ornaments+crimes extended / polymorphic 4x16bit CV generator

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

Moderators: lisa, luketeaford, Kent, Joe.

Post Reply
User avatar
oudplayer
Common Wiggler
Posts: 102
Joined: Sat Apr 16, 2016 8:12 am
Location: Brooklyn, NY
Contact:

Post by oudplayer » Mon Feb 11, 2019 12:38 am

forestcaver wrote:
oudplayer wrote:
Altitude909 wrote:when you plug the teensy in to the USB, does it identify it as a 3.2 and the firmware is 97% full? You should have it set to auto, press the button on the teensy and it should upload the firmware
Just looked into this; yes, when I plug the teensy into the USB I see 3.2 and 97% full (well, with a couple of the earlier firmwares I think one was 98%, but back on 1.3.4 now). Despite this, I have the same problem when I connect the module via the 10 pin header and usb is not connected—a flashing teensy. I'll try measuring voltages a bit later and report back with that, I'm hoping the problem's just a sloppy soldering joint.

I don't know if this makes much of a difference regarding anything, but the version of the teensy I have looked slightly different than the website pictures with one regard: it appeared that the trace had already been cut; well, at least, there was a black line on the board where the website shows a superimposed red line. I haven't plugged in the USB and 10 pin header at the same time so that shouldn't be a factor, right?
Hard to tell from your photos but your solder joints on the teensyheaders look quite suspect in the photo - especially theone below the usb port - a number look very cold - but it really is hard to see inthe photo....
Thanks guys for your help so far... made a bit of progress tonight, reflowed all the solder on the teensy, power headers and oled, joints look fine to me, and now at least I don't have a flashing teensy anymore, it's the proper dim solid color when powered. BUT: OLED not working at all, though, and when in my eurorack, all the o_C does is spit out a noisy pulse about once every 5 seconds (from all 4 outputs), regardless of whatever's thrown at the inputs. I've tried playing around with the app selection and such to see if a different program might kick in, but can't see what I'm doing! Nothing seems to change the behaviour. Need a new multimeter too, my fluke it appears has died. In any event, wondering if this once-every-5-seconds noise is something anyone else has experienced themselves, whilst I get down to replacing my multimeter and all that fun...

rthorntn
Learning to Wiggle
Posts: 17
Joined: Tue Nov 06, 2018 12:07 am
Location: Sydney, Australia

Testing potentially damaged DAC

Post by rthorntn » Thu Feb 21, 2019 12:24 am

I lifted one of the DAC pads by accidentally hitting it with a recoiling solder sucker, locked thread here.

mxmxmx kindly replied:
it kind of depends which trace we're talking about; it might be easy to find a neat work-around, or not. it sounds like it's one of the traces going to/from the DAC though, so probably it'll be tricky.

i'd say simply get a new board and start from scratch .... "ridiculously early stage" doesn't sound like much wasted effort. the DACs themselves are rather robust (re hot air, that is, i've reused some of them like 10 times for prototyping, testing boards etc), so that should be fine, unless you ripped off a pin or the like...

to answer your question though: to test the DAC, you'll mainly need to build the power section + the caps around the DAC. if the basically DAC works, it'll output 2.5V at the VREF pin (the reference is enabled by default, you can test this w/o a teensy attached); if it really works, it'll also output some voltage at outputs A-D, which you can tap at the 24k9 resistors; here's a simple test sketch, which you can adapt to suit your needs; as is, it'll update the outputs every 2 seconds in 0.25V steps (the comment in the code refers to the outputs after the op amps, hence 1V), until it reaches 2.5V, then starts from 0V again etc.
I just had a look with a stereo microscope and did some continuity checks, the DAC trace is lifted but the pin is soldered to it and there is continuity to the via.

I'm fairly sure all of the DAC pins are soldered OK.

The DAC would have taken some heat so that's my only concern but (as you say) it's pretty robust it might be OK.

If I was to tell you that I have access to variable bench power and an oscilloscope, does that make any difference to how I test the DAC?

UPDATE:

I just realised I don't have any machine pin headers for the teensy and so have had to order those.

Dag-nam-it, it never rains but it pours, I have just removed a pad on the 1.3" OLED header (used a Hakko FR301 to desolder the old square pins to put machined pins on there and missed a blob of solder under the header plastic, should of snipped the headers back), the RES (reset) pin it is, I assume its required and so I'm going to get some adhesive copper tape so I can try to salvage it (I do have spare OLEDs). Gotta laugh!

Thanks all!

