on a schedule that will depend on the lead times of production sub-assembly suppliers. This is 90 days, on average.
One reason the EMpick design is taking so long is I’m designing it to be produced by a turn-key manufacturing service. It has to go together perfectly, predictably, and with no time-consuming kludges like having to glue them together as I did with the Wond. With a part-time staff of two, I made and installed all the Vo-96s. I hand-made every Wond on my own. I will personally build the BETA EMpicks but afterwards it must go on autopilot. A manufacturing service will make them, test them and ship them. If I get it right, I may continue to make them this way over and over. Who knows, this may generate enough revenue for me to fund the development of some of my other music product ideas. We’ll see how it goes.
The Background Picture:
I don’t want to build and run my own manufacturing business. I never have. When I headed up the design lab at QSC and Yorkville, those 20 years showed me what an endlessly difficult task being a manufacturer is. To start my own traditional product manufacturing company would leave no time for what I really want to keep doing. I’m an inventor and a product developer and a musician. This is what I need do.
The purpose the Vo-96 Kickstarter and the Wond was to bring to life some of my ideas so that established music product manufacturers could see that they were real and viable and marketable and desired by musicians. My door has been open, but so far not one such company has reached out to me. Perhaps the EMpick will change this because it is going to be a great product and I hope will sell well at an affordable price point.
This is why for some time now one of my background tasks has been to look for a company, a manufacturer that would work with me to develop a line of products based upon a fundamentally new paradigm for sound synthesis. “A Fundamentally New Paradigm for Sound Synthesis.” Sounds like any other hyper-inflated marketing pitch, doesn’t it? But it’s real, and it could be the basis for a new branch of the music products business and a new way for musicians to make music, just as electronic synthesis began 70 years ago and has grown into a whole new category of musical instruments.
And yes, this is the world’s best string sustainer technology. That’s the least of it’s capabilities, though I know “sustain” is mostly what many guitarists want today. But back when the first electronic synths came out in the 1960’s, did keyboard players know that it would still be a big and still growing thing in 2019? Did they know they’d one day want a “Moog One” for $7999? No, because they didn’t yet know what anyone could do with such instruments. Such an evolution takes time. It takes pioneer musicians like Wendy Carlos and Keith Emerson and Isao Tomita.
You’d think though that some forward-looking company would see the possibilities inherent in doing sound synthesis physically. After all, musical instruments are physical things. So I’m looking. Looking…
Over the past year I’ve slowly put together a page on this website that is directed mostly toward finding such a forward-thinking company. I invite you to take a look at it, and please feel free to share this link with anyone you know who may be interested or may know someone who is interested. Here is the link:
And yes: May 2020. The EMpick.
- Paul Vo.
October 16th, 2019: Planning the EMpick: ... continued
…continued from the previous post…
Other considerations influencing this decision.
- Since there is now a replaceable externally charged battery, a USB cable is no longer needed for charging. It therefore makes more sense to use BlueTooth and be wireless. Sending a file to a device via BT is now well supported in the OS of most devices. BT is quite easy to implement at the device level these days.
- A good deal of the work is already done. The enclosure needs only to have the USB port removed from the design prior to tooling.
- My plans for the microphone remain the same. It’s really basically a vocoder implementation but using harmonic controlled guitar strings and using any input sound, not just voice. I have some other ideas for the mic too.
- The EMpick will still have a small speaker that when turned on will emit the re synthesized tones at a volume useful in purely acoustic settings. The same tones will be “injected” into the pickups when using the EMpick with an electric guitar and will become part of the guitar’s output signal.
The battery still presents a problem. The AA size, (14500 metric size), lithium ion battery at 3.7 volts and about 800 ma/hrs is going away. the reason is that there are so many devices out there designed for the same size battery, but the usual alkaline batteries provide 1.5 volts. Dropping a 3.7 volt lithium ion AA battery into such devices burns out their circuits and can even be a fire hazard. To address this, the battery industry is doing a pretty smart thing: They are building a lithium ion rechargeable battery in the AA and AAA sizes, but they are also adding a tiny voltage converter/regulator into each such battery. This results in a rechargeable lithium ion standard-sized cell that puts out a steady 1.5 volts, (unlike other rechargeable batteries such as NICAD.) This makes a rechargeable battery that lasts almost as long on a charge as a standard non-rechargeable battery, provides a more steady voltage that doesn’t sag as the battery becomes discharged, and can be recharged hundred of times. It looks like this solution will be here to stay. It’s a real improvement over the old 1.5 volt cells and these new batteries can be used wherever you’d previously used ordinary non-rechargeable batteries.
To us, this means that I must make the EMpick work on 1.5 volts. That’s a bit tricky, but can do, and it is the new core technology that makes it possible as the peak current requirement is much lower. I will maintain compatibility with batteries up to 4 volts, so whatever happens out there in battery land, you’ll always be able to find a battery for your EMpick.
… more to come …
I’ve been test driving the Analog Devices code tools and so far they are easier to deal with than TI’s code tools, as hoped and expected.
Back in 2012 I designed a Bluetooth module into the Vo-96 and it was easy then, but quite expensive and limited. Well, today there are far more BT modules to choose from and the prices are much more reasonable. Even better, most of these modules have a decent general purpose processor and a fairly large memory map. They just drop into a design and work. It really does make sense to go to Bluetooth in the EMpick.
I have a bunch of housekeeping to do. The existing code is mostly in C and C++ so it can go forward. The great thing is, working with higher level coding tools should make things go much faster. It’s like the difference between getting on a bike and riding it to a store, versus having to first build the bike and then ride it to the store. It’s a big difference.
All this is made possible by the new core technology that fell into my head when I wasn’t looking.
Running on 1.5 volts to 4.2 volts so users can run the EMpick on any kind of battery that fits in the compartment, this is the next “detail” I’m going to tackle…
- Paul Vo
October 7th, 2019: Planning the EMpick: Features and Decisions
I’m posting this now, but it is not complete. Please check back here next week for the complete version:
I have a new and much superior core technology. I want to pilot this in the EMpick. The EMpick up until now was using my second generation core technology, a digital version of the engine in the original Wond.
Both my first and second generation technologies can be implemented in analog. The first generation was pioneered in my all-analog design of the the electronics in the Moog Guitar. Then followed the Vo-96, a digital version of the same technology. Going digital was the only way to release the full potential of controlling string vibration in detail down to individual harmonics. With my gen 1 core technology however, the DSP computing engine had to be programmed in low level assembly to run fast enough for real-time control of 96 harmonics across 6 strings. The same situation holds true for the digital version of my second generation technology. This is what was going into the EMpick, and to develop this I had to bring in an engineer who specializes in just the kind of speed-optimized DSP needed for the EMpick. And although I have nothing but respect and accolades for this engineer, (who asks not to be named), there is still the reality that the work he does with me is more of a hobby that must frequently yield to the demands of his larger clients in aerospace.
This is one of several reasons why it makes sense to move to my latest core technology. This new tech works differently and does not need a super fast DSP programmed in assembly language. This means I can turn to Analog Devices and use a SHARC chip for which much higher level programming tools are available. Firmware written this way runs somewhat more slowly but still quite fast enough. I can also make use of a visual programming library for standard functions such as Fourier and wavelet transforms, convolvers and other filters, etc. The payoff here it’s much faster and considerably easier to create firmware code this way.
… more to come …
September 30th, 2019: Planning the EMpick
I’m posting this now, but it is not complete. Please check back here next week for the complete version:
This project has gone on too long. We’ve had two setbacks due to external events — the lithium ion battery change and the “tariff redesign”. But these aren’t the only reasons the EMpick is taking so long. There is an internal reason that is amplified by these external delays: It’s the fact that I’m an inventor. And when a project like this has to be redesigned, I’m likely to get better ideas about how to do things than I had before.
I can’t un-think a better idea once it’s known to me. I can’t be enthusiastic about finishing a product “the way it is” when I know how to make it significantly better. And I can’t muster the persistent focus needed to do this work unless I’m enthusiastic about what I’m doing.
All these factors have combined to make this ridiculous. I can’t blame you if you have lost patience, and I’m very thankful that so many supporters have been willing to hang in there. But it’s time for me to draw the line. I need to finish and ship these EMpicks to you.
How can I get you your EMpick as quickly as possible? This is the question.
I’ve been considering dividing the EMpick into two stages, one hardware and one software. It’s software, (firmware actually), that is currently consuming my time.
I’m posting this much tonight. I’m still working on the options I want to present to you. If I’m going to publish shipping dates I want to make sure they are realistic, which requires pricing and planning from now through the first EMpick production. This is what I’ll be completing over the next week or so. As soon as I am ready with what will be two or possibly three options for you to consider, I’ll post here, so please check back here in a week.
Aug 20th, 2019: Quick note
Okay, really, I just can’t stop to write about my progress at this moment. It’s going well, and soon I’ll be able to show something. Right now it’s just squiggles and code and I just need to keep at it.
I do have a request though: Are you a supporter living in a country other than the USA? If so, please would you search online to see if you can buy the following battery in your country:
Lithium-ion 3.7 volts 14500 rechargeable battery
If this battery is not available for purchase in your country, please email me at firstname.lastname@example.org and let me know.
Please note that many batteries of this 14500 size are made to have an output voltage of 1.5 volts and these will not work in the EMpick. For the EMpick, we will need the 3.7 volt rechargeable lithium ion battery.
The 14500 size of battery is very similar to the ‘AA’ size battery. You can also search for:
Lithium-ion 3.7 volts AA rechargeable battery
Thank you very much!
– Paul Vo
July 14th, 2019: Hard work and details
Well things are back on track but this doesn’t mean it’s easy. I’m happy now with the quieter modulation scheme. Next is the scripting language.
The Vo-96 was not very flexible. Most of its behaviors were hard-coded and could not be defined by the user. (Adjusted yes, defined, no.) The EMpick allows musicians to define their own sounds or to feed in any sound you can hear, via a wave file or through it’s microphone. The EMpick will analyse the recorded sound or wave file into its harmonics and drive and control the string to reproduce the sound by making the string’s harmonics follow the sound’s harmonics. Easy to say, well sort of, but it requires quite a bit of DSP.
Really all of this ought to be called “Physical Synthesis” rather than “Acoustic Synthesis”, because the synthesis is performed upon the physical instrument string rather than mathematically within an electronic circuit.
