Lop-sided supplies for an opamp

[rtarbell]

I am using a +10V regulator (78L10) and a -10V regulator (79L10) as the analog supplies for an opamp. The -10V supply seems to "sag", as its output is only -9.1V (but it is a stable -9.1V). If I have designed the opamp circuit to be powered from bipolar supplies, but the supplies are uneven, will this induce some kind of offset on the output of the opamp?

I am trying to find the source of this voltage sag (I am using a +12VDC wall wart [which is more like +16V], a TC962 to invert the +16V to -16V, and then a 79L10 to regulate the -16V to -10V), but I may not continue to spend time looking if my circuit will operate properly with these uneven supplies.

[Conrad Hoffman]

There used to be much more emphasis on matched tracking supplies, but today I don't think it makes much difference. You might check your regulators for tolerance and be sure no oscillation is occurring. If everything is normal, I wouldn't worry about the voltage difference in most circuits.

[BWRX]

As long as the maximum desired input and output voltages of the op amp are within the supply rail voltages then there shouldn't be a problem.

[Mooly]

Put a 'scope on it The -9.1 sounds out of tolerance. Check for any ripple into and out of the regulator.
As other have said the fact the supplies are not equal is no problem.
The scope will prove beyond doubt whats going on.

Edit. If there is any HF ripple present from the DC-DC convertor a DVM may give a false reading. Scope !!!

[rtarbell]

Thank you for your replies!

Mooly: You are right, I am only using a voltmeter (and a crappy $12 one at that) to measure the output voltage. I should hook up my old scope to see for sure. The extra ripple would make sense - I have used the TC962 in the past with no problems, but at this point, the TC962 is the only part in the chain that is suspect. I swapped out one 79L10 for another, and the new one did the same thing, so I don't think it is a bad or blown part.

[Mooly]

That's good . Let us know

[cliffforrest]

Once again I wonder how anyone can believe any measurement in electronics except what can be seen on a good ol' analog 'scope!

(Only slightly tongue-in-cheek )

[Mooly]

Originally posted by cliffforrest
Once again I wonder how anyone can believe any measurement in electronics except what can be seen on a good ol' analog 'scope!

(Only slightly tongue-in-cheek )

Well there is more truth in that than you realise. As a technician I always insisted on having a 'scope as a permanent fixture on my bench. No 'scope no fixxy
I used it for a lot of DC measurements, and the number of times it shows immediately the problem is, well I was going to say unbelievable, what is unbelievable is that in professional environments folks don't use it.
Seen it time after time, er this 5 volt rail at 4.6 volts, do think that's all right, near enough isn't it, when it's actually a 7 volts 100khz ripple they are measuring.
Going down all the inputs on 100 legged IC. Use the scope, see what's really there.

Not a dig rtarbell, it just comes from years of experience.
Let us know what you find

[rtarbell]

You guys are absolutely right!

A little bit of switch noise - I believe - fooled my meter. My meter says "true RMS" on it, but like I said before, its total cost was $12.

I wonder - would a really good meter, like a Fluke, be fooled as well?

[Mooly]

It would A scope is the only way to see whats there. You may get a DVM to "show" what you want to see but thats not the same. It might show 20 volts going in to the reg, the reality may be a load of hash and high frequency noise.