Author |
Message |
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Wed Jan 04, 2017 10:29 am Post subject:
Droopy MFOS single buss keyboard |
|
|
Hi everyone,
I built a MFOS single buss keyboard out of a cheap midi controller using reed switches and magnets and I need your help for troubleshooting.
It overall works great and the tuning is good at first. However the voltage drop is really fast, about 50-100mV in less than 10 seconds, which is way higher than what Ray says in the webpage. Component values are all correct, the only change I made is in R5 where I used 470 Ohms instead of 475 (didn't have this value). All resistor are metal film 1%, used film caps in place of most ceramics and used a polystyrene cap for C19. The only thing I can think of is C19 I used a cap with radial leads, so the leads are quite long to fit the PCB, could this possibly be the cause of the fast voltage drop?
Thanks
Stefano |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jan 05, 2017 11:47 am Post subject:
|
|
|
How clean is your board? Flux residue can be conductive. |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 11:59 am Post subject:
|
|
|
It's clean. I tried to change the cap but only had .015uF ones, the voltage drop is even worse with this. I guess the problem has to be c19 though, right? |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jan 05, 2017 12:05 pm Post subject:
|
|
|
What type of cap are you using?
It could also be one or the other of the FETs, Q2 & Q3. None of these components (including film caps) like to get very hot, so they can be damaged in soldering. That would be my next guess. |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 12:27 pm Post subject:
|
|
|
Ok, thanks. Any way to figure it out wuth a scope? Or do I just have to try and replace them? |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jan 05, 2017 12:57 pm Post subject:
|
|
|
Check the resting voltage at the gate of the second fet. It should be significantly negative. If it's not, check the voltage at the gate of the first fet. This should also be very negative, approaching the rail as close as an op amp can drive it. If *that's* not, then pull up the source and gate of that first op amp and see if it's still not negative enough.
"enough" will be defined by the datasheet for the FET, as the cutoff or pinch off voltage.
If it's still not negative enough, then the problem is before the fets. Otherwise it's one of the two. Depending on how many spares you might have, I'd probably change them one at a time. |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 1:03 pm Post subject:
|
|
|
Thank you very much. The gates of both FETs are -10.73V at rest. Think the problem at this point really is C19. |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jan 05, 2017 1:06 pm Post subject:
|
|
|
Yes, I would agree, C19, or something with U3 is draining it off. Cap is more likely.
What type did you use? |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 1:14 pm Post subject:
|
|
|
I'm using polystyrene caps, don't know the brand.
I'm a beginner with soldering so I think heat damage is likely, although I then tried two different ones being very careful and same result (these last two were .015uF though, don't know if this can explain the drain). |
|
Back to top
|
|
|
elmegil
Joined: Mar 20, 2012 Posts: 2177 Location: Chicago
Audio files: 16
|
Posted: Thu Jan 05, 2017 1:17 pm Post subject:
|
|
|
No, the change from .01 to .015 would not explain the drain. It might make it more sluggish to charge, but I doubt it would have a huge impact.
Pull U3 (assuming sockets) and measure the voltage at the top of the cap, see if you still have droop. |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 1:39 pm Post subject:
|
|
|
Ok, I tried that. I see the right voltage just for less then a second and then quickly to ground level or so. I don't see and solder bridges on the pcb, everything looks good. Could Q3 be shorted? |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Thu Jan 05, 2017 1:53 pm Post subject:
|
|
|
unless you have a high end meter with very high input impedance, it's going to drain the cap when you measure the voltage on it.
Did you use the correct op amp for U3? It needs to be a high impedance JFET input type.
It does sound like something is leaking, either Q3, C19 or U3. I've seen another member have a problem with a bad FET leaking, have you replaced Q3 yet?
Are you sure you're using polystyrene and not polyester? _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 2:05 pm Post subject:
|
|
|
I just tried replacing Q3 and nothing changed, although it was one that had already previously used, will try with a new one, but don't have any right now. I'm using LF444 for U3, which I suppose it's the correct one. |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 2:08 pm Post subject:
|
|
|
Yes I'm using styroflex caps. |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Thu Jan 05, 2017 2:31 pm Post subject:
|
|
|
What's the voltage at U2 pin 7 when all the keys are released? Do you get the expected -0.8V? _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 2:43 pm Post subject:
|
|
|
I get -0.57 V with all keys released on U2 pin 7 (same voltage as the bus). |
|
Back to top
|
|
|
gdavis
Joined: Feb 27, 2013 Posts: 359 Location: San Diego
Audio files: 1
|
Posted: Thu Jan 05, 2017 2:52 pm Post subject:
|
|
|
That should be fine. At this point I think I would try removing Q3 and briefly shorting the drain and source connections while a key is pressed, then see if the CV output still drops as fast. That should indicate whether there's a problem with Q3. _________________ My synth build blog: http://gndsynth.blogspot.com/ |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Thu Jan 05, 2017 3:46 pm Post subject:
|
|
|
Sounds like a good test to do. Will try tomorrow. Thanks guys! |
|
Back to top
|
|
|
stefano_cipo
Joined: Jan 04, 2017 Posts: 11 Location: italy
|
Posted: Fri Jan 06, 2017 4:11 am Post subject:
|
|
|
So I did the test and there wasn't any voltage drop. I didn't have any spare 2n5457 so I tried a mpf102 and it works great. I have only about 20mV drop after 4 minutes, which is better than expected. Thanks guys for assisting me in troubleshooting this, was almost sure the culprit was c19. |
|
Back to top
|
|
|
|