With the EMpick it will also be possible to define sequences of harmonics or timbres. This will be similar to what the Vo-96 does with its modulators, except while the Vo-96 has a limited selection of pre-programmed sequences, the EMpick will allow the creation of custom sequences by the user.
A notable challenge in designing features like this is to make it easy for the musician to change from one EMpick behavior to another. There are so many different ways this could be handled. Which way will be the easiest to remember and to use? How will the current setting be displayed? (You will want to know what it will do before doing it.) Instead of answering these questions myself, hard-coding them and having to live with all my wrong guesses, in the EMpick I’m adding a mapping layer between the action, the response and the display. This will make it possible to define the user interface behavior via a text file rather than by re-writing parts of the software. It will even be possible, within certain limits, for users of the EMpick to customize the user interface.
It is so much easier to write about all of this than it is to design the software architecture to make it possible. This is what I’ve been working on recently. I didn’t have this level of user interface flexibility in the software definition for the first version of the EMpick I was working on last year. I’ve had time to think about it and to do some use-case experiments and I think no one will be able to judge which user interface actions work well for the musician until one can actually use the EMpick. The solution is to make user interface behavior easy to redefine at run time.
For the beta run I’ll ship with something that works and then listen carefully to your feedback. There is a growing list of supporters who want to be beta testers, but every user will be able to change a few things on their own to suit their own preference.
Thank you, everyone, for your patience. I still do have something to show everyone and I’ll be looking for your feedback on it. And I still need to make a short video to show it. This task just wasn’t on the logical path forward for these past few weeks.
- Paul Vo.
June 11th, 2019: Everything back on track.
This past month I’ve been working on the modulation scheme. The EMpick’s string driving system, although novel for being simultaneously bi-directional in that it both senses and drives the string, has much in common with class-D amplifiers. It is important for both regulatory and performance reasons to minimize and ideally eliminate any energy emissions of the drive signal that are not in the audible band, i.e., 20Hz to 20KHz. This has been quite a challenge, but I’ve found a great way to do this. My previous solution wasn’t really working in all cases, but this one is solid. I’d love to blog about it, but this method should be added to my next patent before I do.
I’m pretty sure the subconscious mind can do real work in the background, perhaps while we are asleep. It’s not free or effortless. It’s first necessary to consciously put in the work by identifying all the factors involved and really trying to come up with a solution. But once you notice your thoughts circling back to the same unworkable ideas, it’s time to stop and become occupied with something else. Give it a few days and then return to the problem. You may see the solution right away. Or it may come to you sometime when you aren’t expecting it. Suddenly you’ll notice you are thinking about it, and there it is!
Now I’m able to complete the design. This is all so radically different than before that I must make what we call a “mule”, a quick single piece prototype to prove everything out. There will almost certainly be corrections needed after this, but then we can move forward again towards BETA. Several supporters have already volunteered to participate in a beta test.
I’m back on track this month and feeling like I actually know what I’m doing again. It had been a bit like the Mr. Bill Show around here. Does anyone remember that Saturday Night Live puppet skit from the 1970’s? It was weirdly gruesome, but weirdly funny. Mr. Bill suffers various… setbacks. Anyway, things are back on track now.
Thank you, everyone, for your patience. I do have something to show everyone and I’ll be looking for your feedback on it. I need to make a short video. Stay tuned!
- Paul Vo.
May 10th, 2019: Slow Progress and business realities...
It’s May 9th now. I’m sorry to have taken so long to write an update. Some of you have messaged me asking what is going on with the EMpick. Here is the full story for you and everyone. The quick news is that the EMpick remains “on”, but if you want to know more, I’m sharing all the background here. It’s a bit long…
Tax prep in April created a “free” opportunity to review the metrics of my business model. My conclusion? Things are not going according to plan, and if I keep doing the same things the same way, there is no reason to think it’s going to get better.
The best way to speed up and finish the EMpick would be to spend more money. An assistant engineer able to take on some of the load would really help. A webmaster handling web site updates would be a great help. A bigger budget would help in many ways.
My original strategy with the EMpick was to spend 2017 as a robot and get it done quickly. This involved first putting into place the web site, the financial and order tracking, and the “integrations” by which these tasks are performed automatically over the internet for me. Automating the “business” side of my business freed me to get deeply into my reliable single-minded get-the-job-done groove where I do my best work and finish things. This held together very well for a few months. By the end of 2017 the task of maintaining and updating these “automatic” systems began to take it’s share of my time. I could still see a way forward though.
It’s a lot easier to innovate in two phases. During the first phase you don’t worry about how expensive it will be to manufacture your innovation. The goal is to get it working. Once you get it to work, you take it through a second phase of design where you concentrate on refining it and reducing what it costs to manufacture. Often you can make it work even better as you reduce its cost. This two cycle method is best, but each cycle costs time and money. It is sometimes possible to combine the innovation cycle and the cost engineering cycle into a single cycle. This is what I tried to do with the 2017 EMpick design, and I came close, close enough I decided. By mid 2018 I was going ahead with a design I thought would work as I intended, but would cost somewhat more to manufacture than I’d originally planned.
I could see a way forward despite this. I’d hire some friends as I’ve done before, and we’d hand-assemble, test and ship out the first few hundred EMpicks ourselves. This would be both faster and less expensive than the set-up fees involved in turning assembly and shipping over to an assembly and fulfillment service as I’d originally intended. We’d ship them out to supporters, and generate enough new sales of these “version 1” EMpicks to fund a cost-reduction cycle on the EMpick design AND to turn assembly/test/ship over to a service. It would just take longer and require more work.
I was going ahead with this plan until the actual impact of the tariffs placed upon electronic components from China became clear to me. My most expensive components were going up in price by 25%. This spoiled the plan completely. Funds would be so tight I’d just barely be able to break even manufacturing and shipping EMpicks. One misstep, and I’d be losing money on each EMpick. A small business such as mine can’t afford to sell things at a loss and you don’t don’t increase your funds by breaking even.
I could not see a way forward with the existing “version 1” EMpick design. There was no point to completing the Version 1 EMpicks. This would just take more time and money, money I needed to fund the second design cycle. Then I evaluated the possible cost-reductions and decided I could keep quite a bit of the existing electronics and just make some key changes. The biggest change was replacing the microcontroller chip with a new chip that was almost twice as powerful yet nearly half the cost.
Short story is, this “just change some of the most expensive things” plan didn’t work out. I’ve ended up having to make enough changes that it really is a complete redesign of the EMpick. During this task, further testing of the haptic feedback idea revealed a weakness that would cause the tactile actuator to fail. The answer was a custom component, but this is a project in itself that I have neither time nor funding to do. Taking this feature out of the design has also lowered cost, so the forward-looking financial argument for creating the EMpick is now back to where it was originally supposed to be. And I’m in about the same place with the design process as I was in with the version 1 design in mid 2018.
Well, no, it’s better than this. We have quite a bit of firmware ready to run on this new design. And key circuit features have been tested, at least at the basic level of hardware functionality. There is a lot more work to do, but I should be able to do a BETA run in August, and if we do a quick-turn circuit test before this, I may have the confidence to manufacture about 1000 “BETA” EMpicks, the first 450 or so going to supporters like you who have waited so long.
This is my plan currently. If nothing goes wrong terribly wrong, I can see through to starting to deliver EMpicks in September or October of this year.
I say “may have the confidence” because as you may imagine, making 1000 copies of something and then discovering that some change needs to be made to the circuitry for them to work properly would be catastrophic. It’s much safer to make first 50, then 200, then perhaps 750, etc. But this takes more time and costs more money. It’s not easy, all of this. I will see what seems doable when the time comes.
This plan gets the EMpick done, but it doesn’t help me with any additional funding. However, something has been going on with the Vo-96 recently that has created an opportunity.
You see, I’ve never wanted to start a manufacturing company. I wanted to create ideas, invent and patent the means for realizing them, and then license those patents to other manufacturing companies. It’s possible to make a living this way, and it’s possible to get new ideas out to the world faster and wider than by trying to grow a new company from scratch and keeping all your ideas to yourself. And you can mostly keep doing what you love, instead of being pulled into all the other issues and challenges of running a larger conventional company.
I had this license thing going with Moog Music Inc. who licensed my technology and commissioned me to develop the Moog Guitar electronics. Since then, I’ve been trying to make this work with the “one person does everything” small manufacturer model. This isn’t working out. It’s missing the revenue I could be getting from licensing my technology to other companies.
Licensing can fund my ongoing work, to pay people to help as needed, etc. Over time, companies that licence from me will be coming out with their own products which will broaden the market and bring others into it. Some competition would be great.
With the help of modern turn-key manufacturing services I can manufacture the EMpick ongoingly and this can eventually, eventually fund my business. But to grow faster and go further, to really get this out there to the whole world of string playing musicians, and especially to make an updated Vo-96 available to guitarists, I need license deals.
With help from friends, I shipped close to 100 Vo-96 kits and guitar installations in 2014. I haven’t promoted the Vo-96 since then because I can’t afford to make them on my own. People message me that they want one, but I can’t build them. Right now there are many artists out there who have continued to play their Vo-96 instruments. There are some who have made a Vo-96 based instrument their main instrument for performance. A few of these people even took the initiative to have custom instruments manufactured around the Vo-96. These are guitarists who bought and paid for their Vo-96 just like everyone else — no special deals. I didn’t make any of this happen. The Vo-96 made it happen, because it does what it does.
This month I’ll be adding a couple of pages to my website specifically to advertise and explain the benefits of licensing this technology. I’ll send out a mailchimp mailing to announce this, but before I do I’ll publish the link here. You may be interested in taking a look at it, and who knows, perhaps some among my EMpick supporters are also potential licensees.
It’s just nuts isn’t it, that a guitarist such as Gabriel Marin of Consider the Source is out there playing an instrument that no other guitarist can currently buy, not for any amount of money? He’s creating curiosity and demand when there is no supply. So are others such as Ian Ethan Case. Check out these videos:
In this next one Gabriel talks candidly about his guitar and his rig:
And one last point:
The EMpick is designed around one channel of the future “next Vo-96”. The Vo-96 needs to be modular you see, to make it practical to cover the 4 string and 5 string bass market, as well as all the other “not 6 string” segments. The EMpick will be a great product in its own right, and it will also be the basis for the next Vo-96. It’s really all just one effort, one basic concept, (detailed computational control of physical vibration). Out of this one new technology, a great many different products are possible. If I can get some wind in my sales, I’ll make some of them happen.
