MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

TSNM flashed right, booted right once, now doesn't...
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Author TSNM flashed right, booted right once, now doesn't...
jschussler
I got the TSNM build done and flashed, and the first boot it seemed happy: showed the "TSNM 3.0" firmware text, then finished showing me just "C" which I assume means it's ready for calibration. Tap the pads, they seem to work, the LEDs light, etc.

Reboot. Now I get what you see in the vid: a single line, no other response from the display. I've reflashed the Teensy, no change. Doesn't look to me like a build issue, since it booted right the first time, but....

OIP
yeah that's weird never seen that display before.

the 'C' is just the actual note C which i think is default. as far as i know most of the problems with these builds come down to:

diode polarity

bad capacitive touch board

teensy pin issues

general soldering issues

so those are what i would check. also worth checking the actual output in case it's an issue with the LED panel driver circuitry or pins.
guest
is it possible the bootloader got overwritten or the boot vector reset? i imagine youre using the standard teensy ui, which i dont think allows that to happen. but what you describe sounds something like that. the first time the code goes in the bootloader sets the starting point, but every subsequent reset the boot vector is wrong, so it goes to the wrong start point.

if you can take the teensy off, try programming it with the blinkn code, and see if it still uploads and runs code normally outside of your circuit.
jschussler
Unfortunately I cut the trace that allows it to be powered by USB, so now it has to be on the panel to work....

Wouldn't reflashing it fix that in any case?
guest
well, its possible it got modified so it cant accpet new code. maybe take it off, and jumper power to it and try loading code. it would be good to eliminate the teensy as the error source. if it can accept different code, then its either the code or the hardware.
jimfowler
Are you getting any actual voltage output?
jschussler
I'm getting pressure, gate, and trigger outs. Also, I can press and hold the encoder and the display changes (still borked, but different borked).

Also, I turned it on this morning and it showed the boot text TSNM 3.0 initially, then C for a few seconds, then went wonky.

So I think the Teensy is fine. Seems like there's some other connection that's either intermittent or wrong.
jschussler
Reflowed some more, am now getting the entire display to light up, but still borked. And no voltage out on 1v/o. Just on the others (gate, trig, pressure).

Here are some pics.


[/img]
djthopa
Ic3...are you sure all pins are making contact with the pads? Eg correctly soldered?

You can check with a small dentist tool or tiny wood pick, if the ic leg moves its not soldered properly.

Hope you get it fixed.
jschussler
I wiped off a bunch of solder to make sure there weren't any bridges. I'll give them a touch more.
jschussler
Does anybody have any known voltages for different sections of the board? I'm wondering if I can narrow this down by probing different components, but since I don't know what voltages I'm looking for (keep in mind I'm not an EE, and have only limited experience with assembling modules).

I've reflowed again, and just can't figure it out. I did replace C16 with a 1uf cap that was 0603 rather than 1204 (or whatever that larger size code is). Didn't make a difference.

The behavior is the same regardless of whether the front panel is attached, so it's not the interface there. I've swapped both the Teensy and the MPR121, so it's not those.

I can't decide whether to use the cry emoticon or the seriously, i just don't get it . Or maybe just Dead Banana
jimfowler
The schematic is available on Doboz's site. Have a look and I'd start at the DAC and work backwards since that's where you Vout comes from.

- Jim
OIP
check the diodes?

it's a pain but it should be possible to track down by going methodically through the schematic as it's low enough part count.

hope you can fix it! it's a great module. if i get a chance i can try and test some voltages, but i don't think there's anything too exciting going on.
jschussler
The problem I'm having is that I have voltages all over the place (including the DAC). Nothing is obviously not passing current. I just don't have any idea what the "right" values should be....
jschussler
Doboz, you're killing me.

As it turns out, the PCB is wrong. It shows diodes D1 and D2 backwards from what's shown in the schematic. When you said "check the diodes" it didn't occur to me you mean "check that they're backwards from what the PCB says."