Richard

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Thu Feb 21, 2019 8:36 am

Is there any way to download the firmware with "flip 180" already enabled? I'm having difficulties installing the arduino IDE, and honestly I wouldn't know what to do with it even if it would install. Trying to mount a couple uO_c's in my sequencer skiff.

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Fri Feb 22, 2019 4:52 pm

Or how about: where can I get a windows installer version of teensyduino 1.35? On their website I can only get the installer for v1.45.

User avatar
Timmy
Super Deluxe Wiggler
Posts: 1236
Joined: Fri May 06, 2016 7:08 pm
Location: Sydney, Australia

Post by Timmy » Fri Feb 22, 2019 5:37 pm

izmond wrote:Or how about: where can I get a windows installer version of teensyduino 1.35? On their website I can only get the installer for v1.45.
Apologies, I've just updated the documentation at https://ornament-and-cri.me/firmware/#method_b to point to Windows and Linux versions of the Teensyduino v1.35 installer, as well as the Mac version. Please forgive the Mac chauvinism.

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Fri Feb 22, 2019 9:48 pm

I should've known it was a Mac conspiracy... Thanks man, all good now.

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Sun Feb 24, 2019 12:22 am

Now, how about, (in the sequins app) can we get the sequence editor to show only the values that are included in the chosen mode or scale? Or, even better, can we get those relevant values translated into note names?

User avatar
lisa
Super Deluxe Wiggler
Posts: 4813
Joined: Fri Jun 02, 2017 7:00 am
Location: Sweden
Contact:

Post by lisa » Sun Feb 24, 2019 1:01 am

izmond: If it only shows the notes of the choosen scale then how should it handle a situation where the scale is changed after a sequence has been added?
New track! Drum synthesis and melodies. ✨


pld
Ultra Wiggler
Posts: 942
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany

Post by pld » Sun Feb 24, 2019 1:33 am

izmond wrote:Now, how about, (in the sequins app) can we get the sequence editor to show only the values that are included in the chosen mode or scale? Or, even better, can we get those relevant values translated into note names?
It'd probably be easier to write a note-based sequencer than try to retrofit that because it's just not designed that way. In addition to what lisa mentions, what note names should it display for user-defined/microtonal scales? To avoid all kinds of special-casing an abstract value is orders of magnitude less effort...

ricwadsworth
Learning to Wiggle
Posts: 13
Joined: Tue Feb 07, 2017 12:53 pm

DAC shorts

Post by ricwadsworth » Sun Feb 24, 2019 1:50 pm

Just building my first o & c and have shorts between DAC pins 5 and 6, 16 and 15, 16 and 14, 15 and 14.
The only other components on the board are the OP Amps and voltage regulator. Forgive electrical ignorance but just wanted to know if that was expected behaviour as all joints look fine and bridge free.
Thanks

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Sun Feb 24, 2019 3:00 pm

Points taken.

In other news: I successfully compiled and downloaded the firmware with "flip 180" enabled. The gate inputs and cv outputs all correctly remapped, but the cv inputs did not... So if I want to use channel 1 of meta-q I have to set the cv source as cv4. I tested it in several apps on two separate uO_c and both devices and all tested apps had the same behavior. Any suggestions on that?

In the faq it shows two additional lines of code that need to be changed, but in the compilation options section of the manual it says that "flip 180" being enabled would remap all the I/o.

pld
Ultra Wiggler
Posts: 942
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany

Post by pld » Sun Feb 24, 2019 3:56 pm

izmond wrote:Points taken.

In other news: I successfully compiled and downloaded the firmware with "flip 180" enabled. The gate inputs and cv outputs all correctly remapped, but the cv inputs did not... So if I want to use channel 1 of meta-q I have to set the cv source as cv4. I tested it in several apps on two separate uO_c and both devices and all tested apps had the same behavior. Any suggestions on that?

In the faq it shows two additional lines of code that need to be changed, but in the compilation options section of the manual it says that "flip 180" being enabled would remap all the I/o.
Maybe try and build an older revision, IIRC the new ADC/DMA code might have broken the FLIP_180 option.

pld
Ultra Wiggler
Posts: 942
Joined: Thu Mar 05, 2015 5:15 am
Location: Germany

Re: DAC shorts

Post by pld » Sun Feb 24, 2019 3:56 pm

ricwadsworth wrote:Just building my first o & c and have shorts between DAC pins 5 and 6, 16 and 15, 16 and 14, 15 and 14.
The only other components on the board are the OP Amps and voltage regulator. Forgive electrical ignorance but just wanted to know if that was expected behaviour as all joints look fine and bridge free.
Thanks
If in doubt, check the schematic. 5, 6, 14, 15, 16 all connect to GND.

