Re: gEDA-user: Opamps et al...

[Peter Clifton <pcjc2@xxxxxxxxx>]

On Mon, 2006-11-20 at 19:23 -0500, al davis wrote:

> > http://encon.fke.utm.my/nikd/latest/sloa067.pdf
> 
> It can be misleading ..
> 
> I have often tied the inverting input to the output, and ground 
> the noninverting input, perferably through a resistor.  The 
> paper says "NO!" to this.

I thought as much. I woundered if they had their +/- symbols round the
wrong way, implying the designer was attempting to use +ve feedback to
saturate the O/P at one rail.

> The paper is correct if you use a single supply.  With a single 
> supply, grounding an input will force the output to a rail,as 
> the paper says.  With a bipolar supply, grounding the 
> noninverting input is equivalent to biasing it to the midpoint 
> between the supplies.  It will put the output in the middle 
> where it belongs.  Therefore, the "NO!" circuit is actually a 
> good approach with a bipolar supply.

> Since the original question mentioned an "instrumentation 
> amplifier", and a concern about drift, it will probably use a 
> bipolar supply.  Since there is a concern about drift, I am 
> assuming that DC coupling is a requirement, with both positive 
> and negative inputs.

Unfortunately so.. I'm not sure how I'm going to achieve that yet, as I
would have liked it to run off batteries, and in a small box. Like I
said earlier, I may end up using a lab PSU for now.

The intended application is a current shunt amplifier to sense the load
current of a chunky MOSFET + resistor bank.

A further opamp (separate from the boxed instrumentation amp stage)
driving the MOSFET gates, fed with a reference signal should make it a
constant current load. (0-40A @ say 15V). The current shunt is 2mV/A,
and I wanted to measure to say 0.1A at the worst, 0.01A is better.

I will be feeding the amplified signal into a DAQ card at 0.1V/A,
(although the DAQ card I have won't read to 0.01A for the signal range I
need.) 

I wanted to avoid having trimmers in the circuit, so here I am... trying
to design precision analog circuitry on a budget. I now appreciate the
skill which must go into producing proper instruments and DMM etc.., and
realise some of the difficulties in making computerised measurements.

Perhaps if we had the budget to buy some fancy National Instruments card
this would all be easy, as it is, I have one of their cheapest 12bit DAQ
cards.

Peter C




_______________________________________________
geda-user mailing list
geda-user@xxxxxxxxxxxxxx
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user