matched pairs/quads (linear)

[Cal Grant]

On Thu, 15 Jul 1999, Martin Czech wrote:

> In linear mode (Ube1=Ube2, balanced) it looks much like a differential
> amp. You want that the reference current is perfectly mirrored into
> the timing cap. Therefore quite high hfe would be nice.  A low offset
> voltage will enshure low offset drift, and therefore the base impedances
> (resistors) on both sides should also be equal, otherwise base currents
> cause offset and drift.

Martin,

The LM394CH satisfies all the above criteria, and in my mind is worth the
extra cost.  My main concern these days is the value of the emitter
resistor coming from the op amp output, and going to the emitters of the
matched pair. My present setup uses 4.7K, but I have seen as high as 47K
used. 


> 
> We always look at the differential pair, but for low currents the opamps
> become important as well.

I have found the Linear Technologies LT1056 to be satisfactory.  Also, I
have not found it difficult to obtain the 3300ppm/*C PTR's. One just
calls their favorite company (I use KRL) and catches them when they are in
production for the value you want, and then add your request to the order
made by some large company. They will sell them at the bulk price at that
point. Be prepared to alter your design value slightly to work with the
value they are making that day. 

As well, has anyone experimented with the value of the resistor used with
the bulk-compensation diode, a rarely used method apparently.
 

cj