ornaments+crimes extended / polymorphic 4x16bit CV generator

From circuitbending to homebrew stompboxes & synths, keep the DIY spirit alive!

Moderators: Kent, luketeaford, Joe.

Post Reply
User avatar
pld
Super Deluxe Wiggler
Posts: 1077
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by pld » Thu May 14, 2020 4:49 am

wuff_miggler wrote:
Thu May 14, 2020 3:22 am
just wondering if anyone experienced in the zen of o_C can let me know where the best thread/place would be to post about prospective collaborative concepts for user created firmware for this beast?
Hm, between this and "user" thread there hasn't really been a "best" place for firmware development discussions.

Part of that is perhaps because (IMO) the firmware is kind of a one-off and was never intended as a base for long-term or community development. At least not without some fairly extensive groundwork and cleanup (some of which I have started in private github branches), but the hardware platform will always be limited. And of course Hemispheres created its own abstractions.

User avatar
wuff_miggler
Super Deluxe Wiggler
Posts: 1285
Joined: Wed Feb 19, 2020 11:11 pm
Location: Disney, Australia

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by wuff_miggler » Thu May 14, 2020 5:01 am

all good! Thanks for letting me know pld! :)

i love the skews which currently exist :D i guess my main question was regarding adding buttons, small controls on front panels which i'm working on a sketch on now. will start a new thread in music DIY

Sideshow
Common Wiggler
Posts: 144
Joined: Sat Aug 01, 2015 3:17 pm
Location: Norway

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by Sideshow » Tue May 26, 2020 12:59 am

Dais3 wrote:
Sat May 02, 2020 4:32 am
Hi all
I've just completed building a O&C module for MOTM format. All good, but what I'd love to do before I design a panel for it, is to switch to using a 2.4" display - essentially the SSD1309 (I'm old and my eyesight is terrible). It is very similar to the SSD1306, but different enough it doesn't work 'out of the box'. Looking back on the dev history I can see O&C firmware was using U8glib which supports the 1309, and that pld has switched to a custom driver based on potions of that lib to optimise for O&C only on the SSD1306/SH1106. I'm happy enough getting into some code and compiling (I was a developer many moons ago), but wondered if a) anyone has done something like this, and b) if pld or anyone else has some pointers on the complexity of doing this, and if feasible how difficult/easy it might be?
many thanks
Cool!
I've been thinking about porting O&c to MOTM-format myself complete with PCB's to go flush with panel and PCB-mounted jacks.
In my panel mockups (which is all I've really done) it's planned to be a dual O&c behind a 3 unit width panel.

However if you are already porting O&c over to MOTM, I might just jump on (if possible).
Also I do not really have time to jump into making O&c for a while anyhow..

Please keep us updated over at the 5U-forum, I guess there are other ponytailers who would be interested in this aswell.

-Peas out :rastanana:
What is this?!?!?! A synthesizer for ants?!?! It needs to be at least.... ....three times this big!!
- J3RK / Zoolander

vcowl
Learning to Wiggle
Posts: 4
Joined: Sun Jun 28, 2020 1:53 pm
Location: Minneapolis, MN USA

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by vcowl » Sun Jun 28, 2020 2:02 pm

Hello, first time post but I've been leaning a lot from the search history as I build my system!

I'm working on a uO_C kit from synthcube, and running into a few problems. I'm wondering if anyone may be able to help.

First off, what's working:
- Encoders, buttons, screen OK
- No problem flashing teensy and navigating menus
- T1/3 respond to clock
- C1-3 respond to CV in debug and references are OK

What's not working:
- All outputs stay at ~11.2V constantly
- T2 and T4 don't respond to clock
- C4 doesn't respond to CV

What I checked:
- reflowed DAC pins
- all IC orientations appear to be correct
- I couldn't see any missing solder joints

Unfortunately the standard O_C guide doesn't help me much with test points since I'm not quite good enough at electronics to translate between the original and micro boards. This is also my first SMD project, so it's entirely possible I did some joints wrong.

