Arduinos: Some Afterthoughts

[Coenraads]

Those of you who have been following my posts, know that I've been on a mission to promote the use of Arduinos in the organs/MIDI environment. If you have some electronics/programming background, I would encourage you to take the plunge since it can be very rewarding to build your own hardware. And I have yet to destroy one despite many connection errors being made during my experiments.

I'm not a professional programmer or electrical engineer and am largely self taught in this area. My computing background goes back to FORTRAN (on punched cards), BASIC and Pascal and you'll see from my code, that the computing constructs I use are pretty elementary. Certainly not up to current C programming standards. I.e., despite the power and sophistication of C, it is still an easily accessible language.

Arduinos have a reputation of being a toy for hobbyists, but that doesn't do justice to its ease of use and flexibility. In the Arduino MEGA configuration, there are seventy pins many of which can be configured in software to perform a variety of functions.

You don't want to bother with diodes and want to parallel scan a 61 note keyboard? You can turn the pins into digital inputs and you can even turn on internal pull-up resistors to keep the pins high when the switch is open.
You need 64 outputs to drive 5V compatible drivers? That too can be done.
If you need analog inputs for reading a swell pedal, some pins can be defined to do just that.
If you need serial ports to let the Arduinos talk to each other or for MIDI IO, you have four ports to choose from.
If you need a Pulse Width Modulated output to accurately control that DC motor turning your Zimbelstern, it can do that. The flexibility of this platform is astounding.

The Arduino has three kinds of memory. Flash memory stores the uploaded program even when the power is turned off. Volatile RAM is available for running the program. And finally there is EEPROM that the program can write to and these data are then saved, like having a small SSD available. This is one capability of the Arduino I have yet to exploit. I'm thinking of writing code to turn the Arduino into a standalone combination action for use with both virtual organs or pipe organs. Stay tuned. This will probably be my last effort.

If you are using Arduinos yourself, or I have inspired you to try them, let me know what you are up to. I appreciate the positive responses I've received so far.
John

[Eric Mack]

Thanks for sharing this journey, John. I think the arduino represents some fantastic capabilities. (By the way, I started learning Fortran using marksense cards as our school did not have a keypunch. Still took 3 days to get our programs back by courier. Now, to have these tiny processors like Arduino for just dollars? That's amazing)

[Mr. Polecat]

I am planning to use Teensy microcontrollers and parallel shift registers for my VPO project. 2 manuals + pedalboard + some stop tabs and switches + a couple of variable speed leslies.

I want to replace the AC motors in the leslies with a fairly torquey DC motor of some sort (not sure what yet) and see if I can rig up something with a power transistor or something to control the leslie speed via PWM generated by the uC (or maybe use half of a stepper motor controller for that, if it can source enough current for the motor). That, plus a cardboard wheel and an optical sensor on the pulley to feed RPMs back into the uC, and I ought to be able to run the leslies at specific frequencies, which might be interesting.

However, I am a bit dubious of 3.3v keying voltage on old Allen manuals. I may have to do some transistor and voltage divider magic there, and use 24v keying voltage to keep the contacts clean.

I have the functional innards of a Conn 720 tube organ, too, and I think it would be an interesting project to eventually midify it, too, as an output device. But that is the last thing on my list, hehe. I am not sure yet how to use the uC to adjust the volume on the thing, either, at least in an analog way.

[johnbeetem]

I want to replace the AC motors in the leslies with a fairly torquey DC motor of some sort (not sure what yet) and see if I can rig up something with a power transistor or something to control the leslie speed via PWM generated by the uC (or maybe use half of a stepper motor controller for that, if it can source enough current for the motor). That, plus a cardboard wheel and an optical sensor on the pulley to feed RPMs back into the uC, and I ought to be able to run the leslies at specific frequencies, which might be interesting.

However, I am a bit dubious of 3.3v keying voltage on old Allen manuals. I may have to do some transistor and voltage divider magic there, and use 24v keying voltage to keep the contacts clean.

Motors: You might check out Brushless DC (BLDC) motors to replace AC motors. I haven't used them personally, but I've studied the technology some and it's pretty nifty. There are microcontrollers with software to produce the waveforms that drive the coils.

Keying voltages: I'm going to add VPO to my Rodgers Trio 321C at some point. The Trio uses +12V keying for Solo, Great, and Pedal. The Accompaniment keyboard uses +12V if the FF tab is down, otherwise it uses about +6V. Fun, eh? My plan is to use CD4000 series CMOS ICs, which allow supply voltage from +3V to +18V, so I can power them from each keyboard's voltage rail. The logic threshold is half the supply voltage, whatever it is. Then I only need to level shift the few connections to my microcontroller.

