importance of capacitor choice (an observation)

Go To Last Post
3 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

The tacho in my car died recently - when I say died I mean it was wildly swinging between 5000 and 9000rpm. After pulling it apart I saw a suspect greencap and replaced it, the value was 15nF. The only caps I had that were suitable were 10nF ceramics, so I put two in series (to get 5nF) and then a third in parallel with the series'd ones to make 15nF.

The only problem was when the car was cool, the circuit would over-read by about 500RPM. When things heated up (parked in the sun all day) I was getting about 50% of the correct RPM value, or even less! It got to the point one day where my car was idling at 0RPM and redline was about 3000RPM (rather than 7000). Good news is the needle wasn't wildly swinging, and once I replaced my bodge job with a proper 15nF greencap I got an accurate tacho back again.

I am aware ceramics are non-linear with regard to temperature and the applied voltage but I am very surprised as to the degree of problems with what should be a fairly simple circuit. 10-15% error maybe but this was wild. I couldn't find a datasheet on my tacho IC but I assume it's a frequency to voltage converter and the cap is used for charge storage (not really a ceramic's forte!).

So use the right cap type, kids :P

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

V to F circuits usually call for polystyrene or polyprop. caps which have low dielectric absorb.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Alwelch is right regarding matching capacitor type to application.

Also watch for the different classes of capacitors in ceramics. I've been burnt by this in the past. I had a smps that would jump from 6.9V regulated to 7.2V regulated after the capacitor heated up.

This is a good read.

http://my.execpc.com/~endlr/ceramic.html