-1
\$\begingroup\$

I want to make a cheap variable gain amplifier using an op-amp, using a MOSFET as the variable resistance. The circuit is for an audio application at line (signal) levels (ballpark of 1v peak to peak).

I need help in finding a suitable MOSFET. At the moment I'm having to look at datasheet one at a time, looking at its graph of Vds vs Id (with multiple plots for different Vgs values). I have no idea how to refine what datasheet I choose to look at. Tips? What criteria can I use to narrow search results?

Also, the graphs often neglect to show the curves for Vgs values close to the gate threshold voltage - which is important to know, since the linear region for those values span the smallest Vds range. (If that makes sense).

What am I doing wrong? I'm sure this is a very common use case, yet it seems an uphill battle.

Edit: I'm making a so-called audio compressor that maintains a more constant volume when the input varies. I believe this is also called AGC, although a compressor tends to offer more control, such as user settable attack/release rates, thresholds and so on. The gain will be varied by about a factor of 5. (For my application, the input audio source is already normalised. The output of this stage will go to a totally separate stage that actually controls the volume in the regular fashion. (From muted up to full volume.)

An example of what I want to do is here, but I want to explore other ways of doing it, for educational purposes really. Also, amplifiers with voltage controlled gain seem to be very expensive, and I don't trust myself not to blow it up. A design with more numerous but cheaper components means mistakes will be significantly cheaper!

\$\endgroup\$
  • 1
    \$\begingroup\$ I doubt that you will find a suitable MOSFET as most of them are designed to be used as a switch with low Rds_on. To achieve that low Rds_on they have a very large W/L so a small change in Vgs will result in a large change in conductivity between drain and source. It will be a huge challenge to keep that Rds in the region where you need it to be. \$\endgroup\$ – Bimpelrekkie Aug 19 '17 at 19:23
  • \$\begingroup\$ Don't be so sure \$\endgroup\$ – Tony Stewart Sunnyskyguy EE75 Aug 19 '17 at 19:23
  • 2
    \$\begingroup\$ @TonyStewart.EEsince'75 What a useful comment, care to elaborate so that I and everyone else here can learn something ? \$\endgroup\$ – Bimpelrekkie Aug 19 '17 at 19:25
  • 2
    \$\begingroup\$ Maybe a MOSFET designed for RF use would be more suitable. The "oldfasioned" way of doing what you want is using a JFET. Another way which might result in less distortion is by using an OTA, for example LM13600. And the "modern" way is of course to just use a chip designed for volume control. \$\endgroup\$ – Bimpelrekkie Aug 19 '17 at 19:28
  • 1
    \$\begingroup\$ Still not getting your point, those IRF532/IRF9532 have a very low Rds_on. In that amp they're source followers so mainly used in saturation mode, then you actually want large W/L to keep Vgs small (and not lose much headroom). Still unclear how that would help when used as a variable resistor for volume control other than on/off. \$\endgroup\$ – Bimpelrekkie Aug 19 '17 at 20:32
2
\$\begingroup\$

Check out the CA3080 CA3080A, this might be the best way to do what you want. Bob K.

\$\endgroup\$
  • 2
    \$\begingroup\$ Note that the CA3080 is ancient, it is an OTA just like the LM13600. \$\endgroup\$ – Bimpelrekkie Aug 19 '17 at 21:29
  • \$\begingroup\$ I didn't know an OTS was what I needed, thanks for the tip! LM13600 is half the price per channel, and surprisingly cheap in the first place - perfect :) \$\endgroup\$ – CL22 Aug 20 '17 at 3:12

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.