strange behaving ARM cortex analog input

[e.oelberg]

Hi all !

I have a synthesizer here on my bench where one of the 10 kb pots behaves strange. Actually it is a simple voltage divider circuit 3.3 volt in -> tapper to arm analog input. while all other pots work fine, this one goes within the first 1/4 turn from ccw to cw from zero volt to 3.2 volt. I'm pretty much an electronic novice , but what could be  the reasons ?

The arm cortex input works.

n

 

[Andy Peters]

Hi all !

I have a synthesizer here on my bench where one of the 10 kb pots behaves strange. Actually it is a simple voltage divider circuit 3.3 volt in -> tapper to arm analog input. while all other pots work fine, this one goes within the first 1/4 turn from ccw to cw from zero volt to 3.2 volt. I'm pretty much an electronic novice , but what could be  the reasons ?

The arm cortex input works.

The pot's broken. Replace it.

 

[e.oelberg]

thats what I did, but I found the problem: broken solder joint at the input of the arm cpu. reflow did the trick !

 

[JohnRoberts]

Yup solder joints can do that.

Broken pots are a common field failure for some power amps (I suspect the packaging could be better for those, and users probably kick/step on the volume pots by accident while amps are sitting on the floor.)

I have even seen broken pots in new builds but the worst offender for that was because one pot manufacturer lost control of their process and were over-curing the substrates to bring the final resistance into range. The green resistance of screened ink pot elements changes after being cooked so they target the green resistance a little high so it cures in range.. This pot manufacturer over cooked their pots trying to make the final resistance meet spec and instead made the pot substrates brittle. These pots would break just from hand assembly into PCBs by factory workers who were not mishandling them.

Another failure mode with mature pots that have been in service for a long time is wiper wear or contamination with dirt.

The step discontinuity like you described sounds like a broken pot with a cracked substrate. Low output until it jumps high. So I would have seconded Andy's guess if you didn't already fix it.

JR

 

[mhelin]

The arm cortex input works.

n

Some ARM Cortex based micros are known to have less than perfect ADC's. However, the SMT32F3  mixed-signal series chips should be fine like used over there (for anyone interested in Eurorack synths): 

http://ebrombaugh.studionebula.com/synth/f373_mod/index.html