24 motorized fader mixer with Arduino

Great. Thanks alot, OP. Now I have another project on my to-do list. $)

Where are you getting these faders anyway? And I second the question -- how much are they costing you? If they're built worth a crap, that looks like quite the feature set.

Where are you getting these faders anyway? And I second the question -- how much are they costing you? If they're built worth a crap, that looks like quite the feature set.

They are some faders from an old mixing console. My University proffessor got them few years ago and they are working allright. No idea about the cost, but is really hard to get this kind of components in Argentina in an affordable price.

f you use the AD7147 chip to detect the touch, you don't have to periodically read in your Arduino loop.
The chip can set an Arduino interrupt when anything is touched or released. (page 29)
( www.analog.com/static/imported-files/data_sheets/AD7147.pdf )
It handles 13 touch contacts so you would need 2 of them.

That would be ideal. I hope I can get them here in Argentina.
If I dont, maybe I could get 1/0 data from touch sensitive track and activate writing mode from code (no idea how to code it yet) in the touched track. But yes, this way means that the arduino loop must read periodically those 24 digital inputs.
I could try to buy those chips in sparkfun and ship them to argentina, but electronic importation is really hard here as well.

Thanks to you all! And sorry for any grammatical or explaining mistakes.

I will get the sample tomorrow and start to do some tests this weekend.

Mytx:
They are some faders from an old mixing console. My University proffessor got them few years ago and they are working allright.

Just my luck. I haven't found a place that sells motorized faders. I guess that trend hasn't hit the DIY market yet.

And sorry for any grammatical or explaining mistakes.

I'm always amused when folks outside native English-speaking countries say this. The wording is sometimes a little peculiar, but the text is usually spelled and grammed better than just about any American poster on the Internet. :wink: As if I could even write a coherent sentence in any language outside my native tongue anyway.

regarding buying faders, you can buy them from any brand that uses them on their mixers/controllers ?!? depend on what you want !
Behringer probably will be the cheapest.

Sorry It took me so long to come back here.

I have the fader, I attached some pictures. The fader is different from the schematics page my professor gave me. It has 1 more pin.

I can't find any schematics of this model on the web, please, if anyone finds it, that would be really helpfull.
I don't know how to find out witch signal is in each pin. As this is an old and expensive fader, I dont want to mess up and burn something. Any suggestions on this?

This far, I reached to move the servo motor one way with a 9v battery. I will add an H-bridge to make it bi-directional.

Next steps (after knowing each pin function)

  • Controlling fader servo from Arduino PWM output bi-directionally
  • Reading Analog input data from position fader
  • Reading Touch sensitive track

After this, I will start messing with software and midi or OSC protocols.

Cya!

http://www.faders.com/
Go straight to the brand and ask them. Im sure they will be willing to help.

Seems they were from a top console called API ( a very valued brand mind) Ill get in contacrt with an expert on them.

Thanks iyahdub!!
I just sent them an e-mail asking them for a datasheet or at least a pinout connection explaining.

Yesterday I made the motor run one way using an Arduino mega PWM output and controlling PWM pulses with a rotary potentiometer.
Strange thing happened, when the PWM output value was less than 255, a high frequency sound was made by the servo while it was running in a slower velocity.
Could this sound has something to do with a non-filered output? Or this could mean that this servo wants only high velocity pulses?
I will try to get a MC33030 chip as soon as possible so I can control the servo both ways with the rotary potentiometer.

ive asked to some people who actually work and have much more experience with those given consoles.
One of them said he had no experience with the specific ones, so im waiting for the others.
If i can, ill help !

Mytx,
Why not try bi-directional direct from the ARduino using 2 digital pins. No PWM.
Two pins, one to each wire side of the motor. Hi, Lo moves one way; Lo, Hi moves the other way.
No pulsing, no PWM. I don't think the fader motor was designed for that. Full DC only. Instant reversing.
Pulsing works well on most dc motors because of the flywheel effect. The momentum of the heavy shaft, motor armature, or even flywheel smooths out any effect of the pulsing. These faders have none of those. They're designed to immediately stop when current stops.
Some lower frequency component of the PWM frequency resonates with the mechanics and causes the whine.

Note that though the ARduino pins can source or sink current, many of the digital pin expansion chips will only sink current (to ground).

OK, here is what i got as response.
"These were used on multiple different consoles as an aftermarket add on.. I recall several of my customers who used them and I even borrowed rigs to demo at trade shows...
I suspect the small connector is for the audio fader and larger connector the motor drive and servo track.
It may be less simple to chase down motor control interface, and roll DIY control software.
*IIRC the cheap alps motorized faders used simple up/down motor, these were more complex IIRC but it's been decades since I've seen or touched one." *

Another
"Due to the placement of the cables I suspect the circuit on the pcb is a motor driver as opposed to an audio buffer or amp. Aren't the parts assembled on any of your faders? Then you have to build a driver yourself anyway. Simple H-bridges are readily available for cheap and completely integrated, they are easily controlled from any µC using PWM or a DAC if on-board. Depending on the expected supply voltage of the motor and the parts you use you might need an additional level shifter.
Still, I'd try to get hold of some sort of documentation/datasheet. If not for the whole fader, if the motor has a parts number printed on it you might get lucky and find a datasheet at least for that."

