Four quadrant multipliers:Results of my searching

[Magnus Danielson]

From: jhaible at t-online.de (jh.)
Subject: Re: Four quadrant multipliers:Results of my searching
Date: Sun, 24 Dec 2000 01:55:05 +0100

> > For a DIY yes, but when in commercial buissness, no. Each trimmer is
> > expensive then. Laser-trimming reduces the need.
> 
> Does this mean the 633 requires no trimming ?

"no trimming" is a matter of taste. The laser-trimmed version would
mean no trimmers for more applications than non-trimmed
versions. Then, when you are really picky, then you would need
trimmers again, but then covering a much smaller range and then that
would be less prone to aging of trimmer positions, so there is added
benefit right there anyway.

> I thought there was an expensive no-trim version, and the "cheap"
> one needed trimpots. Maybe I mixed that up with another device.

Possibly. Analog Devices calls this "Low Cost Analog Multiplier".

Their high-spec would be AD534 and AD734 multipliers. The fun part
with those is that they include divide functionallity.

> Now what's the remaining offset voltage (in dB's or % from full range)
> of a trimpot-less 633 circuit ?

They claim it to be within 2 % of full scale, and then we are talking
about "guaranteed total accuracy".

This means output offset voltage of +/- 50 mV maximum and +/- 5 mV
typical. The input offsets have similar numbers (30 and 5 mV).

The benefit of using a device like the AD633 is that if you add
trimmers, you can directly adjust the inputs and output separately and
that by taking only 3 trimmers and 6 resistors. The datasheet gives
suitable values for these to give the +/- 50 mV trimming range.

Looks like a nice little 8-pin chip to me. If only AD could deliver
their backlog that is...

PS. Christmas calibrating my OB-8... ;)

Cheers,
Magnus