NNTP-Posting-Date: Fri, 10 Jan 2003 09:42:18 -0600
From: Walt Jung
Subject: Re: Zeroing differential amplifiers (AoE)....
Date: Fri, 10 Jan 2003 10:41:10 -0500
X-Newsreader: Forte Agent 1.8/32.548
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
On Tue, 24 Dec 2002 03:50:13 GMT, Robert Baer
>Jay Andersson wrote:
>> I am trying to zero a differential amplifier using a technique similar to
>> that in AoE Chapter 7, where an integrator controlled by a FET switch
>> provides an offset voltage back to the amp inputs.
Unfortunately, most of this wordage below is not at all
correct, *today*. It was certainly correct in the late vacuum tube op
amp days, and in the early modular solid-state op amp days. But that's
some time ago (nearly 40 years!).
> Chopper stabilized op-amps are inverting and have the non-inverting
>side referred to ground.
Nope. They work either NI or I modes. See some modern part #s,
such as AD8551, LMC2001 and LTC1050 (plus many others).
> They are not useable for differential applications, and by definition
>CMRR is not relevant.
Not! See same parts. CMR is relevant to NI mode operation, but
fortunately these modern amps excel in this area. See typ DS #s; in
excess of 100dB.
> If you want to use a chopper stabalized op amp for differential work,
>then one ugly way is to float the supplies.
Not! Use them as you would any other op amp. Just be reminded
that many of today's parts use low voltage CMOS processes, as opposed
to the classic +/-15V.
> A nicer way is to use two, each as an inverter (for + and - inputs),
>the outputs driving a third amplifier (this scheme is called an
> The offset voltage drift is fixed by the chopper scheme, and the CMRR
>of the third amp is improved by the gain of the drivers - assuming
> There are so many off the shelf op-amps now that have untrimmed
>microvolt input offset voltages, that the low bandwidth and higher noise
>of a chopper system is almost unacceptable.