Thanks in advance for any ideas!

TL072
Learning to Wiggle
Posts: 44
Joined: Sun Sep 06, 2015 3:42 pm

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by TL072 » Sun Jun 28, 2020 6:13 pm

vcowl wrote:
Sun Jun 28, 2020 2:02 pm
Unfortunately the standard O_C guide doesn't help me much with test points since I'm not quite good enough at electronics to translate between the original and micro boards. This is also my first SMD project, so it's entirely possible I did some joints wrong.
I know this is going to be a really annoying and tedious answer but 95% of problems I have had with SMD builds are eventually from soldering problems with fine-pitched pins.

I swear to you check some of my forum posts if you do not believe me. You will think you have checked the pins 2-3 times - but some how after you post a help on an internet forum you will try again and realise that somehow you missed something. For Micro o_C I think it is the DACs that are the tiniest pins.
  1. check all the fine-pitched pins for bridges.
  2. Use a multi-meter and if you can check for continuity between the pad and the pin (e.g. not enough solder)
Most of the problems I have had with SMD come down to either a solder bridge between pins or a pin that is not actually soldered down. For (1) If you do not have one already buy a cheap loupe to inspect them (I recall <10 USD/EUR). I wind a piece of wire around the ends of a multimeter probes if they are two big and go along the fine-pitched stuff to make sure they are not bridged (occasionally two pins are actually to ground so check against the schematic if you find continuity between two neighbouring pins).

For (2) above, you can often touch the top of the pin and the trace independently and if there is no beep you know exactly where one of the problems is. The issue becomes particularly clear if a little bit of pressure (very light!) on the pin leads to continuity. The way to solder those down is often to get a bit of solder on the iron and if simply flowing the solder on is not enough give again - very very light - pressure on the pin. Sometimes this has been enough to get the pin soldered.

User avatar
pld
Super Deluxe Wiggler
Posts: 1077
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by pld » Mon Jun 29, 2020 12:27 am

Yeah, "recheck and/or reflow and/or clean yer board" as default answers are annoying, but often successful :)
There is a dedicated thread for the uo_C that might have reference points.
~11V anywhere but the rails is likely a short, but can get there with incorrect components around the DAC.

User avatar
Altitude909
Super Deluxe Wiggler
Posts: 3761
Joined: Wed Aug 24, 2011 5:13 pm
Location: Meesheegan

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by Altitude909 » Mon Jun 29, 2020 6:18 am

open connections on the 2172s can give you those numbers also

vcowl
Learning to Wiggle
Posts: 4
Joined: Sun Jun 28, 2020 1:53 pm
Location: Minneapolis, MN USA

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by vcowl » Mon Jun 29, 2020 2:21 pm

Thanks for the ideas! I found that pins 14, 15, & 16 on the DAC were testing positive for continuity between each other. I've tried desoldering them but am having a tough time getting it with braid, so I've used this dilemma as a justification to buy a cheap hot air station :) I'm going to try completely removing the DAC, cleaning up the pads, and attach it again.

I tried reflowed a suspicious leg on one of the 2172s as well.

Thanks again for the quick replies - I'll check in with how it goes.

User avatar
emmaker
Ultra Wiggler
Posts: 834
Joined: Sat Mar 10, 2012 5:07 pm
Location: PDX

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by emmaker » Mon Jun 29, 2020 2:30 pm

vcowl wrote:
Mon Jun 29, 2020 2:21 pm
Thanks for the ideas! I found that pins 14, 15, & 16 on the DAC were testing positive for continuity between each other. I've tried desoldering them but am having a tough time getting it with braid, so I've used this dilemma as a justification to buy a cheap hot air station :) I'm going to try completely removing the DAC, cleaning up the pads, and attach it again.
You need to look at the schematics.

