Adapting a Rane Magnetic DJ Fader for use with an Arduino?

Projects General Guidance
1

I've built a Teensy-based USB-MIDI 14-bit DJ crossfader, using an ADS1115 ADC to do better-than-stock ADC conversion.

I'm quite pleased with how it's working so far, but here are some technical problems with.

Firstly, it looks/sounds like this (in conjunction with Max/MSP):

The problem I'm running into is with the mechanical fader itself.

I'm using a 45mm 10k potentiometer by TT Electronics (PS45G-C1LBR10KN), which has a 20% tolerance, which isn't great, but I account for this in the Teensy code by having a calibration routine.

The problem I'm having is that the first 1-2mm of the movement of the fader, the actual resistance goes backwards before it goes forward, meaning that regardless of how well I calibrate things in the code, there is a bit of the throw of the fader that I can't access, making the end of the fader feel sluggish (with bad "lag" to use the turntablist lingo).

So I've been looking for alternatives and options to replace the fader with and came across the Rane Magnetic Contactless Fader.

I've emailed the company to see if it's possible to find out more information, but as far as I can tell, the fader itself houses only two hall sensors and the mechanic fader with a neodymium magnet in it:

Here is the board on its own without the fader in the way:

1000×889 118 KB

Now where I'm running into trouble is that I've not really worked with hall sensors much, nor have I ever used an OEM part this way, so I'm not sure if it will even work, and since the fader is quite expensive, I wanted to ask for some input first.

I did find a service manual which has some schematics:
http://bee.mif.pg.gda.pl/ciasteczkowypotwor/SM_scena/rane/Rane%20TTM%2056%20Mixer%20Schematics.pdf

And a patent document document:

Here's a relevant bit of text from the patent:

  1. The fader control system of claim 2, wherein the magnet has a stabilized flux density of approximately 11,000 Gauss, the flux detectors have a separation distance of approximately 37 mm, and wherein the first flux detector has a direct current output voltage between approximately 2.5 volts and 4.2 volts and wherein the second flux detector has a direct current output voltage between approximately 0.8 volts and 2.5 volts, the direct current output voltages being dependent on the linear position of the carrier.

So my basic question is would I be able to implement the hall sensors with a Teensy/ADS1115 or will there need to be other circuitry involved?

And a more specific question with regards to the code would be, would the "actual position" be something like the average between the two hall sensors?

2

You should be able to read both the hall sensor voltages with the ADS1115, one per channel and the relative voltage of each will give you the position of the fader.

I'm sure you will have to run some tests and take some measurements to see how linear the output is with respect to the position.

3

I "scanned" the schematics, but can you point us to where the fader is? (Just the page and the approximate area or quadrant on the page.) ...I see a lot of regular pots.

The patent page says:

to provide a constant, direct current voltage output.

If that's what you're doing with the regular pot, it may just be a matter of connecting it correctly (if the voltages are correct for the RANE device).

Or, maybe you can just find a better pot?

Or for crossfading between two sources (two turntables, etc.) you can probably compensate for everything in software... You don't need a "perfect" 1% / 99% mix or anything like that... Typically, you're just looking for a smooth A->B transition with 50/50 in the center.

4

blh64:
You should be able to read both the hall sensor voltages with the ADS1115, one per channel and the relative voltage of each will give you the position of the fader.

I'm sure you will have to run some tests and take some measurements to see how linear the output is with respect to the position.

That's promising. I just wasn't sure if there was other signal conditioning stuff that needed to happen before the hall sensor hit an adc.

So I would just wire it up as per any old hall sensor diagram (one to 3.3v, one to 0v and one to the ADS1115).

DVDdoug:
I "scanned" the schematics, but can you point us to where the fader is? (Just the page and the approximate area or quadrant on the page.) ...I see a lot of regular pots.

I had a look and I couldn't find it either, but I'm not great at reading schematics... I'm guess just the VCA of the mixer is represented and the circuit for the fader isn't listed?

DVDdoug:
Or, maybe you can just find a better pot?

In the interim I'm actually doing something along the lines. When I ordered that fader from digikey I ordered 5 just to be safe. And in measuring them with a meter, some are "better" than others. So I've picked the one with the least amount of "lag" and have 3d printed some tiny shims to stick in there to physically compensate for the problem (for now). Not ideal but it should at least work until I figure something else out.

DVDdoug:
Or for crossfading between two sources (two turntables, etc.) you can probably compensate for everything in software... You don't need a "perfect" 1% / 99% mix or anything like that... Typically, you're just looking for a smooth A->B transition with 50/50 in the center.

My intended usage isn't going to be for actual crossfading, so it will end up going through a lookup table-type thing anyways, as well as other signal processing to extract additional gestural information (velocity, schmitt-based direction toggle etc...).

But what is critical is having a measurement that is very responsive at the end, even if it isn't perfectly linear.

From the looks of the patent the response is kind of an "S" shape. Although that may be just from the conditioning/processing down the line. But if that's the case, I'll probably come up with a lookup table to linear-ize it, and then apply different functions to that output.

Oh, another question I had was if it is generally possible to power hall sensors off whatever voltage is available? (i.e. the Teensy's 3.3v (instead of 5v or more)). I'm assuming that the hall sensor's output will just be scaled to whatever the available voltage is, but I don't know if there's a minimum voltage for a sensor like that.