Seventh Circle Audio

hey...

I recently finished an a12 card, but the phantom wouldn't work. Doing the molex connectors, I may have very well shorted one a b+/- to ground, blowing a fuse. At this point it's been a awhile, so I'm a little hazy with the play by play idiocy of that day.

In any case, the card is powering up, and there's proper (+/- 18v) at the pins for the opamp (1122 Avedis), but when the op amp is inserted, things go wrong. Mainly, I get a pulsating, chugging bit of sound at the output. It definitely sounds the way a wave looks on a screen. I think. Yikes. And, the opamp (and the card in general from what I can tell) also starts to heat up.

I guess I'm wondering what I can rule out if I'm getting proper voltages on the opamp sockets.

Thanks!

kelly

I was wondering what happened to you. I haven't seen any posts lately so I figured you got it working, but I guess not. It does sound like you've killed another op-amp, probably by having only one power supply rail connected for an extended period of time. Try disconnecting the load and measure the voltage at the output of the op-amp. The easiest way to do it is by lifting one of the load isolators (or jumpers, if you didn't install the load isolators). Let us know what you find.

Hi Tim,

What do you mean "lift" one of the load isolators? Do you mean disconnect it from the board? I want to make sure before I go desoldering things. And which one? Maybe it doesn't matter, I'm just being extra careful.

Thanks,

Kelly

Yes, disconnect it from the board. I'd lift L1, pin 1, and connect a probe to the loose end. Power up the board and measure the no signal voltage. It ought to be just about zero. Let us know what you find.

Hey...

Vdc is .04 at L1 pin 1.

Thanks for you help. Does this suggest the opamp as culprit?

Kelly

No, it doesn't. Are you making the right measurement? If you lift the jumper at L1, pin 1 and then measure the voltage at L1, pin 1, with respect to PGND, anything other than 0V would be a big surprise. What you want to measure is the no-load voltage at the output of the op-amp. If the op-amp is damaged, it's likely that the output voltage will pull toward either power supply rail. If the op-amp is wired to the transformer primary, you'll see little voltage swing but a lot of op-amp heating. Make sense?

BTW, if you've robbed R16 from the A12 to put in your N72, all bets are off. You must have R16 in the circuit, otherwise the op-amp is running open loop.

Hi,

I'll have to wait until I get another r16 then. I assumed (and wrongly) that I could simply jumper pins one and three on the board to take the resistor out of the circuit. If it's not attenuating, it's not in the circuit...that was my theory. This is wrong, I guess.

I'll keep at it. I've got another opamp and r16 on the way.

Thanks.

kelly

No, your logic is correct - you can run with no r16 if you jumper from 1 to 3.

Tele is right, a jumper there will work fine. Or you could stick a 10K resistor in there instead, which will give you the same maximum gain as the pot. The jumper will cut the maximum gain in half.

tpryan \$m[1]:What you want to measure is the no-load voltage at the output of the op-amp. If the op-amp is damaged, it's likely that the output voltage will pull toward either power supply rail. If the op-amp is wired to the transformer primary, you'll see little voltage swing but a lot of op-amp heating. Make sense?

Actually, I don't really understand this, but I was measuring without the op amp actually installed. I sent it back to Avedis to test.

Still, if you have a second:

1. What is no-load voltage?
2. What does it mean that ouput voltage will "pull" toward either supply rail?


Thanks.

Kelly

1. What is no-load voltage?

Pretty much what it says, the output voltage with no load. The transformer primary represents a very low DC resistance which will affect the measurement I was asking you to make. Disconnecting it allows the op-amp output to swing without having to deliver current to a load.

2. What does it mean that ouput voltage will "pull" toward either supply rail?

Very briefly, you can think of the op-amp output circuit as two valves, one after the other (in series) connected between the positive and negative power supply rails, with the output connected to the juntion where the two valves meet. If you open the top valve while closing the bottom valve, current flows from the positive supply to the output. If you open the bottom while closing the top, current flows from the output to the negative supply. If one valve (output transistor) is damaged (shorted), there's no way to shut off the current to the output, and the output voltage is "pulled" toward that supply rail.

And, just to be clear, the test Tim suggested would require the opamp to be present.

Thanks for the explanations about opamp stuff. Now that part makes sense.

kelly

Hi...

Finally...the card is working again (with a replacement 1122 from Avedis).

At least, it's working out of the box, on the bench with a jumper for r19.

Two quick questions:

1. (If you have a second...I can probably figure this out on my own.)I lifted L1 to take a measurement before, and now it's getting messy getting it back in. If it gets bad, where do L1 pins 1 and 2 connect? On the top of the board or the bottom? I'm pretty much deciding whether the top or bottom should take the brunt of the soldering/desoldering heat.

2. Tim, at L1 I'm getting roughly 8mA when lifted from the board. Sound right to you?

Thanks again...I'm so close.

kelly

1. (If you have a second...I can probably figure this out on my own.)I lifted L1 to take a measurement before, and now it's getting messy getting it back in. If it gets bad, where do L1 pins 1 and 2 connect? On the top of the board or the bottom? I'm pretty much deciding whether the top or bottom should take the brunt of the soldering/desoldering heat.

Pulling one of these leads out shouldn't damage the board. If it looks like a pad is pulling loose, you're using too much heat. Just be careful.

2. Tim, at L1 I'm getting roughly 8mA when lifted from the board. Sound right to you?

If you mean 8mV, that sounds about right. You should be able to adjust it to 0mV using R19.