In any case the addition of PWM is unnecessary and complicates the project, no to mention adding audible vibration and possible picked up audio noise. Who said you need to slow down the slewing at all? In fact I think the MC33030 chips are either off, full forward or full reverse until the feedback slider voltage matches your control voltage.

Therefore a simple test sketch is in order. Use two digital pins (A and B) connected to the two motor pins.
Say from your 9v battery test it takes 1/2 a second for the slider to go 0 to full.
In your arduino loop, output:
Pin A High and Pin B Low,
delay 250 ms
Pin B High and Pin A Low
delay 300 ms
Pin B Low

delay 5000 ms
The slider should zoom to about halfway, zoom back,
and then repeat after 5 seconds

With all this someone had to note the fact that seems only two wires coming out of the motor ?!? If so, should be quite straight forward !!

iyahdub:
With all this someone had to note the fact that seems only two wires coming out of the motor ?!? If so, should be quite straight forward !!

It is actually quite straight forward, I already know witch pins are connected to those two wires, they are the first two of the large pin header. What I don't know is the rest of the pins functions

In any case the addition of PWM is unnecessary and complicates the project, no to mention adding audible vibration and possible picked up audio noise. Who said you need to slow down the slewing at all? In fact I think the MC33030 chips are either off, full forward or full reverse until the feedback slider voltage matches your control voltage.

Digital control is a really good idea, but I have some dobuts on this. I started using PWM because I thought it would be better in cases of slow movement of the fader. For example, in a 10 seconds fade in. With PWM the servo travels from one point to the other in a constant movement. What happens if I use digital?

Ideally, the servo should be able to move 1/1024 of fader's full length with a digital input. thats like 0,01 cm per step. I don't know if this is possible yet... But I dobut it, is a really small distance

I will try what Techylah suggested today and give you some news soon

Thank you for helping guys

For example, in a 10 seconds fade in. With PWM the servo travels from one point to the other in a constant movement. What happens if I use digital?

That's a good point, Mytx, but for the full DC voltage approach. This type of slow fade feature comes out more reliable and more flexible if implemented in software.
You would have a software routine that takes as arguments the index of a stored setting, and a number of seconds with which to fade, and possibly a time increment, probably defaulted to say 100 ms. The routine would then go in a loop which computes a new temporary set of values, each n 100ths of the way there, where n is the loop index and repeatedly sets all the sliders to those intermediate values.

That way the overall fade takes exactly 10 seconds in 100 imperceptible steps and you don't have to worry that the fader motors will get stuck if given only 25% PWM. Or that it is hard to calculate the PWM that corresponds to 10 seconds.

Another feature (for once you have this linear fade to Settings #3 in n secs command).
You can do a non-linear or exponential fade. In the above example of a loop 100x, instead of using the loop index "n" for the number of 100ths of the way there, use "en", where en = (int)(0.5 + pow (n/100.0, 1/2.0) * 100).
This zooms faster at first and slows down towards the end. (example: at index n=25, which is 1/4 of the way to 100, en is the square root of 1/4, or 1/2 of the way there, or index 50. You can even change the exponent by increasing the "2.0" for more effect or bringing it down to 1.0 for none or regular linear fade.

Wouldn't the fader positions be in a MIDI / automation track? You wouldn't be calculating long-term fades on the uC, that'll already be in the automation data. You just need to go to a given position.

Well, almost no calculation.. The only interpolation is how long it should take to get there (slew rate) given large or small differences from the current position, and whether there should be any non-linear acceleration applied.

SirNickity:
Wouldn't the fader positions be in a MIDI / automation track? You wouldn't be calculating long-term fades on the uC, that'll already be in the automation data. You just need to go to a given position.

Well, almost no calculation.. The only interpolation is how long it should take to get there (slew rate) given large or small differences from the current position, and whether there should be any non-linear acceleration applied.

Youre quite right there. This is not the case of an automated, self sufficient console, but instead one to use with DAW's.
So the console does take care of the audio and the motor translate a given data from the midi automation

Hello people

Well I've been doing some testing yesterday to understand pinout configuration. The only thing I'm missing is about the touch sensitive (TS) track. How many pins should carry TS singal?

Doing some tests in this 3 free pins I just got a 3 milivolt difference between touched and untouched status of the TS (feeding sensor with 3.3v from Arduino. Is too small difference, dont you think? Maybe is enough to pick it up with AD7147 chips Techylah reccomended?

I've been calling many electronic shops here but no luck with MC33030 chip... I may order it in united states and try to ship it, but if i get an alternative chip with similar functions in Argentina is way easier.

I just got a 3 milivolt difference between touched and untouched status of the TS (feeding sensor with 3.3v from Arduino. Is too small difference, dont you think? Maybe is enough to pick it up with AD7147 chips Techylah reccomended?

I suspect that the 3mv may just be noise pickup, with body acting as an antenna. This is unreliable and is why capacitance sensing is used.

If you can't get the AD7147, which is a special purpose single-sourced chip, perhaps you'd like to roll your own capacitance sensor.
This circuit uses the much more available hex schmitt trigger chip, 74HC14, but uses a bunch of additional components. Since you are using multiple sensors, you can save and have them all share the oscillator in the top part of the schematic. They would all need their own lower part, the part that includes "touch button".

Alternatively, this wiki page lists some other difficult-to-get chips.