4
\$\begingroup\$

There is a 332 1kv ceramic disk capacitor in a battery charger that I use on my desk. The capacitor emits a very high frequency and headache-inducing "singing" noise. I understand that this is normal operation for a ceramic capacitor if the circuit is not designed properly. How can I rid of this noise? Is there an equivalent capacitor that I can solder in that will not vibrate at an audible frequency?

I have read it is common to put 2 capacitors in parallel to reduce noise, if this is an option should i be replacing it with 2 equivalent capacitors or with 2 capacitors of half the rated capacitance value?

\$\endgroup\$
  • 1
    \$\begingroup\$ How do you know it's coming from the capacitor and not, say, the transformer or choke? Also, where exactly is this in the circuit? The 1kV rating implies there may be safety implications. \$\endgroup\$ – Spehro Pefhany Nov 13 '14 at 20:56
  • \$\begingroup\$ A hammer comes to mind :^) As Spehro said 1kV may be dangerous. Sometimes noise will be louder due to mechanical coupling. Maybe a piece of foam under the charger? \$\endgroup\$ – George Herold Nov 13 '14 at 21:11
  • \$\begingroup\$ Unless you have a schematic and understand the function of the cap and any safety implications I would hesitate to make any changes. It's pretty rare for a ceramic disk cap to make that much annoying noise, usually it's the SMT caps that couple mechanically to the PCB causing it to act as a speaker. The leads on the disks provide some isolation. So as Spehro said, it may be a transformer or inductor singing. \$\endgroup\$ – John D Nov 13 '14 at 21:54
  • 4
    \$\begingroup\$ Ceramic capacitors exhibit piezoelectric effect; that's why they can create audible noise. When I had a similar problem, I have replaced a ceramic capacitor with a polyester film capacitor. However, this approach requires spare room, because a film capacitor will have larger mechanical dimensions, compared to a ceramic capacitor with the same capacitance and voltage rating. \$\endgroup\$ – Nick Alexeev Nov 13 '14 at 23:14
  • \$\begingroup\$ @NickAlexeev, I like the idea, but I don't know about the 1kV how big is a 3.3nF film? (and is it AC or DC?)(~$1 on DK) \$\endgroup\$ – George Herold Nov 13 '14 at 23:55
2
\$\begingroup\$

I suspect that the noise is not actually coming from that capacitor. Yes, some ceramic capacitors can exhibit enough of a piezoelectric effect that they can make audible noise under the right circumstances. However, this pretty much has to be helped along by being mechanically coupled to something that acts like a sounding board, usually the PCB. You said "disk" capacitor, which implies thru hole. The size fluctuations of such a disk capacitor would couple to the PCB quite poorly, and the disk itself is quite small and can't radiate a lot of sound.

Much more likely, there is a transformer in this battery charger, and that is what is vibrating. Each little section of wire experiences a sideways force due to the current thru it and the magnetic field around it. It's quite common for transformers to whine for that reason. Magnetorestriction can also occur, but usually plain old magnetic forces outweight that considerably.

This is apparently a line-connected high-voltage circuit. Anyone that has to ask here shouldn't be messing with such things. Even if you don't electrocute yourself, you could be making the device unsafe so that is electrocutes you or someone else later, catches fire and burns down your house, etc. The simplest solution is to get a different charger. Well designed chargers have audible whine taken into account as part of their design. Something slapped together for the lowest possible cost for the mass market from some untracable factory will be more prone to whining.

Another possibility is to put the charger inside a cardboard box. However, be careful that it does not overheat. If the box is several times the size of the charger, then enough heat can still probably be dissipated. Of course that's going to make it bulky. Unfortunately, most things that dampen sound will also thermally insulate. Again, getting a better charger is a better answer.

\$\endgroup\$
1
\$\begingroup\$

This effect that you mention can be caused by an electrical or mechanical disturbance. If the effect were mechanical you would see that the vibrations are causing an analog resonance within the capacitor which sends a signal somewhere. You see this a lot in small PCBs made for wearable devices. This electrical variant is different in that the circuit is creating the resonance, but the fact remains that this resonance can be killed.

For a lower rated capacitor I would say just throw in a tantalum, but since this is a high voltage capacitor you will have to keep the ceramic and add something else. A ferrite bead is possible. The added resistance will create a damping effect. You need to be careful with this, however, that you don't drop too much voltage when adding line resistance. Decoupling caps are likely the way to go, however. Without looking at the circuit I can not say for sure, but they will probably stand up to any spikes.

You touched on this in your question. Either put two in parallel that add up to the same value as the original, or you can actually add a separate path to ground through a decoupling capacitor. Again, without looking, I can not tell you exactly, but if this interests you just look up decoupling caps and your answer will find you. Hope this helps.

\$\endgroup\$
  • \$\begingroup\$ Can you substantiate any of what you've written? \$\endgroup\$ – EM Fields Nov 14 '14 at 12:20
  • \$\begingroup\$ I refer you to this document on multilayer ceramic capacitor vibrations. Do let me know if you have further questions on this. link.springer.com/article/10.1007%2Fs00542-014-2209-5#page-1 \$\endgroup\$ – mcmiln Nov 15 '14 at 20:07
  • \$\begingroup\$ PART 1: Not a bad find, but hiding behind a document with a 35 USD price tag is hardly in the spirit of this site. Moreover, it's rather common knowledge that the Hi-K dielectrics used in many ceramic capacitors are piezoelectric but, more to the point, since the dielectric's dimensional changes are more than likely the cause of the acoustic annoyance, I question your assessment of the "solutions" available. \$\endgroup\$ – EM Fields Nov 15 '14 at 22:12
  • \$\begingroup\$ PART 2: For example: 1. Since tantalum capacitors are electrolytic and, generally polarized, why would you substitute tantalum instead of something with a more passive dielectric - metalized film, say - without knowing the capacitor's function in the circuit? 2. Even considering your caveat, without knowing the function of the capacitor in the circuit or, apparently, that of a ferrite bead, suggesting the use of a ferrite bead in the circuit seems counter-productive. 3. I fail to see how decoupling capacitors are going to ameliorate the situation; could you expound, please? \$\endgroup\$ – EM Fields Nov 15 '14 at 22:32
  • 1
    \$\begingroup\$ ANSWER PART 1: As to the article, all apologies. I just use my school code to get articles so I forget about accessibility. As to the type of capacitor, I fully agree with you that there are other types that may work better, I merely suggested something that I have tried before. I am most familiar with tantalum capacitors. Here is a better article which is free, avx.com/docs/techinfo/mlc-tant.pdf. Look at the section on microphonic effects ameliorated using tantalum. \$\endgroup\$ – mcmiln Nov 16 '14 at 17:22

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.