Panels edited for 2.5mm mounting holes here. Ive submitted a pull request @infovore
heads up: I’ve added a note up top and on the project website to suggest that if you want to use hashtags elsewhere on the internet around this project,
#16nfaderbank is probably a better bet than the shorter option, for reasons.
I just fixed mine
I look forward to having a reason to use this hashtag.
Thanks for doing the legwork
Just fixed mine up
For anyone using with Teletype, the address needs to be changed for I2C communication to work
Thanks for this, all.
I’ve got a PR open with the new addresses for 16n and Ansible in there. Once @bpcmusic or @Nordseele have eyeballed that I’ve done the right thing, I’ll get that all merged into master and update the main release with correct addresses. I’ll also do a
.hex release of the master mode firmware.
There are a few issues on the Github I’m working through now I’m out of a super-crunchy period of work. Hopefully can wrap all that up within the week.
Thanks for this - I am sorry that I didn’t catch that things had changed. I wish that I’d taken the time to test all the latest firmware - but I’m so late on this expander build that I’m not allowing myself any distractions till I’m done. Looks like @Nordseele has had luck with the proposed address changes. Let me know if you need anything from me. I’ll happily pause to lend a hand if needed.
the brilliant think about many eyes is that they make light work of bugs. So: if you have firmware that dates from before right now, I2C is broken for you. The build we shipped, it turns out, had some old I2C addresses in that didn’t work.
The current code in
master on Github fixes the I2C bug, both in MASTER mode (for ER-301) and non-master mode (for Teletype). You can go and get it now.
The DXF files for the panels now have the correct size holes.
Finally, there’s a new feature in a branch that I’d like you to test if it’s of interest: the ability to use the faderbank upside down. The
REV flag in
config.h already let you reorder the faders left-to-right; this new branch adds the ability to set a
FLIP flag to reverse the read direction of faders. What this means is you can set
FLIP, and then the faderbank will work upside down: useful if you want to place the I2C connection, rather than the CV jacks, at the ‘top’ of the board.
I haven’t merged flip into master yet. It’d be good if some more of you could test the flip branch which as mentioned, is here. I think then I’ll be in a spot to merge that to master and build a new version of the firmware.
This also means I’m probably going to split out hardware version numbers and firmware versioning.
Finally: I fixed the DXF files in the repository. They have the M2.5 holes they should have had, not the giant countersink-sized holes.
Would this reverse behavior of the CV out jacks as well? I could see it being slightly confusing if not, as that’d make the i2c and cv behavior inverse. That being said, I could see that slight confusion may be outweighed by the potential ergonomic benefit.
No. The CV outs are raw, meaning no code will ever alter what they do.
You are right: this is confusing if you are using the CV outs simultaneously with I2C or MIDI. The functionality was added for people who just wanted to use I2C and reorient the faderbox so the cable was in a better position.
I agree it’s confusing, but only if you’ve forgotten you’ve got an upside-down firmware on the board later. For most people, this is not a feature of huge relevance. I was just looking for testers.
Forgot to say: big shouts to @Nordseele for some good bug reports, some patience as I fixed things, and his code suggestions.
Thanks for this quick update, yes it’s an ergonomic improvement but it’s really useful. It reduces the length of cable required to connect to a TXb (or a fake TXb in my case). I didn’t encounter any issue with a much longer cable (150cm) but less cables on the desktop is better… there’s already the keyboard, the grid… well, it’s better now
Do you think the Teletype TXi PARAM OPs will ever find their way onto the FB?
They could - depends if we want to add that complexity to the 16n. Which ops would you be interested in.
I just finished two of these. One works without any issues, on the other one there’s a weird interaction between the first pair of faders (from the left) and the second to last pair of faders where when the second to last pair is down at 0V, settings on the first pair don’t result in any output (edit/clarification: both on the CV outputs/jacks, but via usb/midi) and when the second to last pair of faders is high, the outputs of the first pair of faders produces wild jumps between 0V and ~3.5V when I pull the faders up (I hope that description is clear? I’m having some trouble coming up with a better one).
It sort of seems like a short to me, but I’m a bit unsure about where I could’ve even shorted these two pairs out. I can’t see anything immediately on the board, but maybe I’m not looking in the right place. Anyone with more intimate knowledge care to give advice? I’d really appreciate it.
Already very happy with the other one, though. Really cool project!
I briefly had a similar issue with my first build. You should reflow the IC’s and check that all the pins of the faders are properly soldered.
Yep, that did the trick. All well now
I’ve had this problem too. From what I vaguely remember: it means a joint on an op-amp isn’t right, and so the divided-down signal isn’t being fed into the mux - meaning that sometimes 5V-ish goes into the mux and you get crosstalk on the channels. This is what that little webmidi webpage was useful for - working out which IC probably had a dud connection and was making a mess.
Glad you got it sorted.
Yeah, this is a really nice solution. I’m going to work on a release.