Where i work we do not use tantalum caps because there down sides are pretty big in terms of safety.
when you over work a tant, they get very hot.
It should. The other thing to check is ripple: if you're getting enough to make a ceramic sing, the ripple current might be a problem for Ta. Check the ripple voltage and frequency across the cap and use the resistance and capacitance ratings to deduce a current.Yeah, the ripple voltage is pretty high, about 0.5vpp at 4.7uF and 14mA average draw. Hence my other thought, which was a big electrolytic and a medium sized cermic/tant.
If you have that much ripple, you need more capacitance. more of a reason to have a current trace.a high resolution voltage trace would also give the volts per second to deduce the current, knowing the ESR.
Steve On Sep 19, 2007, at 10:30 AM, DJ Delorie wrote:
The PCB for my alarm clock buzzes.Having poked and prodded, and getting a piezo unit that did nothing, I finally figured it out (the trick was to use a nail, not a toothpick, to poke and prod - I was using that as a better sound conductor for the piezo). There are not one but TWO buzzing caps. If I replace both of them with electrolytics, the buzzing goes away. If I replace them both with two new ones of the same type/size, it still buzzes. The caps are 4.7uF 16v. They're decoupling/bypass for a +12v rail (well, one is, the other is VcomH, which seems to be 7v). They're the biggest 0805's with that voltage rating. I can fit a 1206 in that spot. If I replace those with 10uF 25v, do you think they'll not buzz? Another alternative is 22uF 16v. I don't know if the buzz is due to insufficient capacitance, or damage due to being too close to the voltage rating. Alternatively, I can stack two 10uFs on top of each other to make 20uF 25v, but that's ugly. 1210's won't fit. Note that adding a 47uF 35v electrolytic across the 4.7u reduces the noise greatly. In fact, a 4.7u electrolytic helps much more than a second 4.7u ceramic. If I replace just the +12v cap with a 22u elec, the +7 cap still buzzes. I suspect the elec alone can't react fast enough to get rid of the ripple. _______________________________________________ geda-user mailing list geda-user@xxxxxxxxxxxxxx http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Attachment:
smime.p7s
Description: S/MIME cryptographic signature
_______________________________________________ geda-user mailing list geda-user@xxxxxxxxxxxxxx http://www.seul.org/cgi-bin/mailman/listinfo/geda-user