0
\$\begingroup\$

If I put 4 of 1uFx100V cercaps in series I will virtually have one .25uFx400V capacitor.

As I know I should use balancing resistors for even distributinon of the voltage across the capacitors. However if I skip them in the simulation I did not find any diviance in the voltage - no matter which ESR or parallel resistance deviance I put into model.

So my conclusion is that in case of AC (no DC) there is no need for balancing resistors. Is that correct?

\$\endgroup\$
4
  • 3
    \$\begingroup\$ Did you try your simulation with slightly varying capacitance values? Tolerance means that no 2 caps will have exactly the same value ... \$\endgroup\$
    – brhans
    Commented Sep 4, 2018 at 17:03
  • \$\begingroup\$ Thanks, @brhans. That's really made the difference. However balancing resistors did not help in this case either :( I tried 100k resistors (less resistance made the the power dissipation too big for my application) \$\endgroup\$
    – Roman Matveev
    Commented Sep 4, 2018 at 17:12
  • 2
    \$\begingroup\$ Parallel resistors will only help DC voltage distribution due to mismatch is leakage. Depending on application you are more likley to encounter problems due to +-20 % difference in capacitance. \$\endgroup\$
    – winny
    Commented Sep 4, 2018 at 17:58
  • \$\begingroup\$ In case you are, you should not use 4 ceramic caps in series to run on 240Vac line. That requires an X1 rated plastic cap. These will be bulky since the Dk is much lower but AC line must withstand HV spikes. \$\endgroup\$
    – Tony Stewart EE75
    Commented Sep 4, 2018 at 20:52

0

Reset to default

Browse other questions tagged or ask your own question.