0
\$\begingroup\$

I want to use a 10k trimmer pot with a 250ppm/°C tempco for offsetting input offset voltage of an LM741 op-amp.

Cermet potentiometer temperature coefficients partially answers my question. And the following goes into great detail about why null offset op-amps are rare nowadays: Modern opamps with null offset.

But I'm asking here, now, to clarify a few things:

  • how to calculate maximum combined temperature coefficient and drift, given specs of an LM741 and the trimmer pot.
  • why "you can expect 10-20:1 improvement if you draw negligible current from the wipe"
  • why "the best approach is to first minimize the adjustment range", and how this might be done. (Is this by "shunt[ing] a cermet pot element with a low tempco resistor? And how would that look in the LM741 application?)

(I prefer using components I'm familiar with, and while I could find one with low VIO offset, it's hard finding one which is as cheap and versatile enough in the ways a 741 is).

\$\endgroup\$
2
  • 4
    \$\begingroup\$ Well, the 741 might have been called versatile in the 1970s but Starskey and Hutch were considered cool back then. Welcome to the modern world. \$\endgroup\$
    – Andy aka
    Commented May 9, 2017 at 11:43
  • \$\begingroup\$ With 100 suppliers of a "ua741", each with different layouts and processing, and different resistive implants for the onchip offset-trim networks, I'd not try to optimize the temperature behavior. Pick a modern low-offset opamp. \$\endgroup\$ Commented May 9, 2017 at 16:37

1 Answer 1

3
\$\begingroup\$

"how to calculate maximum combined temperature coefficient and drift, given specs of an LM741 and the trimmer pot."

That's easy. You can't. You need to pay more attention to both your parts and the data sheet. Look closely at your 741s. What part number? I'll lay very good money that they are LM741C. Now look at the data sheet. See where it says "Average Input Offset Voltage Drift"? Check the right-hand column, under 741C. Notice anything missing? Like a number? That's because there is no spec for offset drift for the 741C. Likewise check out the offset current drift. Nothing.

So, even if by some miracle you have 741s instead of 741Cs, just try to find a max number for either of these quantities. You can't. They simply are not specified, so you take what you get.

Your underlying problem is that, as the first cheap op amp which required no compensation, the 741 was in no position to worry unduly about small temperature effects. There's an old saying, "Faster, cheaper, better - pick any one." and this applies to the cheaper 741. As a counter-example, consider the OP27, which came out a few years later. Go to page 14 and you'll find a fairly detailed discussion of offset/temperature adjustment. The early days of op amp production were marked by major advances in die processing and laser trimming, so it didn't take long to leave the 741 in the dust.

I can understand your attachment to an old favorite, but "it's hard finding one which is as cheap and versatile enough in the ways a 741 is" is just wrong. The 741 isn't versatile (it absolutely requires +/- 15 volts, although you can usually get away with +/- 12), and it isn't all THAT cheap. For instance, on eBay you can get OP27s for the same cost as 741s. Oddly, 741s are a good deal cheaper at Digikey than on eBay, but on a per-amplifier basis something as old as the TL084 is very nearly the same. And a 084 has 4 orders of magnitude lower input currents and can run down to +/- 3 volts. Plus, of course, it's got 4 amps per package.

\$\endgroup\$
3
  • \$\begingroup\$ @Jodes - I appreciate the selection, but you should be aware that it's a good idea to wait at least 24 hours before making up your mind. You never know when somebody else will come up with a better answer, or even (gasp!) prove me wrong. \$\endgroup\$ Commented May 9, 2017 at 13:18
  • \$\begingroup\$ "Faster, cheaper, better - pick any one." - I thought it was any two? \$\endgroup\$ Commented May 9, 2017 at 14:18
  • \$\begingroup\$ Except when you're dealing with 741's. \$\endgroup\$ Commented May 9, 2017 at 15:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.