I can probably get SOIC parts with 50 mil pitch. If so I can design a single-sided PC board with no holes, making it really cheap to manufacture and easy to hand-assemble.

[lmat]

I am planning to use Teensy microcontrollers and parallel shift registers for my VPO project.

I used TeensyLC and 16:1 multiplexers to scan my two manuals and pedal board. I connect to the PC via onboard USB (on the TeensyLC). It works pretty well.

However, I am a bit dubious of 3.3v keying voltage on old Allen manuals. I may have to do some transistor and voltage divider magic there, and use 24v keying voltage to keep the contacts clean.

Tell me about this. What is the problem with 3.3v keying voltage? I may be having trouble in this regard, but I will wait until I hear back from you to go into further detail.

I found some of these which should get the keying voltage up to 12v without problem: https://www.sparkfun.com/products/12009

(here's my organ: https://www.youtube.com/watch?v=n-vsQvJ7Yoo)

[Mr. Polecat]

Supposedly (and I say "supposedly"), those whisker contact manuals require a voltage high enough to cause a slight electrical arc when the contacts close. The arc keeps the corrosion burned off of them. When keying with too low a voltage, the whiskers won't stay clean. Supposedly.

The manuals I am hopefully going to use are bussed, but I think I can modify them into a matrix arrangement. I figured I could just use a transistor on each of the rows to switch 0 or 24v on them, then when a switch is closed pull the column high through a voltage divider that would knock it back down to 3.3v on the Teensy's input pin. Shouldn't be too awful many discrete components sprawling everywhere, with a matrix setup. 8 transistors and 24 resistors per manual, or double that if I can make them velocity sensitive.

Maybe I am full of poo though. It's quite possible that I don't really know what I am talking about. :3

Edit: Hey, neat. I was originally using an Allen T-12, too. But then I got a Conn 720, which is significantly wider. Wide enough that I can use my stage piano for a 3rd manual, which will be good I think. I'll take out the stop tablets and put a shelf there, which the stage piano ought to fit on just about perfectly. I still want to use the Allen manuals in the Conn cabinet though, because they are WAY nicer than the Conn manuals.

[johnbeetem]

A matrix arrangement usually requires a diode at each switch if you're going to close more than two switches at a time. Otherwise there are "sneak paths": if you close three switches that are the corners of a rectangle, it looks to the scanner as if the fourth corner is also closed.

[Mr. Polecat]

Oh yeah. That slipped my mind!

There's already a diode bridge in the organ that I am planning to use. But I guess if my plan to add velocity sensitivity comes to fruition, I'll have to make a second one.

[lmat]

I am getting some inconsistency in contact, and I'm hoping that bumping up to 12v will help with that. (I have 12v available because that's how I power my audio equipment.)

[John Kinkennon]

It's been my experience that the phosphor-bronze whiskers really need the 12v keying. I've had good luck with 3.3v for the Allen 11x6 matrix manuals providing the contacts are burnished a bit if they are already intermittent. Your mileage may vary. It may be that I just haven't been successful in cleaning that stubborn tarnish off the "whiskers".

[johnbeetem]

What is the problem with 3.3v keying voltage? I may be having trouble in this regard, but I will wait until I hear back from you to go into further detail.

There's an interesting summary of what contacts are doing on at the microscopic level at Wikipedia: https://en.wikipedia.org/wiki/Electr...contact_theory

Basically, all contacts are resistive to some extent. My understanding is that if you have low voltages you really should have gold-plated contacts and you must have low currents so the thin layer of gold doesn't burn off. DC switching is quite nasty at the microscopic level: when you turn a switch off it draws an arc and if the arc persists it damages the contact surfaces.

Non-gold contacts have higher resistance. One way to compensate is to use higher voltages. My understanding is that the resistance will have less effect at the higher voltage, and the higher voltage burns off oxidation and prevents build-up. It's also good to have "wiping contacts", where the contacts slide against each other to scrape off oxidation.

[lmat]

My original idea was to wet the contacts with mercury; I'm pretty sure that would solve all my current problems. Unfortunately, it would bring an host of new problems xD

Great, thanks for the input. (Sorry for the delay, I'm not getting e-mails on this thread; hopefully that's rectified now.) My currents are very low, and the whisker contacts in the Allen T-12 do rub each other. In fact, I would think the constant rubbing might increase the a-spot?

...the higher voltage burns off oxidation and prevents build-up.

Is 12v considered high voltage? I can't imagine much arcing and burning at 12v and the high resistance values I'm using. Even without the oxidation burn-off, I'm quite confident it will be better.

Thanks again!