What I really want to do next is to design an entire guitar with the latest Vo-96 technology in the core, and with other improvements and features. Yes, that will probably be a work of art that will take me a long time to finish. There’s the best reason for licensing this technology to companies big enough to get some products out fast.
Okay, this is all for now. If you go back and read previous entries, you’ll fill in some other details. Believe it or not, I left some things out of this long long blog post.
- Paul Vo
April 2, because no one takes anything published on April 1 seriously. 🙂
April 2, 2019:
Just a short note here. During this past month I’ve removed the haptic element and I’m sewing up the resulting loose ends. It’s been tax-prep time. Yes I know I should send out a Mailchimp mailing soon — it has been months. I should probably also advertise the EMpick on Facebook again. The thing is, these things take time. I’m so late on the EMpick design; I feel some things can wait. Things like tax season and other business matters can’t wait though.
March 31 Design Notes:
I’ve received notes of encouragement from supporters about removing the haptic feedback feature from the EMpick design. This was such a difficult choice to make because I spent quite a bit of time getting it to feel strong enough to be convincing to the user. It provided a convincing tactile sensation telling you the EMpick was very close to a string, and if you swept past the string the tactile feedback was strong enough to feel something like plucking the string with a pick.
I’m not going to revise the plastic enclosure design. It feels good in the hand just as it is. The plastic features specific to tactile feedback will remain in the design. This will save some time, and if it becomes possible to develop the durable custom haptic feedback component that the haptic feature needs, I won’t have to revise the plastic enclosure but only the circuit board to ad haptic feedback back into the design. A custom component would definitely add to the cost of the EMpick though, so this would have to be a more expensive “V2” of the EMpick.
The design revision I’ve been working through gives us twice the processing power of the original DSP chip, and at almost half the cost. The semiconductor industry continues to follow the Moore’s Law phenomenon of value multiplication. Just wait a bit and you’ll get a lot more for a lot less. It’s amazing. If only we could make such constantly outstanding progress everywhere else.
Battery power and run time — this was also stressed by tactile feedback. It takes a surprisingly large dollop of energy to produce a convincing tactile feedback sensation. But guess what? If we wait a while, we’ll begin to see lithium ion battery technology with even greater energy storage capacity.
Okay, back to work. When will this be finished? This depends how much I spend on outside services and assistants to help me get this done. I have had an initiative in the works to improve this situation on the back burner for some time now. I’ll reveal this quite soon. At this point, spending more on the design would really speed this project up into the Beta stage. I’m doing what I can. Stay tuned to this station.
That’s all for now. Back to work…
Feb 28, 2019 - Hmm, why won't this mountain move?
Two months into 2019:
I’ve made good progress on most aspects of the redesign, but I have to admit I am not going to be able to deliver on haptic feedback. This is very frustrating to me. It’s not that haptic feedback won’t work. It does work and it is very convincing. The problem is that to do it in a durable way that will keep working for many years, I really need to design our own haptic feedback component. Read on if you are interested in the details…
I’ve been working with an off-the-shelf component. It forces me to make too many compromises. One of those compromises is intolerable: The part is prone to failure in this application. This isn’t the part’s fault, its that the most energy-efficient way to implement haptic feedback is to arrange for the touch-feedback device to be in direct contact with your thumb when you hold the EMpick. This provides the most vivid emulation of the feel of a pick on a string. The downside is that it also exposes the touch feedback device to your thumb’s motion.
The haptic element I was working with is basically a voice coil. Anyone who knows voice coils knows you can’t push them laterally within the gap. I designed 3 plastic ball bearings to ride beneath your thumb and roll on a platform driven by the voice coil in response to lateral motion. This arrangement made it impossible for your thumb to push laterally on the voice coil. This tactile feedback system worked quite well, but not for long. I won’t detail the failure mechanism except to say that the voice coil element is not designed to withstand wide low frequency excursions.
This is a gradual sectioning of the tactile feedback element it was designed into the EMpick. Sorry if this takes a moment to load.
I’m quite sure I can design our own custom component that will provide good tactile response in just this same way and will last for many years. However, developing a custom component is a serious project in itself and would need its own budget and timeline. It’s frustrating, but considering how late the EMpick already is, I can’t responsibly go down the road of developing a custom haptic component. This is why I have to remove the tactile feedback feature. There is no sense shipping something that will definitely quickly fail.
Removing the haptic element from the EMpick design database will also take a bit of time, but it’s not too bad.
I have an alternative purely mechanical idea that has been in the EMpick design all along. What is needed is a way to direct the EMpick over the strings, moving from string to string quickly and easily, staying very close to the string with the EMpick coils, and yet never colliding with the strings. Tactile feedback would have helped with this by providing feedback to the player’s thumb as the EMpick came within control distance of a string. My alternate idea is an entirely mechanical method. Maybe if I can get over my haptic frustration, I can come to see this alternative as superior. It certainly is simpler.
What is is? Stay tuned to this blog and I’ll show it to you. At this moment though, I need to get back to work…
Jan 29, 2019 - Still moving the mountain a few inches.
Start of 2019 Update:
My apologies to those who found voinventions.com inaccessible over the weekend. An automatic plugin update appears to have crashed my site. It does not seem to have been due to a hack of the site. It took a bit of work to get back online. Unfortunately I’ve lost the last blog post, but I think tomorrow I can rewrite it and perhaps add a few points.
I’m going to try a different way of doing these. I’ll just grow this entry as things come to me.
I’m pretty happy with the basic shape of the EMpick. It looks a bit fat where the battery lives, but actually it fits into the hand very comfortably. Here are a couple of CAD images of the design, and then a couple of images of a 3D printed test piece of the actual plastic enclosure. I’m holding it and it really is just the right shape. One thing; I’m going to add a chamfer to the edge of the coil housing “pick” to make the edge rest against the side of my finger more comfortably.
That’s all for this moment. More to come soon…
- Paul Vo
Dec 23, 2018 - Out with the old, in with the new, like moving a mountain just a few inches.
There are 3 things going on. The EMpick processor change is one of them and we’ve made good progress there. Still, when you make a change like this you have to re-examine everything. It’s like moving all your stuff to the apartment next door. Sounds easy, but you still have to handle every little thing. “It’s like moving a mountain just a few inches.” This will result in a better product though, and I won’t have to raise the price. My market research tells me that $249 is about the limit that people will pay for a hand-held guitar playing device – no matter what it does.
The second thing is a continuing involvement with the medical project. This is rewarding; I mean, it really could help people. I can’t talk too much about it, I’m an inventor/consultant on the project but it’s not my project. The science and sophistication of the EMpick design is quite a bit further out on the “bleeding edge”, but thinking about finely controlled magnetic fields in a medical context brought me a very useful insight I’ve applied to the EMpick operating modes. The bottom line financial reality factors in here too: The medical project helps keep my budget balanced while I complete the EMpick. It lets me concentrate on the tech and not on finding new supporters. I’ll be all about advertising the EMpick once I have one to demonstrate, and there is nothing more important than getting to this goal ASAP.
The third thing is simply, “Everything Else”. When I first conceived of the Wond II Project and the EMpick, I had this not-quite-human idea that I could just power through it in a year. I’d spend the year doing nothing else, I thought. Well, this didn’t quite work out. You end up slowing down. And no matter your desire, life intrudes and you must deal with the realities. So I’ve had to find a balance where my energy is replenished; a sustainable drive forward rather than an exhausting all out sprint.
The latest reality is tariffs. Suddenly the cost of everything has become uncertain. Virtually every component I’m using has a tariff warning attached.
Here is a letter from my main components distributor to their customers. They are doing a good job trying to manage this for their customers, but there is great pricing uncertainty out there right now:
Our Valued Customers,
The tariffs announced by the Office of the United States Trade Representative (USTR) under the Section 301 Action went into effect on July 6th, 2018 with additional products added to the list on August 23rd. All items identified, as classified, under the Harmonized Tariff Schedule of the United States (HTSUS) with a country of origin of China are subject to a 25% duty upon importation into the United States. See List 1, List 2, and List 3.
Digi-Key continues to work toward minimizing the impact of the tariffs to our customers. We are working with the ECIA (Electronic Component Industry Association) and our 800+ suppliers to minimize impact on the supply chain yet adhere to the law. Due to our inventory position, we have been able to delay the immediate impact of the tariff in some instances.
Digi-Key is responding to customer requests to identify products impacted by this tariff in the search results on our website. Again, the tariff only impacts products with specific HTS codes, with a listed country of origin of China, and consumed within the United States. There is no impact on pricing for products within Digi-Key’s stock shipped outside the United States.
This is a dynamic situation and we will continue monitoring the situation and exhausting all avenues to mitigate any impact or uncertainty in your designs or supply chain.
President & COO
It’s nuts, this tariff thing. I discussed in my previous blog entry what we are doing to overcome this. It has been a big job, and it is ongoing.
I’m taking a holiday during the holidays. I need to have a week to reset and restore. When I return in January, it will be pedal to the metal. The rework is almost complete, and this time it is already tightly integrated with the plastic enclosure. In January I’m going to publish a time line for completion, BETA, and then production.
I know this project is very late. I will complete it though, I always do. Thank you very much for your patience and understanding. Happy holidays to everyone, and may we all have a happy, productive and successful year in 2019.
- Paul Vo
Nov 8, 2018 - It's been a busy month but changes have slowed our progress...
October was supposed to be the month I made a few EMpicks as a final test before going on to the Beta run. Sometimes things just refuse to line up and this has been one of those times. My collaborators were both unavailable during October. More significantly though, in the course of gearing up to make a few I updated the components list and pricing and, well, prices have risen quite a bit. I’ve been aware of a brand new microcontroller from Texas Instruments that is twice the power and half the price of the TM320F28069. I’ve decided the best solution is to move to this more powerful chip.
This isn’t as difficult as it seems. All these chips are in the same family and all use the same coding tools, so you can often just move code from the one to the other, In our case it is somewhat more complex, but not that much. One big boon to using this chip is that more firmware can be coded in C and C++ , a higher level coding language that is easier to write and maintain. However, it does mean I must revise the hardware design and we have to get the new code verified with the new chip. This task is ongoing at present.
