build a millivolt generator

[Hercules1UPW]

I need to build a millivolt generator that has a range of 0-1000 mV.

I would like it to have accuracy of 0.1 mV (if possible) and be reasonably priced to build ($0 - $200).

I would like to be able to hook it up to a female BNC connector.

As a side note, This would be a device to calibrate an oxidation-reduction potential (ORP) meter used in my facility. I'm using sodium bisulfite injection to protect water treatment equipment from chlorine.

All help is greatly appreciated. Thanks in advance!

Hercules

 

[BrianG]

For the laboratory the simplest answer is a voltage standard and a precision potentiometer, however I imagine you want a small portable unit, probably battery powered?

You need to supply some more data like: what input current your ORP unit takes. You have said you would like 0.1mV accuracy but what about resolution, i.e. "range 0-1000mV" how would you expect to set it to a desired reading - "decade box" control steps perhaps (0-9mVv, 0-99mV, 0-0.9V) ?

 

[IRstuff]

see:

http://www.analog.com/UploadedFiles/Data_Sheets/AD768.pdf

http://www.analog.com/UploadedFiles/Data_Sheets/AD5680.pdf

The gain error alone is 100 times your required accuracy. You are requiring $6000 performance for only $200. You can, of course, calibrate your way through that, but you'd still need instruments accurate enough to do the calibration itself. And if you did, you could just as easily use those instruments to do what you're asking for.

TTFN

Eng-Tips Policies FAQ731-376

 

[itsmoked]

Just buy a calibrator!

How about one of these. I have one works - better than great!

http://www.transcat.com/PDF/alte422.PDF

If not this one check this site as they have any calibrator you can think of. Maybe a 0-20mA calibrator with a resistor, to convert mA to voltage, would work well too.

Check:

http://www.transcat.com

Keith Cress
Flamin Systems, Inc.-

http://www.flaminsystems.com

 

[8060A]

Actually, you can make a highly accurate millivolt source using matched thin film resistors in a kelvin - varley divider configuration. I have designed and built many millivolt sources of various complexity and resolutions. It can be done with only a 6.5 digit multimeter for matching resistors and reference trimming. High accuracy is obtainable without pots.

Kelvin-Varley dividers have very high resolution capability. However, the output resistance can be high, and therefore it is only accurate into a high impedence. Buffering the output is possible but you might add more error without a carefully designed buffer amplifier.