ricwadsworth
Learning to Wiggle
Posts: 13
Joined: Tue Feb 07, 2017 12:53 pm

Re: DAC shorts

Post by ricwadsworth » Sun Feb 24, 2019 4:44 pm

pld wrote:
ricwadsworth wrote:Just building my first o & c and have shorts between DAC pins 5 and 6, 16 and 15, 16 and 14, 15 and 14.
The only other components on the board are the OP Amps and voltage regulator. Forgive electrical ignorance but just wanted to know if that was expected behaviour as all joints look fine and bridge free.
Thanks
If in doubt, check the schematic. 5, 6, 14, 15, 16 all connect to GND.
Thanks so much. For someone lacking in electrical knowledge it’s those sort of tips that help the learning process. :banana:

User avatar
mxmxmx
Super Deluxe Wiggler
Posts: 2243
Joined: Thu May 31, 2012 1:54 am
Location: berlin

Post by mxmxmx » Mon Feb 25, 2019 7:25 am

pld wrote:
izmond wrote:Points taken.

In other news: I successfully compiled and downloaded the firmware with "flip 180" enabled. The gate inputs and cv outputs all correctly remapped, but the cv inputs did not... So if I want to use channel 1 of meta-q I have to set the cv source as cv4. I tested it in several apps on two separate uO_c and both devices and all tested apps had the same behavior. Any suggestions on that?

In the faq it shows two additional lines of code that need to be changed, but in the compilation options section of the manual it says that "flip 180" being enabled would remap all the I/o.
Maybe try and build an older revision, IIRC the new ADC/DMA code might have broken the FLIP_180 option.
ups, yeah. i forgot about this when putting in the DMA stuff ... i'll try to fix this later/after work. in the meantime, please use the previous; or you can try with this (i'm not 100%, i can't test now): https://github.com/mxmxmx/O_C/tree/dev136

izmond
Learning to Wiggle
Posts: 48
Joined: Mon Jan 30, 2017 6:18 am
Location: OHIO

Post by izmond » Thu Feb 28, 2019 7:34 am

dev136 appears to be working. Thanks mxmxmx.

duno
Learning to Wiggle
Posts: 15
Joined: Sat Nov 17, 2018 3:20 pm

Post by duno » Fri Mar 15, 2019 8:50 am

edit-

Deleted. Wrong thread!

rthorntn
Learning to Wiggle
Posts: 17
Joined: Tue Nov 06, 2018 12:07 am
Location: Sydney, Australia

Built

Post by rthorntn » Thu Mar 21, 2019 5:04 am

Hi All,

To recap I lifted one of pads on the DAC footprint and I put a bit of heat into the DAC chip fixing it all up, continuity checks led me to believe the lifted pad is still working and the DAC leg is attached to it, I soldered the power section and I could see 2.5V VREF and on the four 24k9's I could measure the DAC stepping through voltages running the DAC output test (thanks again mxmxmx!).

I went ahead and finished the build and uploaded the firmware.

It all powers on, OLED fires up, encoders work.

I'm new to eurorack and so I thought I would look at the ADC debug page (with no signals patched) the rightmost values for all four inputs are fluctuating wildly, they all look to be in the 2600-2800 range but they are all moving about a lot, maybe +/- 50 so quickly that the display is lagging and so it's really hard to tell what it's reading, I read and understood it should be like maybe +/- 5 from 2730 so I'm a bit worried.

I do have access to a DMM and an oscilloscope and a replacement DAC chip (the lifted pad will probably make swapping the DAC dangerous though).

Thanks.

Richard

User avatar
mxmxmx
Super Deluxe Wiggler
Posts: 2243
Joined: Thu May 31, 2012 1:54 am
Location: berlin

Re: Built

Post by mxmxmx » Thu Mar 21, 2019 12:18 pm

rthorntn wrote:[...](with no signals patched) the rightmost values for all four inputs are fluctuating wildly, they all look to be in the 2600-2800 range but they are all moving about a lot, maybe +/- 50
that doesn't sound right ... they'll fluctuate even in a healthy module but not so much. it's hard to say what might be causing this, you probably want to start with reflowing the MCP600x op amp(s) (ie depending on whether that's the 8HP or 14HP version), and the passives around the it/them. also reflow the LM4040

rthorntn
Learning to Wiggle
Posts: 17
Joined: Tue Nov 06, 2018 12:07 am
Location: Sydney, Australia

Re: Built

Post by rthorntn » Thu Mar 21, 2019 11:08 pm

mxmxmx wrote:
rthorntn wrote:[...](with no signals patched) the rightmost values for all four inputs are fluctuating wildly, they all look to be in the 2600-2800 range but they are all moving about a lot, maybe +/- 50
that doesn't sound right ... they'll fluctuate even in a healthy module but not so much. it's hard to say what might be causing this, you probably want to start with reflowing the MCP600x op amp(s) (ie depending on whether that's the 8HP or 14HP version), and the passives around the it/them. also reflow the LM4040
Thanks mxmxmx, I fluxed around the (only) MCP6004 and the LM4040 including the nearby passives and applied hot air until I could see the solder soften, still the same issue. A video:

ADC video

UPDATE: I'm wondering if the crazy fluctuations in the middle column are relevant, not sure if that column should fluctuate at all?