The other task I’ve worked on this past few weeks is a new patent. The new hardware and firmware methods that go into driving fat heavy strings like bass guitar strings will probably end up replacing my original methods. For a while I was thinking to just let this go. After all, I hold several other patents and frankly it’s not clear these expensive documents are repaying me for the work and cost of writing them. But I have concluded it would be silly to just publish my very best work to date without obtaining a patent. So, once again I’m writing another patent application. This usually takes me a couple of weeks before I get it into the proper shape.
This is probably the right time to let everyone know that the other reason I’ve blown my target dates on the EMpick is that the whole funding thing has been more difficult than I’d hoped. As a result, earlier this year I took on an additional electronic design consulting task. This is in the field of medicine, which is completely different than music of course. Nevertheless, it turned out that in this specific area my expertise with tight control of magnetic fields allowed me to come up with a very good solution for the problem at hand. This isn’t my project, so I don’t want to say much more. However, if you are curious and know how to search for published patent applications at the USPTO, you can find the publicly published pending patent there under my original name, “Ierymenko”. While I have spent the majority of time on the EMpick this year, taking on this other creative work did slow me down this year.
The thing is, had I not been fortunate enough to be asked to join this medical project, I may have had to cancel the EMpick project due to lack of funding. Instead, I’ve been able to self-fund the EMpick project due to this medical commitment.
The other news this month is that I am negotiating with a company regarding my technology and if this goes well it could make a huge difference in the resources I’m able to fund to get these kinds of projects completed.
So you see, there is a lot going on. I’m also trying to remain somewhat balanced through all of this. Yes, I’m having an actual life too, if only for a few hours here and there. If I fail to enjoy this incredible world and the wonderful people in it, what really is the point of creating anything?
I’m sure this is trying everyone’s patience and I do apologize. Rest assured the EMpick will get done and it will be a great new musical device to add to your tool chest.
We will slip into 2019 with EMpick delivery. Do you remember that the iPhone from Apple came out two years later than originally promised? You may not; most people don’t. What counts today is that the iPhone is a truly great product. We are working very hard, really without compromise, to make the EMpick a total blow you away upside surprise. There is a LOT you can do with this tech. So please, stay tuned here to see the latest developments. Your EMpick remains reserved for you and its coming — just more a lot more slowly that I originally thought.
Thank you again for your support!
– Paul Vo
Oct 10, 2018 - Validation of the Enclosure Design
I have a few pictures here of a physical 3D printed EMpick enclosure. This enclosure has been fabricated using a special plastic and a machine capable of nearly perfect precision. The plastic material is not cosmetically pleasing, but this isn’t the point of this exercise. the purpose here is to validate all the features of the enclosure such as the snaps and screws that hold the enclosure together, and of course to just make sure everything fits and interacts as it should. I’m pleased to say I seem to have nailed this design. The little speaker “nest” is a shade too tight, and the snaps need a bit more overlap, but those are just improvements. The design works properly just as it is. Here then are some pictures of the first “real” EMpick enclosure:
I’m still working on the tactile feedback system. I would really just like to design exactly what I want this component to be, but at this point I think it would just take too long. I’m getting good results with the off the shelf system I’ve chosen.
I’d like to say hello and welcome to the supporters who have recently joined the project!
– Paul Vo
Sep 23, 2018 - Minor loss of time and balance...
The plastic parts all fit together correctly. And I have made a decision regarding the haptic actuator. Nevertheless, I had to attend to a personal matter that took almost 3 weeks. And along with this, it took longer than I’d hoped to arrange financing for the next leg of this project. I can’t complain, because the financing has now come through, so it’s full steam ahead again.
I’m debating if I want to share the nature of the personal matter. I may – but later. when you do a thing like this and it takes a long time, real life goes on of course, and sometimes there are issues that make it impossible to work. This isn’;t a sad story, it’s a very good outcome, but there was a time of worry and actions were required. Later I’ll share details, perhaps in the project epilogue.
Pictures of the actual plastic enclosure are in the works. I just wanted to get a few words out first. Please check back in a few days and see the test 3D print of the EMpick enclosure in “action” live and on Youtube. Thank you all for your continued patience.
Aug 31, 2018 - Haptic Happenings
A note on the enclosure: It is complete in every way – all 3 pieces, the Top, the Bottom, and the Battery Hatch.
We are experiencing a snag with the haptic component. Simply put, the off-the-shelf but ever-so-slightly-customized motor we need for the haptic feedback system has hit a snag. The company I’ve been working with isn’t being responsive The problem isn’t in design, it’s in the unwillingness of the producer to slightly alter their assembly process. I want them to leave out the terminals they normally install for electrical connections and just let me deal with the electrical connections using a pattern on the EMpick PCB. The terminals in question are in the way and must be removed for the part to function in the EMpick.
This is something I have to resolve very soon. It has been dragging on. I wish they’d say yes or no, but instead they say they are looking into it. Then a week of nothing. So you wait. Then you ask again. Still nothing. You ask one last time. Nothing again. At some point I guess nothing = no. This makes Boolean sense I suppose. It just takes a long time:
“Yes” = 1.
Nothing = 0.
0 = “No”.
I have figured out an alternative using an off-the-shelf component that can be used as-is without customization, and I’m now waiting to receive a sample to test this before committing to it. If we go this way, I will have to change an internal “nose” feature within the plastic EMpick housing to hold the alternate solution in place. Then I can release the plastic for initial fabrication.
I really want to get this haptic thing right this time. I tried to achieve this before in the Wond using a cell phone vibrator device, but this didn’t allow the degree of control needed to create the right tactile response. In the EMpick I’m using a different motor with much better characteristics for this task. If I can get this right, it will make a huge difference in the usability of the EMpick. I want it to feel like a pick under your thumb, so that as you touch the string with the EMpick’s magnetic field, you experience the same sensation as plucking a string with a flat pick. Hence the name: Electro-Magnetic Pick = EMpick.
We’re close: Prove out the new haptic, which nails down the plastic. Check everything several times. Order a small test run of real EMpicks. It’s close.
July 29, 2018 - The BIG Battery change
My blog entry from last month announced, “The Enclosure is Complete”. And so it was. At that point I decided it was time to order the custom lithium ion batteries for the EMpick Beta run. Along with this I reviewed what I’d learned from our original Wond shipping experience regarding the regulations for shipping these batteries “contained in equipment” from the USA to the other countries. We are thrilled to have supporters living all over the world. It would be truly brainless to design a product we couldn’t ship to them.
Those of you who were also with us for the Vo-96 production and shipping may recall that the regulations around shipping lithium ion batteries were changing and tightening right in the middle of it all. It turns out the rules have changed again. After looking at transportation regulations of several countries I concluded that to do this properly with a custom battery, we’d have to go through much the same expensive certification process that battery retailers must go through to introduce a new battery to the market. And we could not begin the process until a final production model could be sent in for testing. The labs that do this testing are backlogged because every product on the planet containing a custom lithium battery requires the same tests. We’d also have to maintain a special shipping license and still we’d pay a ‘hazard’ premium in shipping charges. In many places you can’t ship any lithium batteries by air but only by land and sea, etc. etc. The more I looked at it, the more involved it became. If you’ve read this far your eyes are starting to cross. See what I mean? Even just within the USA, approvals and shipping restrictions would raise costs and introduce a long delay between completing the EMpick and being able to ship it.
The bottom line is that today, developing a product containing a custom lithium-based battery is for larger companies who can justify the expenditures and keep up with the changing regulations. A small company like mine shouldn’t try to do it. It’s better to use a stock standard battery of some kind…
The solution: Commercially Available Lithium Ion Cells…
I’ve been aware of the alternative solution for some time. The solution is to use an off-the-shelf generally available battery. So, after a week of evaluating the options I decided the best way forward was to use an AA – sized battery. Several companies make these batteries, which are used in various devices from flashlights to E-cigarettes. I first saw these on the market I think in 2015, but I wasn’t sure they would “stick” because an AA-sized lithium ion battery puts out 3.7 volts while a regular alkaline AA puts out 1.6 volts. It is possible to damage devices designed for standard alkaline batteries by installing one of these powerful lithium batteries. Moreover, the length of these seems to deviate some. The international AA size is termed “14500”, which means 14 mm in diameter and 50 mm in length, but the actual lithium ion batteries called 14500 vary from being slightly smaller to being 3mm longer and almost 1 mm fatter. However, despite these variations and voltage incompatibility, these 14500 lithium ion batteries are here to stay. The product lines have expanded. There are more companies making them. The prices have dropped. Their power capacity has increased. They come with built in protection circuits. One particular battery that is very suitable for the EMpick even comes with a built in charger. You plug a micro-USB directly into the side of the battery and charge it from any standard 5V USB jack. Very cool:
The cell above is made by “Nitecore”. It’s model # NL1475R and you can obtain one together with a USB charging cable from Amazon HERE. This link may not be reachable from all countries, but it seems there is always a way to obtain this battery or others like it. If you use a search engine to look for “lithium ion 14500” (without quotes), you’ll easily be able to find a retail source. ( If you are a supporter and are unable to find a supplier for a 14500 lithium ion battery, please let me know by email at email@example.com ) A suitable battery must be lithium ion, AA sized, or 14500 sized, and must be the usual round cylindrical cell with a button cap on the + end. The battery should include a built-in protection circuit. (most of them do). The Nitecore with the built-in charger has a capacity of 750 mAh, (0.75 Ampere/hours) which is quite good and should last a couple of hours of intense use between charges. You can also use a more powerful one that needs a separate charger. (inexpensive). This one stores 850 mAh but requires its own charger:
HERE is a battery from a different manufacturer also via Amazon. This one claims to be 1200 mAH and comes as a pair with a charger for just over 10 dollars. I have heard that there are newer batteries that can pack more power into a given size, but I also know some of these numbers represent operation under “ideal” conditions. I think they will all work quite well in the EMpick. It does seem these batteries are getting better all the time. Once we get going with the EMpick, I’ll put out a list of recommendations. At this point, the Nitecore products look best to me.
I’m planning on shipping the EMpick with a standard Alkaline AA cell because I can ship this anywhere with no restrictions. (Provided I can get it to work on a standard AA cell. I’m pretty sure it will but I have to test this. I don’t know how long it will work on an AA Alkaline cell, but if I can manage this it will mean you can buy a few at any drug large food store anywhere, so even if you lose or forget to bring your lithium ion 14500 batteries, the show can go on!)