TL072
Learning to Wiggle
Posts: 44
Joined: Sun Sep 06, 2015 3:42 pm

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by TL072 » Mon Jun 29, 2020 5:53 pm

vcowl wrote:
Mon Jun 29, 2020 2:21 pm
Thanks for the ideas! I found that pins 14, 15, & 16 on the DAC were testing positive for continuity between each other. I've tried desoldering them but am having a tough time getting it with braid, so I've used this dilemma as a justification to buy a cheap hot air station :)
I think this basically happened to me at some point as well.

For de-bridging, there is really goey flux that is almost like honey which i found to be really good and kind of separating the bridged legs almost like dish soap in oil and then the oil wicks onto the iron. This stuff. I think it was the first solution that worked for me and then at some point I needed the hot-air gun (this one works for me still but I am sure based on this recommendation many actual electronics people are waking up in a cold sweat and not knowing why)

vcowl
Learning to Wiggle
Posts: 4
Joined: Sun Jun 28, 2020 1:53 pm
Location: Minneapolis, MN USA

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by vcowl » Wed Jul 01, 2020 9:11 am

Checking out the schematics and re-checking continuity between all the DAC pins, I'm no longer convinced there's any bridges or open connections on the DAC, but I may try reattaching it eventually if other things don't work. I reflowed all four of the transistors for the trigger inputs, which seems to have fixed one of them, but one is still not working. I reflowed all the connections on the 2172s as well, and still there's the problem of all four of the outputs staying constantly high at ~11.2V. Looking at the schematic, I don't really understand what components are shared between all four outputs that would cause them to behave that way, especially since there are two separate opamps on the outputs. Any suggestions on the most likely suspects?

Thanks for all the ideas so far! This is my first time doing any real troubleshooting on a board, since my other projects have been much simpler through-hole builds and just happened to work the first time :) Learning a lot.

ploubus
Learning to Wiggle
Posts: 3
Joined: Sat Feb 08, 2020 11:56 am

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by ploubus » Wed Jul 01, 2020 2:28 pm

The o_C was also my first smd build and I too experienced the constant voltage on all outputs. If I remember correctly this was due to several bad solder joints along the op amps and some resistors as well. What helped me tremendously was that I loaded up the schematic and looked at just one output on the board and tried to figure out which components are involved in this circuit, tracing lines on the pcb and rechecking with the schematic. I then just reflowed all joints along the way and it worked out in the end.

Funnily I also ended up soldering a capacitor face down in place of a resistor at one point due to accidentally tossing around some stuff on the desk and ended up reflowing half the build before spotting it. That was fun.

vcowl
Learning to Wiggle
Posts: 4
Joined: Sun Jun 28, 2020 1:53 pm
Location: Minneapolis, MN USA

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by vcowl » Fri Jul 10, 2020 8:59 am

Got it working! Ultimately, I had a bad joint on the rectifier (found useful troubleshooting info on the uO_C thread), that wasn't supplying -11.3V as it needed to, as well as a bad joint on the ADP150 that wasn't supplying the DAC with 3.3V. In both cases I had tested for the reference voltages on top of the joint rather than testing at a nearby via or component, which gave me false confidence that they were fine.

Thanks much to everyone that helped with ideas!

oldgearguy
Veteran Wiggler
Posts: 663
Joined: Thu May 26, 2016 6:42 am

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by oldgearguy » Mon Oct 19, 2020 7:09 pm

ok, I must be a stump about this, but how do I stay on and then enter the Calibration menu?
I hold down the left encoder on power up, I briefly see the two choices, but then it exits quickly to the last app I had loaded.

What do I do once those two choices are on the screen and the timer bar is going down?

firmware 1.3, 5U build (if that matters)

snowtires
Common Wiggler
Posts: 216
Joined: Fri Jun 03, 2016 12:58 pm

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by snowtires » Mon Oct 19, 2020 11:45 pm

oldgearguy wrote:
Mon Oct 19, 2020 7:09 pm
ok, I must be a stump about this, but how do I stay on and then enter the Calibration menu?
I hold down the left encoder on power up, I briefly see the two choices, but then it exits quickly to the last app I had loaded.

