OP184 offset null pins

Document created by analog-archivist Employee on Feb 23, 2016
Version 1Show Document
  • View in full screen mode

We need OP184 or similar device which has still the NULL pins for offset
adjustment (so we don’t need the latest auto zero parts). Does OP184 have this
feature? Because pin1 and pin5 on page 1 of the former datasheets shows them,
but no details can be found in the rest of the datasheet. The other selection
criteria would be RRIO so rail-to-rail for inputs and outputs. Please let me
know if OP184 could fit or which devices would you recommend.


Currently there is no other single, +/-15V, RRIO amplifier in our portfolio
(only OP184) which meets your requirements. We changed the designation of pins
1&5 from "NULL" to "DNC" (DO NOT CONNECT) in datasheet rev. E. However, OP184
still has the offset adjustment because there was no design/die change which
removed this function. Also, we did not remove the bonding wires to the pads,
so Pin 1&5 can still be used for offset adjustment (rev. J).

Why we suggest to leave pins 1&5 as DNC -> please take the following remarks
into consideration:

We used one of the nulling designs for OP184 that we have for bipolar rail to
rail input amplifiers with two differential pairs. As said, the null pins
should be able to null the offset voltage of the amplifier, which is a couple
hundred of microvolts. We do see customers with several gain stages trying to
use the offset adjust pins to trim the overall system output offset. This is
not recommended, the null pins should NEVER be used to null a system offset.
You should be careful with using the null pins if your input common mode
voltage goes from rail to rail. When the null pins are used, the p type diff
pair is trimmed and offset looks good. When input common mode changes such that
the n type diff pair is active, you’ll get worse performance over the common
mode range. Vos of the n type diff pair is not trimmed with the null pins.
Also, the tempco of the external pot does not match the tempco of the internal
resistors, so even if you squeeze out a few tens/hundreds of microvolts at
room, it could actually be worse at other temperature. From E grade, Vos over
temp is 165uV anyway, adding a pot with a tempco might not be a good idea. With
Ib of 600nA, offset voltage reflected at the output can easily be greater than
165uV (think of a 1Kohm feedback resistor). Also, nulling an op amp offset
usually causes the drift / temp co to be worse.