Simple Discrete Op-Amp

[RoGeorge]

The resistors are Chinese metal film from the same lot, binned with a 5 digits DMM.  The 10k in the base were +/-1 ohm (10.013, 10.014) and about the same tolerance for the 1k resistors in collector.

As a double check, when the transistors are swapped the DC offset changes its sign, probably less than 1-2mV is caused by resistors mismatch, while the best matched BJT pair found so far produces about 9mV of DC offset.

The TC of the resistors seems good enough.  When touched with the finger (through a plastic bag) the offset stays about the same.

The transistors, however, go crazy with the temperature difference.  I knew Is changes with temperature, but I never measured how much would that mean in practice.  By simply touching the capsule through a ziplock bag, the offset can change by many tens of mV.

The pair of BF458 (TO-126) I am measuring now started with 8mV DC offset, slowly raised for about 5 minutes, and reached thermal equilibrium at about 40mV.  But if I touch one transistor the DC offset slowly grows from +40mV to 100...150mV.  That's more than 300% offset increase!    If I touch the other transistor, the DC offset changes with about the same amount towards the other direction, to negative values. 

Now it makes sense why precision opamps are designed to have a thermal axis of symmetry, and with the critical components carefully arranged with respect to the thermal axis.

[magic]

At first I've connected the ACin straight to the base of the transistors (Rb=0), only to notice the output was not very sensitive with the BJT differences in Beta. 

Didn't see that coming.  Then I've added a 10k in series with the base of each BJT, and it became sensitive enough.  With base resistors, it was easy to spot differences in almost identical transistor, for example Beta 408 vs 410 mismatch was measurable.

β only affects input current, so a mismatch shows up when there are unequal voltage drops across significant base resistors.

A difference with shorted bases is mainly due to Vbe mismatch. Suppose it's 18mV, then you get 1:2 instead of 1:1 current split and it easily shows up when the tail current varies due to varying Ree voltage. Typically transistors from one batch are matched to a few mV, though.

Vbe and β are independent errors and they may add or cancel out, and the cancellation may stop working if you change source impedance, tail current or temperature. It's a good idea to untangle them and measure each independently.

[mawyatt]

By touching the transistor you likely changed the temperature >10C. The transistor delta Vbe/T is about -2mv/C, so one would expect a significant change as you've discovered!!

As mentioned a delta Vbe of 18mv produces a 2X Ic change, 60mv produces 10X, so equalizing device temperatures is very important.

This is why George Erdi's brilliant Cross Coupled Quad became a baseline in IC design as it reduced temperature and IC process gradient effects simultaneously

Best,

[langwadt]

try adding a resistor in each emitter