User avatar
mxmxmx
Super Deluxe Wiggler
Posts: 2243
Joined: Thu May 31, 2012 1:54 am
Location: berlin

Re: Built

Post by mxmxmx » Wed Mar 27, 2019 6:20 am

rthorntn wrote:
mxmxmx wrote:
rthorntn wrote:[...](with no signals patched) the rightmost values for all four inputs are fluctuating wildly, they all look to be in the 2600-2800 range but they are all moving about a lot, maybe +/- 50
that doesn't sound right ... they'll fluctuate even in a healthy module but not so much. it's hard to say what might be causing this, you probably want to start with reflowing the MCP600x op amp(s) (ie depending on whether that's the 8HP or 14HP version), and the passives around the it/them. also reflow the LM4040
Thanks mxmxmx, I fluxed around the (only) MCP6004 and the LM4040 including the nearby passives and applied hot air until I could see the solder soften, still the same issue. A video:

ADC video

UPDATE: I'm wondering if the crazy fluctuations in the middle column are relevant, not sure if that column should fluctuate at all?
yeah, it looks rather noisy. also a bit hard to discern this way. either way, it should basically work. have you tried to calibrate the module? (it doesn't look like it.)

rthorntn
Learning to Wiggle
Posts: 17
Joined: Tue Nov 06, 2018 12:07 am
Location: Sydney, Australia

Re: Built

Post by rthorntn » Wed Mar 27, 2019 10:16 pm

mxmxmx wrote:yeah, it looks rather noisy. also a bit hard to discern this way. either way, it should basically work. have you tried to calibrate the module? (it doesn't look like it.)
Thanks, calibrating the DAC channels goes fine but the CV inputs CV1, CV2 and CV3 have the crazy 2 digit jitter and they never adjust to anything that approximates 0, CV4 configures OK.

UPDATE_1: looks like I have found the issue... on the front of the board the silkscreen is cutoff, so for three of the resistors the silkscreen looked like 100K, I need to take them off and put 100R resistors on there.

UPDATE_2: Changed the resistors, MUCH happier.

User avatar
dragulasbruder
Common Wiggler
Posts: 172
Joined: Sun Oct 23, 2011 9:57 pm
Location: Chicago

Post by dragulasbruder » Thu Apr 25, 2019 8:31 am

a feature request from the voltage gods:

When using the Quantermain with the CV source set to LFSR, there is no "LFSR rng CV" option when the quantizer is set to "Off". Is it possible to make one in a future update? Even if I'm not quantizing the output, I'm just as likely to want to control the spread of possible values than the probability of a bit flip--I find myself using Quantermain as a random voltage source more than anything.
Now offering sliding-scale SYNTHESIS and AUDIO PRODUCTION LESSONS during the pandemic--please PM for details!

WTB Wish List: https://www.modulargrid.net/e/racks/view/1224624

MY MUSIC: https://www.soundcloud.com/wish-fulfillment

User avatar
mxmxmx
Super Deluxe Wiggler
Posts: 2243
Joined: Thu May 31, 2012 1:54 am
Location: berlin

Post by mxmxmx » Thu Apr 25, 2019 9:09 am

dragulasbruder wrote:a feature request from the voltage gods:

When using the Quantermain with the CV source set to LFSR, there is no "LFSR rng CV" option when the quantizer is set to "Off". Is it possible to make one in a future update?
hi, not 100% but i think it accidentally just got deleted somewhere along the way ...(here). i've just re-inserted a/the missing line in this branch. works? (not at home right now, so can't try myself...)

User avatar
J_Behan
Common Wiggler
Posts: 53
Joined: Wed Jan 03, 2018 7:57 am
Location: Ireland
Contact:

Post by J_Behan » Tue Apr 30, 2019 7:19 am

Trying to calibrate at the mo.
On channel A, and I've got up to 5v done. But when I try to dial in 6v, the figure in the top right corner has maxed out, and I can't get higher than 4.6v on my meter.

Is this likely down to a build mistake?

Post Reply

Return to “Music Tech DIY”