12
\$\begingroup\$

I'm curious about how MURATA variable capacitors work, for example, parts from the family LXRW0YV330-056. As far as I can tell, these are not semiconductor varactors, but based on some other physical principle; does anyone know what it is? There is no information on the data sheets except that they are "ceramic".

My interest is that I would like a voltage-tuned capacitor where the input capacitance on the control voltage line is as small as possible, but the capacitance swing per input volt is as large as possible. Also, I would like these to work at cryogenic temperatures, where semiconductor varactors tend to fail. Any input would be greatly appreciated!

\$\endgroup\$
5
  • 1
    \$\begingroup\$ Interesting question. As you point out, Murata seems to have deliberately left out any information about the mechanism of action, possibly to protect a trade secret. Could be some type of MEMS variable capacitor structure. \$\endgroup\$
    – John D
    Aug 11 at 20:03
  • 4
    \$\begingroup\$ Maybe the external voltage is used to polarize the media similar to how a bias reduces the capacitance of an MLCC? Could possibly test that by measuring the impedance between the bias port and the capacitor terminals. \$\endgroup\$ Aug 11 at 20:08
  • 3
    \$\begingroup\$ The C-V curve looks very much like a X7R cap \$\endgroup\$
    – bobflux
    Aug 11 at 20:32
  • \$\begingroup\$ There's no tempco listed in the data sheet, and the temperature range is quite restricted. If they are ferroelectric capacitors, I would not expect them to get down to cryo temperatures any better than varactors. \$\endgroup\$
    – Neil_UK
    Aug 12 at 5:37
  • 1
    \$\begingroup\$ My practical experiences from trying that chip: 0.4 x 0.4 mm chip is really small; the capacitance adjustment somehow didn't seem to work very well for wideband signals and is sensitive to DC bias on the RF pins. \$\endgroup\$
    – jpa
    Aug 12 at 9:01

1 Answer 1

13
\$\begingroup\$

See US 9,047,524

The structure shown in the patent incorporates the resistor network R21-R25 whereas the capacitors OP mentions have those resistors (where needed) externally, but otherwise the function is exactly similar.

The principle is a series array of ferroelectric capacitors with bias voltage applied in parallel through high-value resistors that are incorporated in the IC-like structure.

enter image description here

From the datasheet you can see the relationship between the bias voltage (P13 to P14) on the capacitance (P11 to P12):

enter image description here

In this case, C1..C6 are about 200pF each at 0V bias.

I think you could make something like this with small ceramic capacitors and discrete resistors. For example, this Murata capacitor (curve from Murata's online generator):

enter image description here

So with 6 capacitors you'd get a change from about 17.5nF to 13nF for a bias change from 2V to 15V (typically, unlike the purpose-built part there are no guarantees).

If it's not obvious- the AC voltage across each capacitor would be 1/6 of the total voltage, so the resulting effects of the capacitance modulation (eg. distortion) will be much reduced compared to one capacitor. It's not quite as effective as the back-to-back varactor method, of course.

\$\endgroup\$
2
  • 1
    \$\begingroup\$ Thanks for finding that! I expect you are right. So if I understand correctly, I'd expect input capacitance at P13 i that diagram to be gong to be larger than that between P11 and P12 at low frequencies? \$\endgroup\$
    – Andrea
    Aug 11 at 23:37
  • 2
    \$\begingroup\$ P13 and P14 are the bias terminals of the variable capacitor. The capacitance between P11 and P12 is inversely related to the voltage between P13 and P14. \$\endgroup\$ Aug 12 at 0:56

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.