Making the Change
It is actually more difficult and expensive in the EMpick design to use this standard replaceable battery compared to a fixed custom battery. Further below I’ll show how the new battery compartment in the EMpick works. It requires an additional component of plastic tooling, some other parts, a bit more labor to put the EMpick together, and some additional support circuitry. I’ve worked really hard on this and I think I’ve done it in a way that maximizes the advantages:
- Keep an extra battery charged. Never run out of power.
- Lithium ion batteries do wear out. The replaceable battery can always be relatively new.
- Increasingly powerful batteries of this size are being developed.
- Really easy to swap out the old battery for a fresh one.
- Battery compartment accepts all size variations of 14500 or AA batteries.
The main challenge was to design a battery compartment that closes securely and yet is very easily opened without any tools. I think the design I came up with achieves this, but you will all have to tell me what you think once you have your EMpick. What follows are some pictures and a short video on Youtube showing how you swap out the EMpick battery for a fresh one. I think it is quite easy:
Here is a link to the exciting Youtube action adventure movie entitled “EMpick Battery Replacement:”
And next is a CAD representation of the current EMpick design, which is now once again COMPLETE, and going out for 3D printing to check the fit of everything.
Note that for all the images in this blog post, the colors are false colors from my CAD system. LED indicators must be visible through the enclosure. I’m planing on making the real EMpick out of transparent medical grade polycarbonate plastic – tough, durable, hypo-allergenic, a bit more expensive but the “right stuff”. It should look… interesting … with all the electronic bits inside visible from the outside.
June 12, 2018 - Haptic Feedback, the Enclosure complete, and an Ad!
This project has kept me very busy and I’m sorry to have missed an update for May. Everything is in forward motion, from ordering some long lead-time components to arranging terms with suppliers, to doing some advertising, to directing work on the code, to – well you get the idea.
The Enclosure is Complete!
I guess the most visually interesting step forward is in the enclosure design, which is finished now – hooray! I’m just waiting for a sample of the custom haptic feedback component to make sure it’s going to lock into its ‘nest’ in the EMpick. Here are some pics and details. Please note these are still in false color; they are computer generated CAD images. I am planning on making the enclosure clear plastic – the same medical grade polycarbonate we used with the original Wond. The EMpick enclosure is more complicated for a number of reasons. One is that the enclosure halves snap together, and this combined with two screws should make for a very solid device without having to glue things together.
Let me show you around a bit…
Above: I’m quite enthusiastic about the tactile (haptic) feedback system we’ve worked out here. In fairly realistic testing it provides a very clear sense of touching the string with the EMpick’s magnetic field. It does require a custom component and we are waiting for samples of this. The real evaluation must wait until the 3D models, the revised circuit board and this custom component can all be assembled. I can see that point in time down a straight run through some lead-times and a lot of detail. No show stoppers or unexpected delays are in the way!
Below: There are so many details here essential to things working as they ought. This is the inside of the top half:
Here is a detail of one of the clasps. This one is behind the coils:
From the Wond to the EMpick – an advert – with great musicians!
I’ve been working on this video ad for a couple of months just now and then in the background. It’s based on an iPhone video capture of a few moments in the concert by Acoustic Syndicate held on the last evening of operation of the Altamont Theater in Asheville. the Altamont has been a great downtown venue and we are all sorry to see it close. Acoustic Syndicate gave it a final performance evening to remember that final week. In this advert I tried to tie together the original Wond – which Billy uses in the video – to the EMpick, which in the video is presented at an earlier stage of completion than today. These fine musicians make this an entertaining video. Do catch Acoustic Syndicate if you get a chance! The audio here was mono, recorded on an i-phone as was the video, but I used some techniques to give it a fairly realistic stereo field and recover a fuller range of sound.
April 28, 2018 - A major milestone in EMpick development!
A few weeks ago we reached the milestone that had proved to be more difficult than anticipated. We brought the harmonic control capability I first developed in the Vo-96 over the EMpick. It is now running on a different digital signal processor, but more importantly it has been rewritten to overcome the structural problems that made it difficult to modify and add features to the Vo-96. This new implementation is fully scripted, and my plan is to expose these scripts to the user initially in the form of a simple text file so that users with very little computing experience can hack the behavior, features and sounds of the EMpick (without the possibility of damaging the control system) and add unique custom features, etc.
From here on forward we’ll be working on features, defining this scripting language, getting the microphone – analysis – re-synthesis loop working, etc. Hopefully now that the core capabilities are solid, we’ll be able to move forward more quickly. The core capabilities that are in place now include the Vo Harmonic Control system, the USB “thumb drive” OS-independent behavior, charging, and firmware updates via USB by dragging an update file into the USB “EMpick” thumb drive.
Oh — and the beaming of re-synthesized audio into the pickups of an electric guitar is also in-place, though the sounds themselves are not there yet. On this point, those who have used the Wond with an electric know that the raw sound of the Wond’s chaos drive field, (that’s really what it is), induces some very aggressive sounds in an electric’s pickups. Turning down the tone control, playing through a filter like a wah pedal for example, or positioning the Wond well away from the pickups are ways to deal with it. And there are some who like these chaotic sounds as sonic effects. The EMpick turns this around into a major virtue of the device. With advanced processing, the drive sounds become synth sounds and sampled sounds re-synthesized to the pitch of the string being addressed by the EMpick, and emerge from your electric guitar as part of the guitar signal. I believe people will find this to be a whole new thing to explore, and very musically useful.
But at the current stage of development we are validating the basics that make all this possible. Here is a video I made that demonstrates harmonic control. I made it for subscribers, but also as a kind of “We have this technology. Would you like to license it” advertisement. (For this whole new idea of acoustic synthesis to take off, it can’t be just me doing it. I’m starting to look for people and companies to license this technology and gain the benefit of my years of research and patents as they create their own products based on their own ideas about acoustic synthesis. )
I finally have video editor software I’m not completely disgusted with. 😉
Oh – yes, I know about the TC Aeon device. So does my friend Greg Heet, inventor of the Ebow. The Aeon is basically an Ebow, though some may say it borrowed its upright positioning and it’s “headlight” led that shines onto the string from the Wond. Maybe so – but tough for them because the hand grip “L” shape of the EMpick is quite superior. 🙂 Anyway, the device is essentially an Ebow. It even rides along the adjacent strings like the Ebow does. It is in no way an EMpick except that you hold it in your hand and it sustains a string. When you receive your own EMpick, you’ll quickly see what I mean. But then again, the Aeon is only about $80. The EMpick is much more expensive but it will do much more than basic sustain – and it will do basic sustain faster and at higher power. It’s in a different league. My hope is that TC creates a buzz around this type of device which will help sales both of the EMpick and of the Ebow.
Sorry there haven’t been more updates. I have a one-track mind and I don’t really argue with myself when that track is on EMpick development. We have also produced an advert — look for it on Facebook in a week or so. At this stage, growing our base of supporters is also very important.
Thank you all for your patience. I know the EMpick is taking longer than planned, but we’ll get it completed and it will be great – so after it has been out a while no one will remember it was late. This is what I tell myself anyway!
Here are a couple of videos about the Aeon. They do say imitation is the sincerest form of flattery. Greg and I should be flattered I guess!
From TC Helicon: (I’d like to have their ad budget to work with!)
BTW, this doesn’t exist yet either. They say they will begin shipping in July 2018 I think.
… and here is a less hyped evaluation by “Ryan”:
- Paul Vo
March 18, 2018 -- Uh, what happened to February?
Well the long and short of it is that I lost a couple of weeks to the flu in February and this put a dent in our momentum. So February was not a good month. We are back on it now full force.
On a positive note, there is suddenly a lot of new interest in the Vo-96 because Gabriel Marin has been out there performing with his double-neck instrument. He is a monster player and there are many guitarists who follow him. We’ll be doing some joint marketing in the future. This isn’t the EMpick, but it is closely related because the EMpick’s starting point is as a single channel of the Vo-96 and we are building on top of that with new features and the ability to handle all kinds and diameters of string. (except nylon of course).
Back in EMpick land, a lot of code is being written but there is nothing spectacular to show yet. What everyone wants to see is the actual EMpick, in an actual enclosure, with actual electronics and doing its actual thing. This is what we are working up to, one tiny detail at a time. It does feel like forever, but there are only so many details. Eventually you get past the last one and it’s ready. The milestone ahead is when I order what I hope will be the final EMpick circuit boards and a few enclosures to test via 3D printing.
One supporter questioned the tab-type momentary switches. I did some tests on this and decided to go with an actual button instead. As soon as I make this change I’ll post a CAD picture. The problem is not wear and fatigue of the tabs over time because the movement is so very small. The problem is that the injection molding process we’ll use for high volume production may not be able to hold the geometry perfectly, and if the tabs come out a bit bent down they will activate the switches all by themselves. It’s more expensive to use separate buttons and I have to take steps to make sure they don’t rattle or buzz, but it’s the right way to go. Here is the current state of the design with integrated tab switches: (It’s in a false color I find helpful when designing in Solidworks.)
Back to the Vo-96 — here is an unpublished video of Gabriel experimenting with the Vo-96’s interaction with a digital pedal. I’m sorry, I don’t know which particular pedal, but it doesn’t sound like this with any other guitar anyway. There is a secret sauce interaction going on here, a little GM special. Please don’t share this out, it’s meant just for you. There will be many more videos of his work in the future, done to a better quality standard. But this is interesting, for sure. The relevant point is, this is being achieved through injection from the Vo-96 to the standard guitar pickups on his very custom guitar. The EMpick will do this kind of electromagnetic injection into the pickups. I don’t know if *exactly* this sound will be obtainable, but this is the kind of thing the EMpick will make possible. Here is the link:
Once the EMpick is done, I plan on a new Vo-96 incorporating EMpick innovations for all 6 strings – and in a modular form factor so that it can be applied to any stringed instrument from 4 strings to, well I guess 88 strings. This Vo-96 will be much more expensive than the EMpick in any channel configuration of course – and is quite far away in the future.
— Paul Vo
Gabriel will be playing this instrument exclusively going forward. My hope is this will help raise awareness of acoustic synthesis technology and what it can do and help us reach all those guitarists who haven’t heard of my work at all. I run into such guitarists frequently, even here in Asheville. They are too busy with gigs and rehearsing to spend much time on social media, and they won’t find my products in any musical instrument store, or on Sweetwater, etc. So they just don’t know what we are up to over here in Asheville – until they do!
