#355

Hey 16n-peeps… I have Teletype-backpack-TXb-ansible-ansible-TXi-TXo-ER301 all daisy chained. I have tried powering the Teletype/backpack alone and also the TT and TXb simultaneously. In both cases, if I have the 16n jack plugged-in, Teletype hangs. Also hangs if I leave the power off the 16n until everything else is booted. Any suggestions? I made sure I’m running 1.33 firmware.

#356

double check your I2C connections?

#357

If I understand your setup here - you’re using BOTH the TT-backpack and the TXb?

In this case don’t power the txb.

Or don’t use the TT-backpack at all and connect everything to the txb.

#358

Yea, I tried all manner of combinations. Looks like it’s working now with NO backpack and powering the TXb. Also, I have to have the 16n powered up before booting the case. Go figure. It’s working for now!

Thanks.

#359

Huh. Maybe your 16n has pull up resistors installed?

#360

Mine came from Michigan Synth Works. If so, how should that change the behavior? I know you need those to do a direct connection to the ER301.

#361

The pull-ups provide power to the i2c bus. The TT-backback, the TXb and the 16n (if pullups are added) all can provide that. Things can sometimes get flakey if you have more than one of those providing power and pull-up resistance.

Best practice is probably to have only ONE device in your i2c bus providing power and pull-up resistance. Of course this all depends on what’s on your i2c bus - if you have many devices, you may need additional pull-up resistance.

NOTE - there’s many variables in play with these setups and everyone’s situation will be a little bit different.

1 Like
#362

Yea, I have a pretty full i2c bus, so I’m familiar with how it can be a little flakey.

Doesn’t look like I have those resistors on mine. So as it is now, I have only one pull up in the system…TXb.

#363

just to rule this out, are you using the leader or the follower firmware on 16n?

#364

I reinstalled the Teletype friendly (non-master) firmware, just to be sure of that.

And for what it’s worth, things seem to be working for now…

2 Likes
#366

Just tried my 16n for the first time and having an unexpected behaviour with the USB midi - each slider outputs slightly less than the full range it should be controlling. For example if I map a volume gain on Ableton track channel with default settings (min set to -inf dB and max set to 6.0 dB), the highest value sent is 5.4 dB. I get the same behaviour if I try to do something similar with Norns.
Is this to be expected or may there be a solution somewhere?

1 Like
#367

I noticed something similar, the value can be 1 or 2 down on maximum depending on the usb voltage. If there are other devices on the bus, then its not seeing a full 5.00volt. I had an (difficult to isolate) issue where i had an arc on a usb hub with varying current demand. At higher currents voltage can easily drop 100mV or more - solution was to redistribute my usb devices.
If it used a voltage reference this might be able to be avoided. I haven’t looked at the code but imagine it could be addressed in software -for example to ensure values above say 4.8v are always registered as maximum

#368

The CV end of the device uses the 5V from the USB power which is far from a clean, reliable voltage source. A simple calibration function would fix it. The ADC on the teensy isnt that great either, that always has to be trimmed (i.e. OC/TU etc) since those offsets vary from device to device

#369

Damn, I was about to buy a behringer BCF2000 or the new X-Fade thingy once I found a decent one on ebay but now I think I need to do this. It will be for Kontakt control for orchestral libraries mainly.

#370

This is both to be expected and there is a solution here. Solutions to ‘scaling MIDI output issues’ (cc @nattog, @jonnymon) follow:

For reasons, the raw 5V going through the fader (and out to modular) is run through a voltage divider so as not to damage the Teensy. This scales it to ~3.2V p-p, but, because it’s a voltage divider, its output may vary and fluctuate depending on your board, power supply, discrete parts.

These two config lines set the boundaries for the scaler in the code; the typical absolute maximum is 8192, which would represent 3.3V coming into an analog input. We take a slightly lower number as a maximum to reflect that voltage division.

So: if your board isn’t quite giving you full 0-127 p-to-p, you’ll probably need to reflash it having fettled the config.h to set MINFADER/MAXFADER values that give you desired results. It’s not currently possibly to adjust this without recompilation: the magic of compiled code.

(Yes, this is not perfect. It’s a DIY product. Given my time over, I think I’d definitely feed a ~3.2V signal calculated identically into an analog reference input and calculate it from there. I didn’t. It’s on the list for obvious improvements in any 2.0 that may happen. But this would avoid various trimming issues)

5 Likes
#371

If you can find something affordable with motorized faders, do not pass that up.

Especially for a MIDI library that wouldn’t benefit from CV’s higher resolution anyway.

1 Like
#372

Thanks for the advice, the IF is the big problem in these parts though. Plus come on, this looks cool!

1 Like
#373

There is room in your heart for more than one bank of faders. :slight_smile:

1 Like
#374

:joy: I’m afraid there is room in my heart for far too much gear.

1 Like
#375

is it also possible to configure CV ranges this way?

My faders seems to starts a bit above 0 in ER-301

1 Like