What do I do once those two choices are on the screen and the timer bar is going down?

firmware 1.3, 5U build (if that matters)
Do the encoders and switches work normally, otherwise? Can you get hold+left encoder to do what it should in the apps? If the screen saver goes on, does it stay on if you don't touch the module or does it turn off on its own? I had an issue similar to yours and I just had to reflow the switches one by one until I found the one that was bridging closed.

User avatar
pld
Super Deluxe Wiggler
Posts: 1077
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by pld » Tue Oct 20, 2020 1:37 am

Yeah, sounds like switch isn't working. IIRC it should also highlight the selection when the L/R encoder is held during the countdown.

oldgearguy
Veteran Wiggler
Posts: 663
Joined: Thu May 26, 2016 6:42 am

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by oldgearguy » Tue Oct 20, 2020 4:51 am

snowtires wrote:
Mon Oct 19, 2020 11:45 pm
oldgearguy wrote:
Mon Oct 19, 2020 7:09 pm
ok, I must be a stump about this, but how do I stay on and then enter the Calibration menu?
I hold down the left encoder on power up, I briefly see the two choices, but then it exits quickly to the last app I had loaded.

What do I do once those two choices are on the screen and the timer bar is going down?

firmware 1.3, 5U build (if that matters)
Do the encoders and switches work normally, otherwise? Can you get hold+left encoder to do what it should in the apps? If the screen saver goes on, does it stay on if you don't touch the module or does it turn off on its own? I had an issue similar to yours and I just had to reflow the switches one by one until I found the one that was bridging closed.
So just to confirm - when the menu pops up, it should 'just stay there' without me having to do any othe rbutton pushes or anything?

Good suggestions.

More detailed info -- I have a pair of O_C's (FSFX builds). They've been working normally for regular app duties since I bought them.
I decided to change one to more closely conform to 5U specs, (0 - 9v on the cv I/O instead of -3.5 to +6.5), so I swapped the resistors out and rebuilt the firmware with the 0to10V flag set (or whatever it's called). Successfully flashed the Teensy and powered it up to calibrate. The knobs/switches worked fine before. I have not checked the behavior of the other stock O_C, but I'll do that today as well as pull the updated one and doublecheck I didn't dislodge any connectors or have any stray bits of solder on the board.

FWIW, I've been repairing my own gear for 20+ years and have good tools and techniques and my day job is writing software, so I'm fairly confident the resistor swap went OK. The firmware was pulled from GitHub last week and built without a problem.

User avatar
forestcaver
Ultra Wiggler
Posts: 947
Joined: Tue Mar 14, 2017 4:41 pm
Location: UK
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by forestcaver » Tue Oct 20, 2020 6:46 am

oldgearguy wrote:
Tue Oct 20, 2020 4:51 am

FWIW, I've been repairing my own gear for 20+ years and have good tools and techniques and my day job is writing software, so I'm fairly confident the resistor swap went OK. The firmware was pulled from GitHub last week and built without a problem.
With so much experience and knowledge, I’m sure this is redundant, but it’s the first thing anyone will think and ask - are you using the correct toolchain version and optimisation settings? (Embedded stuff can be weird with different toolchains and due to the o_c design it can be exquisitely sensitive to them (but there is an alpha/beta branch that is less so...). To be fair, your issue sounds less like the more common weirdness that happens with a non-supported toolchain version...