January 25, 2018 -- 2018!
I hope everyone enjoyed the holidays!
Yes, in posting this update I just managed to accidentally overwrite the previous BLOG entry from December 21, 2017. It was about achieving USB functionality for software updates. Hopefully you were able to read it before today. Not too difficult to do, but I know better! Darn!
2018: it’s like the cover of a sci-fi novel…
With the holidays and family things and end of year business things and beginning of year business things, I had to put our EMpick development on pause for about 3 weeks to get all of these things sorted. Starting the new year with only a very small backlog and the EMpick on the front burner.
The things that have happened during this pause have all been very good. A lingering business matter has been resolved in a very positive way and I’m thankful for this. Other more personal things have happened, also in a very positive way.
A new collaborator, a creative in the art of media has joined our other collaborators working with me on the EMpick. You will start to see a positive difference in activity on our social media pages, advertising, and web presence. Advertising especially will get a boost as the EMpick gets closer to completion.
Job one for me is designing the EMpick. I do my best work when I can focus on product development. Developing a product is itself a multitasking enterprise with a long list of action items. Collaborators in the area of coding our USB implementation and in optimizing specialized real-time DSP control code are being a great help. I’m personally still working on the design of the plastic enclosure – almost done!
This isn’t a traditional kind of company with executives and employees. I don’t want to be an executive. I want to try new ideas and develop new products. I work with collaborators. This is a more flexible, cooperative and less “position-structured” way to do development. We are independent but we all want to bring something new to the world of musical instruments so we are working together on the EMpick. It’s my project of course, so it’s my responsibility to keep a step ahead and lead the way.
(I may blog more about this business model more one day, especially if it continues to work out well.)
Every ad I’ve run has brought in more supporters. I just haven’t had the time to create very many ads. Now with a new shoulder to it we’ll have better media presence, which hopefully leads to more supporters, which leads to more EMpicks…
Thank you for Supporting the Wond II Project!
We have been back at it strongly for a couple of weeks now this January. We aren’t quite ready with the Vo-96 functionality yet. It’s not just a matter of copying it over from the Vo-96. It is being redesigned around a more advanced way of sensing and controlling guitar strings. Current Vo-96 technology balks at brass and nickel. This new control method works with any magnetically responsive strings from acoustic guitars using regular copper/brass/bronze strings on up to electrics using pure nickel wrap. While the best results will always come from using strings with the best magnetic properties, this new tech will actually drive any steel-core string with as much power as nature allows.
In related NEWS:
The Vo-96 has been in the news lately. Gabriel Marin of Consider the Source is about to perform here in Asheville. He’ll be playing his new double neck, double Vo-96 electric instrument. Gabriel plays a lot of solo lines – melodies, lead lines he delivers with off-planet virtuosity. One string at a time. So the sounds he’ll be using will be very similar to the sounds the EMpick will give you when operated in Vo-96 mode. Here is this amazing instrument:
Look for Consider the Source on YouTube and at a venue near you.
That’s all for now, but as soon as we have the EMpick operating in Vo-96 mode I’ll be blogging about it
— Paul Vo
December 21, 2017 --- We have USB working!
I just accidentally overwrote the original blog entry for December 21 with the new one for January 25, 2018. So here is a summary of the missing blog entry:
Basically, what this blog post showed was a demo video of me showing how the EMpick now becomes a “thumb drive” when you connect USB. This represents a big milestone in EMpick development. It wasn’t easy, especially with the complexity of USB-C. However, USB-C will be everywhere in a couple of years, (It’s called “Thunderbolt 3” on MAC), and USB-C on the EMpick will make a lot of sense then and into the future.
To do a software update and/or charge your EMpick’s battery, you just plug a USB-C cable into the EMpick and connect it up to your computer, or to a 5V USB power source if all you want to do is charge. When you pl;ug it into a computer the EMpick comes up as a little external flash – a thumb drive” named EMpick in your folder tree. You’ll download a software update from our Voinventions site here, drag the file onto the EMpick thumb drive, and that’s it – software update done!
Nov 21, 2017 - Working on the enclosure and testing more code
I’m just going to drop random screenshots here for the next few days. The whole thing is coming together and it’s kind of a brain stretch – little details to deal with in the code, in the electronics, and the enclosure design underway in Solidworks. (By the way, what a fine software application is Solidworks. One of the best conceived and best maintained pieces of software I’ve ever encountered!)
Here is a link to a little video friend and supporter Killick Hinds made after visiting here in Asheville on afternoon. Let me explain what you are seeing and hearing here. I have a 15 inch long 15 gauge plain guitar string I am using to test the EMpick circuit you see in the video. Killick is just lightly tapping and sliding on this string with his bare fingertip. There is no slide, nothing, just the soft pad of a finger. This particular string just happens to sound like a weird charged up penny-whistle, but this could be any sound. The point is, with this technology you could use the EMpick on a string and play instant slide or various notes on that string with just a light touch of your fingertip.
The following screenshots – think of a sculpture. You start with rough block shapes and you chisel them down. Except in software you can also build them up in a different way and move them around. So nothing you see in this blog post is exactly what the final EMpick design will look like. What you see is that final design slowly emerging from imagination tempered by logical necessities.
Oct 30, 2017 - The EMpick wakes up!
EMpick™ — Coming Alive and Starting to Work!
What you are seeing in this short video is the EMpick with all the necessary instrumentation connected for loading and running firmware and testing all the various functions. Here I’m testing the basic sustain functionality. My finger is on a short length of guitar string. The EMpick is driving this with one set of coils — you see these mounted to the “string jig”, but they’d normally be at “pick” end of the EMpick body. (No enclosure for this yet – working on it!)
This is the newest sustain technology. I’m grasping the string between thumb and finger and sliding along to change the pitch. Like playing bottle guitar with just your fingers! This is what I call “sustain” !!
“It’s not just a string sustainer, it’s a hand-held single string guitar synthesizer.”
If you can see where this is going, click here to Join the Project and reserve your EMpick string controller. You’ll be helping me by increasing the resources I have available for development, and you’ll be reserving your EMpick at a really good price.
Please read the previous blog entry for much more information. You’ll find some pics and a video of how to hold the EMpick in the blog entry before that. You can see it’s coming, slow yes, but sure. Slow but sure.
Reserve your EMpick right now using this link to Join the Wond II Project!
Oct 3, 2017 - More About the EMpick...
EMpick™ — Details and Possibilities
I wince and get indigestion every time I hear the term “elevator pitch” but…
“It’s not just a string sustainer, it’s a hand-held single string guitar synthesizer.”
“A hand-held single string guitar synthesizer.” This is one aspect, but it is also a string exciter and sustainer. And it controls the timbre of string vibration. The elevator pitch keeps getting longer!
The following is a very non-elevator but still not completely detailed description of what the EMpick can do and, everything…
The EMpick is a digital acoustic synthesizer.
The EMpick is entirely firmware controlled, it does nothing without some code in it and everything it does is defined by code. On this page is a portion of the ongoing ultimate list of features. All these things are possible, but some require firmware development beyond what we will have in the first release.
The first EMpicks will ship with basic firmware which will do what the Wond does – actually more like what one channel of the Vo-96 does, plus one or two cool other new things. This is a starting point. Continuing sales will be in everyone’s interest because continuing sales will fund continuing firmware development. (So it must be at least somewhat mindblowing when it first comes out, or sales won’t continue.) The product will just get better and better over time with updates. This is the nature of most successful connected digital products these days, when you really think about it.
Firmware Update Process:
In the music products industry there are some very weird and complicated ways to update firmware. I really wanted to make our update process simple and easy. Therefore: The EMpick looks like a little thumb drive to your computer and connects via USB for both charging and firmware updates. You will download the update file from our site and drag it into the EMpick folder that springs into existence when you attach the EMpick to your computer – just like moving or copying a file to a thumb drive. (As far as your computer knows, the EMpick IS a thumb drive.) No other apps are needed on desktop systems or laptops. Windows, Apple Mac, Linux – anything with a file system that will recognize an external flash drive will just work automatically. I don’t think it can be much simpler than this. Right at this moment we are proving out the rather complex code needed to make this happen on the EMpick – so far so good!
A partial list of EMpick Capabilities:
- excite and sustain vibration. [In Version 1]
- control the timbre of vibration. [Basic In Version 1 ]
- control the octave of vibration. [We’ll try to get it in Version 1]
- excite any of the natural harmonics individually, with or without the fundamental. [Simple version to start.]
- inject a sound at the same pitch as the string into the guitar pickups. This sound may be very different than the string sound, i.e., it may be synthesized or re-synthesized in all the creative ways of electronic synthesis. [Simple version to start.]
- Put an octave subsonic below the guitar string pitch – really, acoustically and electrically. [Simple version to start – but yes!]
- Echo a recorded series of notes played within the harmonics of the next note played.
- Accept data (wav) files to define a timbre of string vibration.
- Use the microphone to record a timbre to define string vibration.
- Drive any kind of string from the fattest bass guitar string to the guitar’s high E. As long as the string responds magnetically we are good. (The Wond gets close to this but fat bass strings don’t always work with it depending on brand). [In Version 1]
- All the guitar pedal technologies – modeling, echo, reverb, flanging, ring mod – whatever, can be made available to the Empick user via the right firmware. It’s an ocean of possibilities, so I want to open up the process and allow other people and other companies to make use of it.
On an acoustic instrument The EMpick emits its own sound, (if you want it to), truly adding a new voice to the instrument. On an electric guitar, it emits a magnetic field that couples sound into your guitar signal.
We’ll try to get all this basic structure into the first release and make the [indicated] features surfaced and functional.
It’s going to be a very interesting device.
On the original Wond, I was dissappointed by the tactile system. It didn’t cut it.
The tactile system in the EMpick is a whole new deal, stronger, much more sophisticated, and it acts directly against your thumb. If what I’ve tested in the lab works the same when we load it into the actual EMpick, the experience will be quite amazing. You’ll “feel” when the magnetic field touches the string, so the EMpick will feel something like it feels when you pluck a string with a real guitar pick, except you won’t actually touch the string physically but only magnetically. It’s be magic, if I get this right. It’s tricky. We’re working on it!
