Op amp power supply sequencing question

[Harry Bissell]

Ooooh... can I play too ???

I've never had a problem with electrolytic caps in this fashion. Most of the time
the power down is so infrequent (as a function of total time) that there really
isn't much chance for damage or degradation. If you are using really BIG lytics...
or running the lytics near their voltage rating then careful design should be
followed.

There should be bleeder resistors to discharge the caps. These should help prevent
any reverse bias (this is the worry, right ???). If you like you can put reverse
biased diodes around the caps (cathode to positive). If there is a voltage
reversal the diode will conduct and make brief.

Most opamps are pretty rugged. The problems are usually that if the supplies do
not sequence up properly the opamp may latch to the negative rail (esp opamps with
the
dreaded "phase reversal if the common mode range is exceeded"). I don't use such
(misbehaved) opamps.

The biggest danger is when a circuit has inputs from an external source, and then
is missing a power supply. The is ESPECIALLY true of CMOS logic. This can cause
latch-up and destruction. Series resistance with the input (always a good idea
between CMOS and the "real world") usually prevents this.

Another hidden problem is with 78xx or 79xx (or similar) regulators. If the input
capacitors can go to ground more quickly than the output, then the reg. gets
reverse biased and can blow up. The reverse diode trick helps here. The other
thing that helps is for the larger capacitance to be on the input side of the
regulator. I put real big ones here
so that the "bulk" supply capacitance is always much larger than the distributed
capacitance (like 10X bigger). And I always use the diodes...

I agree with Magnus... not usually a problem but worth keeping an eye on...

:^) Harry (who just fixed his 20yr old design ADSR's with the LM324 on a bipolar
supply... with reverse latch up.....)


Magnus Danielson wrote:

> From: "Mike I." <mirwin1 at istar.ca>
> Subject: Op amp power supply sequencing question
> Date: Fri, 07 Jan 2000 23:33:55 -0500
>
> > Does anyone know whether having different rise-times (and fall-times)
> > for both sides of a +/- 15V supply can damage op amp circuits (eg. when
> > supply is turned on and when supply is turned off).?
>
> I've never hear that it was a real problem for op-amps themselfs. For
> electrolytic caps however it is a problem. It's a known problem that quick
> steps can make a cap fail, and to help solve that you should shunt them with
> a plastic or even ceramic cap. Keep short leads.
>
> I've never experienced opamps to fail. I don't think they are sensitive to the
> normal range of rise and fall times that we experience. In either case, caps
> should take it down to safe levels anyway.
>
> > I think I read somewhere that  V- should never be applied if V+ is absent
> > (even  momentarily). Is this correct?
>
> This may very well be correct for some op-amps. It is really due to the
> internal design for a particular op-amp. The only way to know is to read the
> datasheet carefully.
>
> Cheers,
> Magnus