Author |
Message |
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Fri Jun 28, 2019 3:00 pm Post subject:
Klee Troubleshooting Subject description: Clock runs through sequence only once |
|
|
Hey everyone,
I have my Klee built, but have a problem early on in the test procedure. I press the load button and enable the clock and it will run through the sequence once. Also, when I turn the clock input off the light doesn't stay on the first LED. I also need to turn the clock off to get the pattern to load again. I tried seeing if I could use the manual step to step through the sequence a second time, but I cannot. I can use manual step to step through the sequence once only.
Also, manual load doesn't always work every time.
thanks so much for any help |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Sat Jun 29, 2019 6:14 pm Post subject:
|
|
|
It's not looping back around. Does this happen both in 1x16 and 2x8 modes? |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Sat Jun 29, 2019 6:19 pm Post subject:
|
|
|
If I recall correctly the most likely cause is a miswired/bad cable and the second most likely is something in the chips /signals either before or after.
You can verify some of this depending on which options you've chosen to build by setting it to random, forcing the random input to be on all the time and seeing if you can clock in 1's that way. If not then it's a problem on the input side of the loop. Otherwise it's likely on the output.
Also switching 1x16 or 2x8 helps -- usually you'll find either the A or the B side will loop fine, which also points you to whether the problem is on the output side or the input side.
Which Klee are you building? Is this one you've wired yourself, or the Euro, or the recent Synthcube MOTM panel PCB one? |
|
Back to top
|
|
|
AlanP
Joined: Mar 11, 2014 Posts: 746 Location: New Zealand
Audio files: 41
|
Posted: Sat Jun 29, 2019 11:29 pm Post subject:
|
|
|
Is this old-style wires everywhere, NV's Eurorack adaptation using cables, or NV's new Eurorack no-cables? |
|
Back to top
|
|
|
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Mon Jul 01, 2019 10:09 am Post subject:
|
|
|
Thanks so much for your replies!
This is an old school style wires everywhere MOTM panel. This is one of few panels I didn't make and screen print myself. I think I got it from Bridechamber way back when.
By testing the random input function I discovered that the random/pattern switch is backwards. Everything seems to be working fine. That should've been the first thing I tried, but I immediately jumped to thinking it was something more complex.
Right now I have a new problem with the gate bus. Everything more or less works as it should but some of the switches are always setting back to Bus 2 so I'm guessing I have a bad connection somewhere. It shouldn't be hard to find. The Bus load isn't working. |
|
Back to top
|
|
|
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Mon Jul 01, 2019 2:19 pm Post subject:
|
|
|
So I've identified that Gate Bus switches 5, 10, and 13 are the culprits.
The way I initially did this was leave all bus switches at bus 2 and switch through one at a time while clocking the input and seeing which individually didn't light up bus 1. Some had broken because of the cheap UTP wire I used, inexplicably only on those switches (the rest uses a better quality UTP 24awg stranded wire).
After fixing two switches I've narrowed it down to these 3.
I've checked continuity back to the board by throwing each switch individually and then using a continuity tester and testing the connection from the strap wire (of say bus 1) to the cathode of the corresponding diode of the offending switch. I get continuity. I can't see why there would be this specific problem only on one control pin of the 4066's, so I'm kind of mystified. I also thought maybe I had bent legs on the 4066's so I tested continuity from the anodes of the diodes to the control pins of the 4066's from the top and got continuity. |
|
Back to top
|
|
|
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Mon Jul 01, 2019 2:54 pm Post subject:
|
|
|
I always crimp and solder wires in molex connectors but didn't for 3 in the wire harnesses that connect the analog and digital boards: the three offending gate bus switches. Now working. |
|
Back to top
|
|
|
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Mon Jul 01, 2019 2:57 pm Post subject:
|
|
|
In the bring up procedure, with the part testing the Bus Load Switch: should pattern LED's 1-4 light sequentially over and over?
When I throw the Bus Load switch it cycles through 1-4 once and then goes back to cycling through the whole sequence. |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Mon Jul 01, 2019 4:01 pm Post subject:
|
|
|
That sounds pretty weird to me |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Mon Jul 01, 2019 4:02 pm Post subject:
|
|
|
Although.... Bus load is on gb 1, do you have gb1 activating? |
|
Back to top
|
|
|
jstarx
Joined: Jun 28, 2019 Posts: 6 Location: Ecuador
|
Posted: Mon Jul 01, 2019 5:41 pm Post subject:
|
|
|
If I understand correctly, in order to perform the test I need to to route a gate bus switch to gate bus 1. Gate Bus 1 is sending a gate on that step.
Also, I've been watching more carefully and 1-4 continue to cycle, sometimes every other cycle, sometimes not at all. This led me to think it might be an intermittent connection, but I have good continuity between the switch and GB1, and the other pole has a good connection to Digital Board J3 position 7, so I have no idea. |
|
Back to top
|
|
|
|