Most modern OP-AMPs, even single types, do not have an offset pin, but many older single type OP-AMPs had an adjustable offset.
I was recently surprised to see on TI's website the latest datasheet (revised 2024) for the OPA132 (and also the OPA134) that pins 1 and 8, previously assigned to the offset trim pins, are now NC.
From the old datasheet.
At first I thought it was an error in the datasheet revision, however, but the notes on pins 1 and 5 (which may have meant pins 1 and 8) said
‘Existing layouts for the OPA132 before revision C of this data sheet do not need to be redesigned.’
And when I checked, I found that OPA131 and 130 also had pins 1 and 5 assigned to offset trim pins before, but they are now NC as well.
Even the OPA627, in its moulded package, is now NC, and only TO-99 here retains the trim terminals.
(The OPA604, TL071 and others still have trim terminals as before.)
The ones I own were acquired a long time ago and all can be adjusted by adding trim pots, but does this mean that the newly produced ones cannot be adjusted? Or does it mean that they are still connected internally as before but (as TI) do not recommend offset adjustment there?
Of course the correct offset adjustment should be made at the input, and we are aware that adjustment by a degenerate resistor such as a mirror inside the circuit can exacerbate temperature drift and other.
However, is it really acceptable to make such an essential functional change to these OP-AMPs that have been in use for a long time? I very much doubt it.
I was recently surprised to see on TI's website the latest datasheet (revised 2024) for the OPA132 (and also the OPA134) that pins 1 and 8, previously assigned to the offset trim pins, are now NC.
From the old datasheet.
At first I thought it was an error in the datasheet revision, however, but the notes on pins 1 and 5 (which may have meant pins 1 and 8) said
‘Existing layouts for the OPA132 before revision C of this data sheet do not need to be redesigned.’
And when I checked, I found that OPA131 and 130 also had pins 1 and 5 assigned to offset trim pins before, but they are now NC as well.
Even the OPA627, in its moulded package, is now NC, and only TO-99 here retains the trim terminals.
(The OPA604, TL071 and others still have trim terminals as before.)
The ones I own were acquired a long time ago and all can be adjusted by adding trim pots, but does this mean that the newly produced ones cannot be adjusted? Or does it mean that they are still connected internally as before but (as TI) do not recommend offset adjustment there?
Of course the correct offset adjustment should be made at the input, and we are aware that adjustment by a degenerate resistor such as a mirror inside the circuit can exacerbate temperature drift and other.
However, is it really acceptable to make such an essential functional change to these OP-AMPs that have been in use for a long time? I very much doubt it.
I think, it's an indication these historic products might be phased out sooner or later. No designer would use those in a new design anyway as there are better modern alternatives.
OPA672/637 in TO-99 metal can still has the offset trims but those are already marked as NRND. Then again, the untrimmed offset is really low to begin with. If you need lowest offset and 1/f noise etc, take a chopper amp.
Probably only very few customers ever used the trim and TI decided it is OK to crop them as they only add manufacturing cost (bonding).
The OPA627/637 has a relatively detailed equivalent circuit drawn on the old DS, the input is a cascode bootstrapped Pch J-FET with source to output, whereas the new DS has a much simplified functional block, but is replaced by a general differential pair with Nch J-FETs drain-output .Also, a detailed comparison of the characteristic tables shows that those in the SOIC package also differ in terms of output current, open-loop output impedance, etc. The TO-99's seem to have inherited the values of the old ones. It is curious whether these changes have also been made in the actual circuit.
This may be influenced by the fact that semiconductor production is a process where process updates are important and old process equipment cannot be left behind forever.
This may be influenced by the fact that semiconductor production is a process where process updates are important and old process equipment cannot be left behind forever.