My LM350 regulator with smoothing caps (C1:0.33uF, C2:1uF polarized, R1:150 R2:10ohms+trimpot) does not work. It works without capacitors. I am an idiot. What have I done wrong? I tested all the connections, and they match up fine with the diagram. On Vout, there is briefly a nice voltage, then it drops. When I retest with my DMM a few moments later, the voltage is up again, then it drops.




What is wrong? Hit my stupid skull with your cluebats!

Edit: Had the same problem with a LM317 voltage regulator. I 'fixed' it by removing the caps. But I really want to know what I'm doing wrong here. The LM350 is confirmed working - without those caps.

Edit2: This question has been answered by @sstobbe. I had a bad solderjoint in an inconspicuous location. Thanks for your time everyone.

Working regulator

  • 1
    \$\begingroup\$ Does the regulator get hot with no load and the capacitor? \$\endgroup\$ – W5VO Jun 9 '17 at 13:08
  • \$\begingroup\$ No, it is quite cool. \$\endgroup\$ – user2497 Jun 9 '17 at 13:09
  • \$\begingroup\$ There is most certainly no short, either. I feed it 12V@20A. There would be a smoke signal. \$\endgroup\$ – user2497 Jun 9 '17 at 13:11
  • \$\begingroup\$ The way you have C2 in there looks odd - it is hard tp tell, but do you have the negative side connected to ground? Try removing only C2. If it then works, turn C2 around and reconnect it then see if it still works. \$\endgroup\$ – JRE Jun 9 '17 at 13:21
  • 1
    \$\begingroup\$ R1 & R2 are constantly loading the output so adding a DMM and then the voltage drops is suspicious, its likely a bad solder joint somewhere or its oscillating \$\endgroup\$ – sstobbe Jun 9 '17 at 15:07

Resistors R1 & R2 are constantly loading the output. So, having the output voltage sage due to probing the output with a DMM is suspicious.

Its likely a bad solder joint somewhere. When you apply pressure with the DMM probes, you break a connection.

It could also be that the output is oscillating and the impedance of you holding the probes plus meter is sufficient to detune whatever is oscillating.


You only need a load. The LM devices are "series pass" and can only increase the voltage (by supplying current) but can't decrease it by sinking it; a load has to pull the cap down if it is too high.

  • \$\begingroup\$ I tried a voltmeter that pulls ~30mA. Not enough? \$\endgroup\$ – user2497 Jun 9 '17 at 14:27
  • \$\begingroup\$ Is this the explanation for the shifting voltage potential I read with my DMM? \$\endgroup\$ – user2497 Jun 9 '17 at 14:27
  • \$\begingroup\$ Should be enough. What output voltage are you aiming for? \$\endgroup\$ – John Birckhead Jun 9 '17 at 14:29
  • \$\begingroup\$ 9 volts. I saw it rise and fall with my DMM, testing it every few seconds. \$\endgroup\$ – user2497 Jun 9 '17 at 14:31
  • \$\begingroup\$ What is the voltage at the adj pin? \$\endgroup\$ – John Birckhead Jun 9 '17 at 22:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.