Status and Pricing:
Right now I’m putting out our first Wond II Project advert in quite some time. Milestones should become more frequent now that we have hardware. My plan is to run a new advert at each milestone and simultaneously raise the price a bit — because the wait will be less and the exact nature of the device less uncertain. The EMpick will reach its new price of $249 when it is available for sale more or less immediately from ongoing production.
The discounts I’m giving right now would not be sustainable as an ongoing business. But I’m asking people to support me now for something they won’t recieve for some months. When you sign up today, you help me make the case to financeers that this is a creditworthy venture.
$249 is truly an amazingly low price considering the cost of putting all these components together to make an EMpick. It’s less than the price of many a boutique pedal, but it is way more “boutique” in design, contant and capability. This fair price is doable because I’m using a new model of enterprise made possible by the internet. Every customer and I can have a direct and simple one-to-one transaction, no matter where we are on this planet Earth. I’m excited about this new way of business. It doesn’t look like all those old ways of doing business. It lets me give the customer the maximum value I can give. Seems like a good idea to me.
Please go to the previous blog entry (September 24, 2017.) to follow more of the EMpick story. You’ll also find some pics there, and a video of how to hold the EMpick.
Or reserve your EMpick right now using this link to Join the Wond II Project!
September 24, 2017 - The EMpick Powers UP!
Introducing the EMpick™
David Torn visited my workshop last year while he was in town for a concert, and I was able to show him the Vo-96 and a prototype of the first Wond. We had a good talk about this and that and how, and why. His intuitions and reactions to the Vo-96 and the Wond gave me a lot to think about.
I designed the Wond to be held like a pen or a highlighter. Torn was holding it that way and sustaining a string, but I could see he wasn’t quite comfortable with it. At one point he flipped the Wond sideways to hold it in his palm. I remember he gave me a quick meaningful glance as he did this.
I realized he’d held the Wond like an oud pick. (The oud is one of Torn’s favorite instruments.) What if I reconceived the Wond along these lines? The coils would have to be repositioned. Everything would have to change, but it would be better. Clearly better.
This thought went around in my head for a long time before I did anything about it. When I began the design of the Wond II earlier this year, I just allowed the ideas to flow. At some point I realized I was going with an alternate ergonomic idea. We’d hold the device in the palm. Then it came to me that the coil end represented an electronic pick…
You hold the EMpick in the palm of your hand, and this puts your hand in contact with the length of it. A small change in any of the 3 angles of orientation is thus magnified by the length of the Empick’s “handle”, giving your hand much more to go on as you work to train your motor memory and automate the basic task of positioning it. It’s just vastly superior ergonomically
I’ve decided to rename the Wond II the “EMpick”. The EMpick is shaped like a guitar pick with a handle. You hold it like an oud pick. It just makes more sense this way.
The list price is going up too – but this doesn’t affect existing supporters – you just got an even better deal than you originally thought. If not for you, none of this would be happening!)
Here is a quick iphone video I made about the EMpick.
Below this text is a picture of the very first moment the EMpick electronics I’ve been designing since February were powered up. The very bright light is from the string illumination lamp, which is on FULL because there is no firmware installed yet to tell it to do otherwise. It is a tri-color led capable of producing all the colors including white, and you’ll be able to control it’s brightness and color modes – just another item on the long list of things the firmware takes care of.
What you don’t see here is the firmware. While designing the hardware, I’ve also been working with one collaborator on the west coast and another to the east. So a significant chunk of firmware is ready to be loaded into the EMpick. This is exactly where things are, at this moment right now. I have no definite idea how long the firmware integration is going to take, but my general plan is to first get all the subsystems proved out. (there is an accelerometer to tell the DSP about the EMpick’s motion and orientation; there are color-aware light sensors, there is the tactile feedback system – lots of stuff). Then we’ll make sure the firmware update process is bullet-proof and the whole battery charging thing too. This will give us a very basic level of operation going with basic sustain functionality. My inclination at this point is to ask any of our current supporters to be a beta testers if they want to participate this way. (I know some do, and thank you!). I want to put the EMpick into your hands. We’ll continue to work on the firmware and everyone can watch and participate, (if they want), by providing feedback on what you experience, making suggestions for various EMpick behaviors, etc.
This musical tool can do so much. One of my goals for the firmware is to open it up by publishing a programming reference, (an API). Anyone who wanted to write custom code could get our SDK and go for it. It will be an open, creative-commons license. You see, there are enough sensors and enough computing horsepower here to do many interesting things with an EMpick. The permutations of it’s capabilities could go anywhere, even to different instruments altogether, where perhaps one or more EMpicks could be mounted and used to make music in some completely different way. For my part, I play guitar so I think I can make the best contribution by focusing on the EMpick’s use with the guitar.
This, loosely with lots of room for improvements, is the current state of the Wond II Project and the EMpick product. Pics follow…
The size, relative to Popsicle sticks:
Here’s a picture taken that time David Torn came to my workshop to put some thoughts into my head:
– Paul Vo
September 18, 2017 - Finally have the boards!
The boards are here, and I’m about to be deeply into merging firmware and hardware. I’ll blog about some other mysterious changes and aspects of this, including the term “EMpick” in a few days. Right now I just want to complete the electrical test. Here is what the circuit boards look like: (The Popsicle sticks are a size reference and are not essential for operation 🙂 )
August 14, 2017 - Ordering the first complete circuit boards. (!)
That really just says it all. Of course, I’ve revised every bit of it at least once. When you are making something that starts with a blank sheet, everything needs to be designed A-Z. There is no existing form factor, no re-used circuits, etc. You literally make it up as you go along, in layers from the first hazy ideas to the full details of everything, specified to the nearest 1/1000″. (Actually, to the nearest micrometer these days.)
I’ve been building these layers for 7 months. I can’t push everything out of my head and into computer databases in one giant continuous flow. Instead, it’s take a few steps forward, look around, tune your visualization, then often it’s go back and rearrange things because you can see what is coming more clearly now and you need to have done something a little bit differently before. Like this, back and forth. Forever. But then you are at the end, and it’s all there in the computer, ready to be sent anywhere on the planet digitally to wherever it gets made. But not before going over every bit of it A-Z to make sure the overall design integrity made it through.
So – that’s done, and I’m creating “Gerber files”. A Gerber file isn’t just baby food, it’s a specialized file format specific to the circuit board fabrication industry. Some of these are visually interesting – the whole thing is often conventionally called the “artwork”. I’ll collect some screen shots and post them here after I get the files to the fab.
July 22, 2017 - Obsessing the Circuit Layout
Here it is July. I’m in the middle of converting the schematic to a printed circuit database. It’s a squeeze, putting all this electronic capability into a hand-held device. It’s necessary to think about every component and every square millimeter, so it goes frustratingly slowly. You work on it every day and you get a lot done. Then you zoom out and your day’s work just disappears, it is so small.
Last month I completed the schematic. In some ways, this means the electronic design is complete. However, a schematic is just an abstract presentation of the circuit. To make this actually work, we need to select the individual component packages, (most parts come in several physical forms, some smaller than others), place this graphically on a printed circuit board, (PCB), and connect everything with what will become copper traces on the fabricated circuit board.
The PCB layout process begins with something called a “rat’s nest”:
This is all the parts, and the blue lines are all the connections. This is the same information contained in the schematic, but in a jumbled physical representation, the “rat’s nest”. The task is to untangle all this, place the parts where they need to be, (“need to be” is a complex thing. Some parts need to be closer to others, electrically. Some further from others. Some must be placed to be accessed by the user – that’s you). The CAD, (computer-aided design), software I run helps keep track of everything, but it doesn’t do any of the other thinking for you.
(These are all “screen shots” so they are a bit pixelated.)
I usually turn the connection map off, and start by sorting things out in an organized way:
There, this is better! Looks…easy, maybe…
Next, I usually pick something that needs to be in a particular place and work around this. As soon as I take this step, the overall shape of the thing and the constrained size of the PCB comes into play. In this project, the PCB is only 0.950 inches, (950 mils), wide. Fortunately, some really tiny versions of our parts have recently become available.
Here I’ve turned on some connection lines and moved some selected components into position:
The image above shows the USB end of the Wond II with the USB-C connector.
A few more days of obsessive work, and here is a close-up of a big chunk of completed circuitry:
This is the entire USB-C interface, USB and firmware update memory controller micro, power on/off switch, battery charging circuit, and two precision switching power supplies. All the “housekeeping” stuff. You are looking at just over 1 square inch of circuit. It’s nuts, what you can do today. And this is not yet the smallest scale one can work with. You are looking at circuits smaller and tighter than virtually any audio electronics you may have on hand, but this is still not quite as miniaturized as your cell phone. Cell phone technology is driving this miniaturization, we are just surfing along in its wake here.
Okay, I have more of the same, but you get the idea. This is what is going on right now with the Wond II Project!
It’s a race towards the first working model…
– Paul Vo
June 14, 2017 - Tiny Parts!
The BIG SHRink…
“Electronic components get smaller and smaller.” We’ve all heard this. The number of circuit components you can put on a small printed circuit board has increased at least tenfold since the familiar DIP package became “old”.
Microcontrollers, tiny systems on a chip that can do any task assigned to them within the same tiny piece of real estate on a circuit board, have also become much smaller, more powerful, and less expensive.
Right now components are undergoing a rapid devolution in size. Everything is being scaled down much closer to its essentials. All the extra wire, all the unnecessary plastic in the IC package is being squeezed out of the components being introduced today. Components that were popular just five years ago are now on notice to be discontinued and are not recommended for new designs. There is a real push towards a level of integration as close as possible to the size of the IC chips themselves.
For perspective, in the Vo-96 design I completed in 2013, only one of the ICs I used was available in the new, smallest QFN-type package. (It’s the capacitance sensor IC under the touchpad.) Now, over half the chips in the Wond II design are QFN or some other package at the same scale as QFN. Most of the parts I’m using didn’t exist even 10 years ago.
I hope the Wond II will be in production for many years, so I’m selecting the smallest and most advanced components for its design.
In our case the cliché, “advanced”, means these are smart components – they have more complex and nuanced behavioral possibilities. Almost every one has a little brain. I’m wiring up all those little brains to communicate with our main brain – a microcontroller which will orchestrate everything according to the motions and commands of the musician.
This whole design terrain is very exciting, but the basics have to come first. Establishing all the necessary connections. Choosing and positioning every last thing where it needs to be, etc. This is where I am now – the choosing of things is now over and I’m in the middle of arranging and connecting everything. The PCB layout process.
