I provide the link here:


Here at page 153, a carrier frequency amplifier explained. It is explaining how it is modulating amplifying and then demodulating the signal. What I didn't get is it what is the advantage to use this method rather than simply amplifying the signal. What use is the carrier frequency here comparing to usual instrumentation amplifiers?


Back in the day when op-amps and instrumentation amplifiers didn't have amazing DC input offset voltage levels and temperature stability, AC excitation of strain gauge bridges were used in conjunction with high-gain AC amplifiers.

Using a high gain AC amp meant you just AC coupled the amp to the bridge and forgot all about DC drift problems.

This solved the problem of dc drift but these days a typical op-amp can be bought that is superior by a mile compared to the middle years of the last century and AC amplifiers are nowhere near as commonplace.

  • \$\begingroup\$ What was the problem back in the days? I didn't get it. Do you mean that in the old days the DC amplifiers didn't have zero balance and coarse settings? \$\endgroup\$ – user16307 Dec 2 '14 at 15:45
  • \$\begingroup\$ They had offset null circuits but temperature stability and drift were a real problem. \$\endgroup\$ – Andy aka Dec 2 '14 at 15:55
  • \$\begingroup\$ sorry but maybe i have problem with the terminology here thats why it is still blurry. what do you mean by dc drift? do you mean a kind of offset changing during time? and if so why ac amplifiers don't have that? \$\endgroup\$ – user16307 Dec 2 '14 at 15:59
  • 1
    \$\begingroup\$ Input offset voltage is small but present in all opamps. It isn't a fixed value and drifts with temperature and time. Ditto input bias currents. Try checking some opamp data sheets. \$\endgroup\$ – Andy aka Dec 2 '14 at 17:57
  • 1
    \$\begingroup\$ so dc drift means "changing input offset voltage". sorry for making this longer but believe me the biggest problem im encountering is definitions. it is like trying to understand a sentence without really grasping what its words mean. thats why i insisted on picturing "dc drift". thnx i will read about input offset voltage. \$\endgroup\$ – user16307 Dec 2 '14 at 18:36

As Andy says, such carrier frequency amplifiers (such as "chopper stabilised" amplifiers) eliminate DC offsets, which can be hand trimmed, but may return as temperature changes or the equipment ages.

However, AC techniques like chopper stabilised amplification can also avoid LF noise aka flicker noise.

DC coupled amplifiers do still exhibit increasing noise levels below 10Hz, and laser trimming their offset voltages to amazingly low levels won't help that.

So the value of carrier frequency amprlifiers has been reduced, but not entirely eliminated.

And the ADA4528 linked by Andy is a chopper-stabilised amplifier - with the interesting twist of servoing out the error component instead of simply AC coupling - but nevertheless a simple-to-use, convenient, low power, chopper stabilised amplifier, thus proving the continuing value of the basic technique.

The Application Note describes how this amplifier eliminates (or rather, re-directs to the 200kHz carrier frequency, where it usually won't matter) the LF noise problem.

  • \$\begingroup\$ Maybe take a look at the ADA4528 to see how low its flicker noise is!! analog.com/en/precision-op-amps/low-offset-voltage-amplifiers/… \$\endgroup\$ – Andy aka Dec 2 '14 at 12:17
  • \$\begingroup\$ Moderately noisy, but impressively low power. Thanks for the suggestion!. \$\endgroup\$ – Brian Drummond Dec 2 '14 at 12:27
  • \$\begingroup\$ @Andyaka I wonder where the 1/f corner frequency really is typically- maybe 0.1Hz. It's off the curves just like that nasty 200kHz noise spike is off most of curves (to their credit they put one with it on the front page). \$\endgroup\$ – Spehro Pefhany Dec 2 '14 at 12:46
  • \$\begingroup\$ AC and chopper-stablized are two different things. AC filters out DC, but won't necessarily do anything about input offset voltages of the amp itself. Chopper-stabilized amps sample the input offsets on the amp, and remove them. By their nature, Chopper-stabilized amps, being sampling systems, are subject to aliasing. \$\endgroup\$ – Scott Seidman Dec 2 '14 at 13:39
  • 1
    \$\begingroup\$ I partially answered that in the first paragraph. If that's not enough, it might make a good new question. \$\endgroup\$ – Brian Drummond Dec 2 '14 at 16:36

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.