common mode terminology

Take your attitude problem and go away. ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
I love to cook with wine.     Sometimes I even put it in the food.

n c

he

g

A simple search of the literature answers your simple question

Where else do you think it would be? Read the data sheet. No, it will not work with 250VCM. Unless you add external resistors in series with the inputs. Calculation of the values and their effect on VCM, accuracy, and xfer function is trivial and left as an exercise for the student. Art

AD629 handles +/- 270V.

Its quite simple. Use a resistive divider scheme at the input to stay within the common mode range of the given amplifier. You'll need high precision resistors (0.1% or better).

--
Failure does not prove something is impossible, failure simply
indicates you are not using the right tools...
nico@nctdevpuntnl (punt=.)
--------------------------------------------------------------

What is your application?

John

Thanks for the note Fred.

Looks great. Lower CMRR than the 117,but I guess you have to pay the man for everything, including enhanced input range. Thanks, Andrew.

Thanks for the time you took to reply Art. Yeah, I had already figured out I could screw up my circuit accuracy that way if I so desire, but since this is not an academic exercise, I was looking for the right component for the job. Will keep looking...

Jack

Fuel cell voltage monitoring

Can you create a pseudo ground at 1/2 the cell stack voltage? That way the INA117's will see max +-125V. An opto isolated serial link can get the data to a different ground reference. Art

Is that essentially a battery stack? There are some cute ways to do that. An instrumentation amp per battery is about the worst.

John

You can hand-match them and get excellent results. I did this for a similar problem. Of course, in large production it is better to have the manufacturer match the resistors for you.

Jon

Thanks Nico. I am trying to avoid this approach for a non-trivial reason. The DUT in question is a voltage source that is highly sensitive to current draw. So if I drop the input impedance of the measurement circuit too much, I won't be able to get a clean open circuit voltage reading. This is still a fall-back option, just not my favorite one. Thanks again for your suggestion.

Jack

Lord no... I agree whole-heartedly. In this case, without giving away the whole farm I will say that the invention uses one set of amplifier equipment to monitor the whole show.

After fifteen years in the fuel cell business I have seen and played with personally a lot of too cute, and some things that work pretty to very well.

Present idea will hopefully be less cute and more effective. It is probably excessively cute on the mechanical side to simplify the electronics.

Jack

Now this I like! I gotta go do some homework to figure out how...thank you very much for an outside the box (my box at any rate) suggestion. Sometimes we stare at our own problems for too long...

Jack

If you can multiplex scan the stack, not need continuous monitoring of every battery, you can have an "odd" bus and an "even" bus, and an SSR from alternate battery junctions to each bus. Turn on two SSRs to connect a battery to the buss pair, realizing that half of the polarities will be flipped.

Then use one real isolation amp off the bus pair, or to be really cheap build a flying-capacitor mux, most of which is done already.

John

Why not just use one real isolation amp?

John

mmon

Vdc

r the

ion

hing

d a

The 200V limitation has more to do with keeping the OA inputs within range than anything else. There are such things as high voltage analog switches that operate in this range, check out

They come in octals which you can parallel and won't put a dent in disrupting the balance required for the INA 90dB CMRR. So switching the INA in and out at about 100Hz with say 20us on time buys you lots of time for the output to settle to within 0.1%, reduces loading on the fuel cell by a factor of 500, and also relaxes the stress on the internal 380K by a factor of nearly 50. Use the technique in the bulletin, say pull the reference to -5V, to get you 20x5=3D100 extra volts of working CMR on the high end.