I'm currently using an INA196 high side shunt amplifier in a couple of=20
applications.   As I asked/discussed a couple of weeks ago, one of them=20
involves current monitoring the currents through some P-channel FETS,=20
and am looking at moving the sensing from a sense resistor to sensing=20
across the FETS.

I of course need to deal with the situation where the FET is off,=20
producing a high differential voltage across the amplifier.   I could do=20
this easily in software EXCEPT the INA196 only handles +-18V of=20
differential input, and this happens to need to support up to around 62=20
or 63V... which is fine with the common mode range on the INA196, but=20
like I said not so much on the differential spec.

I also use the INA196'es on a couple of other circuits to monitor=20
customer-provided shunts, and have seen some failures related to what I=20
believe is excessive differential voltage inputs as well.   For this=20
reason, I'd like to avoid fixing the above problem with something driven=20
along with the FET, and instead handle this as a separate circuit.

I think at this point, I need to either protect the INA196 from a high=20
differential input, or switch to a difference amplifier.   However, I'm=20
finding this to be a struggle for the following reasons:

1) I am unable to find any difference amplifier which runs at 3.3V and=20
supports the necessary common mode range of 0 (-2 would be better) to=20
65V.   Some are close (the AD8207 for instance looks great until you=20
realize the +65V is only at 5V).   I'm ok with one which has an=20
integrated ADC as well...    I also am hoping for something which is a=20
$1-$2 part, not a $5 part, since I use a lot of these on these boards.

2) Adding protection to the INA196 seems doable, except it also needs to=20
protect 'open' (I can't have it shorting across the shunt and taking=20
current), and have minimal impact on the accuracy of the INA196.   A=20
depletion fet seems reasonable, except the high Ron which seems very=20
likely to mess with the accuracy of the INA196.  Other options seem=20
prohibitively expensive.  Add the fact that we're turning the fets off=20
once a second for a few microseconds and this circuit has to be=20
transient free... well >18V transient free I guess is more accurate.

I'm hoping someone knows of a good, relatively inexpensive, difference=20
amp which I could use in this circuit, OR alternatively has an idea for=20
a relatively inexpensive protection circuit which effectively=20
disconnects the INA196 when a differential voltage over a certain=20
threshold is seen.

Anyone have any ideas?

-forrest
--=20
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist
.