I've now reversed those, and get this:

- Now I get no display whatsover.

My brain hurts. How is it possible, that with the diodes in their previous wrong position I was getting it to boot and show the display (albeit borked), and now I'm not? Makes no sense.

I'm seeing 3V or so on IC1 and IC2, so my theory is they're alive. Not entirely sure, though, don't now what the values should be on all the pins.

IC3 and IC4 seem fine, but I still don't get a v/oct output. Pressure, gate, and trigger all work fine. So the board in general is working, and the Teensy is alive. (Yes, I've reflashed it).

Anybody have any idea what voltage ranges I should be seeing on ICs 1-4?
guest
D1 and D2 are mislabelled (at least from comparing the schematics to the board image on the build guide), but the silkscreens is correct. they should be orientated as shown on the silkscreen. right now there is only 3.3V getting to the board, and not the +/-15V rails, because those diodes are backwards.
jschussler
Ok. Diodes back to original PCB orientation. I see 12V around the board on various spots, so that seems good. No display, no v/oct out. Makes me wonder if there's a common component that supplies both of those that is dead.

Sadly I don't know enough about electronics to understand from the schematic if that's true....
guest
that first column is controlled seperately from the rest, which are controlled via SPI and the 595s, so im guessing the problem is either with one of those two things. you might be able to put your meter on the SPI lines and see some voltage flickering if data is going across.
jschussler
Um...what's an SPI?

Sorry, I really am ground level here.
guest
the DAC and display are both on the SPI, although so is the whole teensy.
guest
SPI is a serial communication protocol that most microcontrollers support. this board uses SPI and another protocol, I2C, which goes to the touch sensor board. that schematic is really tough to follow. SPI uses 3 pins to talk to a device, and a 4th if the device wants to talk back. none of these talk back, so its just 3 - clock (SCLK), data (MOSI), and select (SS). on IC2 these are pins11,14,12. check to see if those are just stuck, or if they blip every so often. its possible they are working but you cant see it with your meter, as they transfer data really fast.
guest
if pressure, gate, and trigger are still working, then its definitely the SPI. focus on the SCLK and MOSI lines, as they are common to the display and DAC. turn off the power, and measure the resistance to ground and 3.3V on those lines, they might be shorted. might be good to check resistance to nieghboring pins as well (looks like they might be on IC3 - two middle pins on the left in your picture are the MOSI and SCLK lines).
OIP
someone had the same issue here: https://llllllll.co/t/diy-tsnm-touch-sensing-note-memory-by-doboz/1184 7/168

but i can't find record of any particular resolution. that thread reminded me about the extra pin on the teensy which needs soldering, it's under one of the ends, so worth checking. i think it's pin 14 of the teensy. here's it from the other side:



sorry about the diode confusion - i did mean check the orientation but just the basic 'are they the right way around' not that there was any known consistency issue. remember reading a few people had issues with this.

also agree with all guest said about the signal path for SCLK in particular
jschussler
Getting closer. You are definitely right about the DAC being part of the issue. I've replaced it entirely, and am now here:



So display mostly working except for columns 1 and 4 of the left matrix.

As far as the v/oct goes, now I get a noisy output that, when I tap the pads, does change the pitch of the oscillator, but you can only barely hear the change. It's mostly noise.
guest
the first line of the display is on its own piin from the microcontroller, so follow that back and check for signal on it. whats odd, is that was the only line working before. column 4 goes back to IC1 through R7, so check those parts out. its mostly like a short/open somewhere.

as for the DAC, you can measure the voltages directly on the DAC, and see what it does with different button presses. the output goes to R27, and it would also be good to check the Vref out, which goes to R28. you can measure on these resistors rather than at the IC, as the IC pins are really close to one another, and you risk shorting them with the probes.
MUFF WIGGLER Forum Index -> Music Tech DIY Goto page 1, 2  Next [all]
Page 1 of 2
Powered by phpBB © phpBB Group