1. Home
  2. Electronics Design
  3. reverse opamp supply limit (AD820)

reverse opamp supply limit (AD820)

Anyone know how much voltage and current it is safe to feed an AD820 opamp through the supply pins, with polarity reversed?

I'm working on a battery-powered device that might have the batteries (2x CR2032) installed backwards by the user. I can clamp the reverse voltage with diodes; just wondering whether clamping it to two diode drops is enough or whether I need to clamp it to one. Or, for that matter, maybe the internal resistance of the batteries is enough of a current limit; internal impedance is typically >20R and initial cell voltage is 3V, so two cells in series can deliver at most 225mW into an optimally matched load. If a reverse-powered opamp looks like a single diode, it would see 0.7V at 58mA, or 40mW. Would that hurt it?

If it matters, I think that typically the reverse-battery situation would only be applied for a couple minutes at most. But worst case it would be for the time it took the batteries to discharge. At high current that is probably 2-3 hours.

I sent some mail to the folks at Analog but got back some boilerplate about protecting against negative input voltages, which is not the issue here.

Thanks!

Reply to
Walter Harley
Loading thread data ...

Instead of worrying about effectively shorting the reversed batteries, why not put an anti reverse current switch made with a small logic level mosfet is series, that blocks the reverse current? You connect it so that it is conducting backward, through the body diode, in parallel with the turned on channel, during normal operation. But if the batteries are backward, the body diode is reverse biased, and the gate voltage is below zero volts.

Reply to
John Popelish

That's a good solution, probably better than the diode albeit a tad more expensive. But if I can get away with no solution at all (that is, if -6V through >40 ohms is safe) then that would be even better...

Reply to
Walter Harley

Reverse polarity protection is a function of the CR2032 battery holder. see

formatting link
for details.

Reply to
Daniel A. Thomas

It is a function of *some* CR2032 battery holders. But not all of them; and in particular, not the dual-battery holder that fits in the very tiny and awkwardly shaped space that mine unfortunately needs to fit in.

Follow-up:

The fine folks at Analog were unable to tell me anything about what kind of reverse voltage or current their chip could stand. So, it looks like I'll be incorporating John P's suggested MOSFET. Oh, for the days when companies published the schematics of their analog ICs.

As an aside, I was flabbergasted at the response of one tech supporter there, excerpted here:

I'm sure glad this fellow doesn't design life support systems. The idea that it is okay for a system to be destroyed by a single easy-to-make mistake is - well, I just can't believe anyone could actually say that out loud.

Reply to
Walter Harley

Yes, I've grabbed the wrong end of a soldering iron more than once. So putting batteries backwards is certainly within the realm of possibilities...

Reply to
Daniel A. Thomas

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.