User avatar
pld
Super Deluxe Wiggler
Posts: 1077
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by pld » Tue Oct 20, 2020 7:08 am

Also, while at first glance that #ifdef looks purely cosmetic and shouldn't have an effect on the buttons, the amount of "test" coverage with non-standard build flags is pretty minimal (e.g. I've never even compiled it) so I'd definitely try the regular firmware as a sanity check.

oldgearguy
Veteran Wiggler
Posts: 663
Joined: Thu May 26, 2016 6:42 am

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by oldgearguy » Tue Oct 20, 2020 8:41 am

all good points. I didn't state my quals as bragging, but just to focus on other possible non-obvious issues, but you all are right -- it was late in the day, I was moving through the process quickly and I should go back and recheck one thing at a time. I did spend time pulling down the recommended/required versions of Arduino, etc. but I'll doublecheck the settings as well.

Given I have a stock working one should also help troubleshoot. Since none of the hw mods happened on the front panel board nor were related to the button/encoder ports on the Teensy, software (or assembly/connector) type things should be checked.

I'll report back what I find in the next day or so as free time and focus allows.

oldgearguy
Veteran Wiggler
Posts: 663
Joined: Thu May 26, 2016 6:42 am

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by oldgearguy » Tue Oct 20, 2020 3:50 pm

Follow-up --- probably a combination of being tired, impatience, and the module not being screwed in, but I must have let up on the encoder press too early or my hand twitched or something.
Today, under more controlled conditions (and patience) I held the encoder down until the menu appeared and stayed. The source code is scanning for other button and/or encoder presses/releases so it's probable just user error.

Now my only worry is the lifetime number of rotations of the right encoder. lol
Every voltage is off by around 0.25 volts, so lots of encoder twisting.

thanks to all who chimed in.

User avatar
autodafe
Wiggling with Experience
Posts: 406
Joined: Mon Oct 02, 2017 3:14 am
Location: Italy
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by autodafe » Sun Nov 08, 2020 4:30 am

I have a calibration issue with some uO&C I have built.
They all have a fixed voltage of about 1.15V at their outputs, when trying to calibrate the various -3V, -2V and so on.
I will double check all DACs but they look OK and I don't think I have bridged like all 4 of them...
Maybe a wrong component on them?
Any idea for this fixed voltage at each output?
Thansk

User avatar
pld
Super Deluxe Wiggler
Posts: 1077
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by pld » Sun Nov 08, 2020 5:14 am

autodafe wrote:
Sun Nov 08, 2020 4:30 am
I have a calibration issue with some uO&C I have built.
They all have a fixed voltage of about 1.15V at their outputs, when trying to calibrate the various -3V, -2V and so on.
I will double check all DACs but they look OK and I don't think I have bridged like all 4 of them...
Maybe a wrong component on them?
Any idea for this fixed voltage at each output?
Thansk
Check again for shorts and bridges around the output opamps as well, and/or post an image (although perhaps in the uOC build thread). 1.15V is somewhat interesting.

If it's a fixed offset, that smells like the v_bias if off. It's generated from the 2.5V reference from the DAC via 2x 47K (or 49K? Best check the schematic) -- it should be 1.25V.

User avatar
autodafe
Wiggling with Experience
Posts: 406
Joined: Mon Oct 02, 2017 3:14 am
Location: Italy
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by autodafe » Sun Nov 08, 2020 5:53 am

Checked DAC and Opamps with a microscope, no solder bridges..(I don't think I made the same bridge mistakes on 4 of them, it's not my first uO&C)
I'll check the resistors as you suggest

User avatar
autodafe
Wiggling with Experience
Posts: 406
Joined: Mon Oct 02, 2017 3:14 am
Location: Italy
Contact:

Re: ornaments+crimes extended / polymorphic 4x16bit CV generator

Post by autodafe » Sun Nov 08, 2020 6:42 am

pld wrote:
Sun Nov 08, 2020 5:14 am
Check again for shorts and bridges around the output opamps as well, and/or post an image (although perhaps in the uOC build thread).
I have posted pics and additional infos in the microO&C thread you suggested, thanks

Post Reply

Return to “Music Tech DIY”