The next step, once I’ve finished this arranging and connecting, will be to send the design off to any one of several circuit assembly houses offering a prototyping service. In a week to 10 days after that will be testing this first prototype and loading in some firmware, (“software”). Being a programmable system, it has no idea what it is until the firmware wakes it up and tells it what to do.
Firmware development has been going on in parallel for almost 2 months now. I’m working with a great programmer located in my old Southern California neighborhood. We’ve worked together for many years before and I’m very pleased to be working again with Robert on this project.
Thank you everyone for being patient. Once the first prototype exists there will be things, finally, to see. Once we get it going it becomes all about the firmware, we’ll have videos of the actual prototype as it becomes a very interesting musical device for you to enjoy.
I’m not doing any promotion just now. Promotion will begin again once I have the prototype and can drop some iPhone videos onto Facebook and send a promo out to all the subscribers who aren’t yet supporters. Right now it is all about getting to that prototype.
I’ve given much thought to the Ideas and perspectives you’ve messaged me about – thank you for your thoughts. When you think of something, just email me at firstname.lastname@example.org and it’s in the mix.
– Paul Vo
May 16, 2017 - Battery Charging , Firmware updates and other data
Several supporters of the original Wond have expressed concerns and given suggestions for a better charging interface.
The old Wond charging jack leaves much to be desired. The Wond ii has a slightly bigger battery to charge. The ii is a programmable device so we also need data communications capability for downloading new firmware, archiving user settings, and ultimately much more.
I’ve done a lot of work on this. The very best solution is clearly USB Type C.
Everyone is at least somewhat familiar with USB. You see it with some printers, phones, flash drives – you know. There are many devices routinely charged over a USB connection. This seems to be less so in the Apple universe because of the proprietary Lightening Connector, but the future belongs to USB type C. Apple will be increasing its support of this important standard connector in the near future. Eventually, you’ll have one kind of data cable for almost everything.
So I’m designing the Wond ii with a USB type C connector. There are several levels of communication standards and power/charging standards written into the type C specification – it’s a massive document of 500 pages. The Wond ii will charge from any Type C receptacle including wall plug charging receptacles, laptops and desktops. For legacy equipment without USB type C connectors, there are all kinds of cables available on Amazon and elsewhere. The type C connector is already in use and people are using their old chargers and laptops with a “USB type C to USB type A” cable – standard patching that will be around for years or until all the old USB stuff dies out completely. When it comes to new products, Apple will lead us all to a more uniform type C environment and the Wond ii will be in it.
Data communication with the Wond over USB-C gives us some very interesting possibilities including the ability to add wireless Bluetooth communication to the Wond using a little add on plug-in device.
All this has been taking a considerable amount of time but it’s very important to get this kind of detail right. Being able to update the firmware easily and reliably is absolutely essential. As long as the firmware can be updated, the door to future possibilities is open. (Not to mention bug fixes.)
I’m pleased to announce that your support has made it possible for me to book the services of a freelance firmware engineer, a person I’ve worked with before, so I know he’s good. Robert will be implementing all of this USB implementation and other such things while I will be focusing on the core DSP and vibration control system.
I am very grateful and appreciative of all of you who are already supporting this project. And I’m humbled by your willingness to support me right from the beginning. When the Wond ii emerges from this design process and becomes a real thing that you can hear and take videos of, I’m sure we’ll pick up many more supporters who have a more ‘wait and see’ attitude about it. I’m especially grateful to you because, obviously, if everyone took a ‘wait and see’ approach, this would not be happening! Thank. You.
I haven’t been sending out campaigns or doing very much at all to advertise the project. I’m just far too busy actually doing the job of making it real. And enough of you are supporting the Wond II Project already to make this possible.
Here are some links to learn more about USB type C, etc.
Your comments and suggestions about the Wond Charging jack and USB have been very helpful. Please do email me at email@example.com to share your ideas, concerns and questions.
April 16, 2017 - Work in Progress
Firmware Updates: The plan is to make the Wond simply look like a thumb drive so you can copy firmware files using any OS without needing an app. This is the basic function that makes all future updates possible.
I’ve mostly been making decisions about which component to specify, one component at a time. Simple resistors and caps are easy, but the more complex elements take time. Nevertheless, the Wond II schematic is coming along nicely…
I have two viable ways to achieve the tactile feedback I’m looking for. I’ve chosen one…
A few days were spent designing and testing an electro-stimulation biofeedback system. Your thumb would be in contact with a small set of electrodes that would send a tiny current through the skin of your thumb across a small distance, like 1/4″ or so. By careful construction of this current pulse it really is possible to produce the sensations I’m looking for that would allow you to “feel” the string when you touch it only with the magnetic fields of the Wond. However, this idea must remain experimental for now. The sensation varies a lot depending on moist or dry hands, thickness of skin, etc. I can imagine ways to automatically compensate for these variables, but this is just too “iffy-tricky” at this stage. And of course the idea of using electricity to mimic touch may be disturbing to some of you. It was once used on some cell phone interfaces, so it’s not a completely crazy idea – but who can say what years of electro-stimulation of your thumb pad will do?
Instead of electro-stimulation, I’m going with a new motor-based technology. This is well-behaved and fully programmable. It’s not completely inaudible, (the main virtue of electro-stimulation is you hear zero), but then neither is the sound of your pick on the string. It feels great – more realistic than the electro-stimulation method because it actually produces a bit of motion of the kind you feel when your pick touches the string.
So for production Wond II, I’m going with the tactile motor devices. I may make a very few Wonds modified with the electro-stimulation idea because I do find it fascinating. Must be the Frankenstein in me.
Please remember, I value your comments and suggestions! Just email me at firstname.lastname@example.org
March 14, 2017 | In the Thick of it | A Video!!
I have a couple of things to show you now.
I’ve put together a video showing how well the new vibration control technology drives a thick bass string as compared with the performance of an Ebow and a Wond on the same string. My test instrument here is a Martin Backpacker guitar that fits nicely on my bench, but then I change to an old Ibenez bass for a more realistic test. I should emphasize this is a lab test to highlight just one aspect of the technology — it’s powerful ability to drive a heavy bass guitar string. I’m not going for tone, the device is not made pretty or compact here, the string is driven so hard it rattles against everything, etc. You probably already get this, but I just want to make sure not to give anyone the idea this is a demo of the actual Wond II. It’s a tiny isolated piece of it, significant only because it shows a capability of my new technology that wasn’t possible before.
In this video, before demonstrating the new control technology I show the Ebow and the original Wond for comparison. The Ebow produces a nice tone but with very slow attack and little volume. The Wond is more powerful — but… please watch the video:
Many bassists have expressed a desire for a hand-held device that has enough power and control to drive heavy bass strings. I’m happy to report that the new vibration control technology I developed in 2016 finally achieves this goal. The Moog Guitar and the Vo-96 will not drive a thick bass string at all. The Wond does drive it, but — well, watch and listen to the video above. The new tech – it can drive literally anything that responds magnetically. It will probably drive the side door panel of a pickup truck. (Can’t wait to give this a try!)
The development work I’m doing right now is not visually rich so there isn’t anything to show. I’m grinding through some math, and moving routines from the Vo-96 to a different DSP microcontroller we’ll be using in the Wond II. The architectural specification is also in process, though I haven’t added to it since a week ago. There is a great deal to do, and I am not cutting any corners!
If you have any comments or questions, you can reach me at email@example.com
– Paul Vo
February 21, 2017 | Coming Together | by Paul Vo
I’ve been working on the “Architectural Specification” for the Wond II. Once this is finished I’ll publish it here for comment. I’ve decided to keep things simple and ask you to just email me at firstname.lastname@example.org with any comments you may have.
I’ve lined up some people ready to step up, (as paid consultants), and help me get this product developed. In the past I’ve tried to do almost everything myself. I do have the full set of skills so this ought to result in the most fully integrated “work of art” of a product, right? Well here are some things that are wrong with this approach:
- It takes longer to do everything oneself. Nothing happens in parallel.
- It is emotionally difficult to throw away a bunch of your own hard work and do it over to make it better.
- It’s easier to be objective about the work of another, easier to bite the bullet on changes and improvements.
- It’s possible to do many easy things in parallel when you have a team. This saves time for finding better solutions to the most difficult challenges.
- In short, doing is often at odds with guiding. My most important task is to guide the Wond II Project towards excellence.
- Despite having said all this, I will certainly be personally doing a lot of the design. I’ve just started working on the enclosure shape in Solidworks. I’ll show pictures when it looks like something.
The bigger our development budget, the more great talent we can afford to bring in. We are doing well, but more is better! Your support funds are going right into creating the Wond II. If you know others who would be interested, please introduce them to the Wond II Project.
– Paul Vo
February 15, 2017 | Getting Configured | by Paul Vo
It’s Feb 15th now. I’ve spent the past week wrestling my computer systems and engineering tools into the right configuration to handle this project. All this is now ready to go. Important updates to my engineering software are in place. These days it is engineers and gamers who need the fast, deep machines. I’m ready to crunch some serious simulations.
I’m still working on the blog feature on this web page. I want to keep it simple, and tonight I’m thinking that pages are simpler to manage than posts, because posts come with automatic baggage and then it’s, “well okay, but I don’t want this string of text fragments on every post”, and it takes an hour of reading to discover how to remove it, except when you do remove it something else changes too. I’m comfortable with pages, and eager to design the Wond II rather than furthering my web page design knowledge.
Today I also took the Wond II Project public on my various Facebook pages via a Facebook Event. It’s all still low key, but we have months to raise the temperature. There will be a day when something physical is working — then we’ll have videos. And there are some well-known guitarists already on board and more to come. The more supporters the better – it will make an uncompromising design process possible – if we need something custom, the tooling cost won’t stop us. So if you’ve already joined, it’s in your interest to bring in other musicians you may know who would enjoy this journey and the Wond II product.
Okay, it’s 3:30 AM February 16th. Goodnight for now.
- Paul Vo
February 8, 2017 | Introduction | by Paul Vo
Hello! Paul Vo here,
This is the internal project blog visible to supporting members of the Wond II Project. I’ll be blogging here about the project. Visit this link any time to get the latest info.
If you arrived here by the communications test message, great! We are in communication! If you have comments about the project, please email me at email@example.com — I may not always answer but I promise I will read every comment.
Thank you for supporting the Wond II Project!
– Paul Vo