Stumped... Liquid Drive voltage readings off

[remmelt]

Hello all!

I've been trying the Liquid Drive but can't seem to get it right. I've built it with all stock parts, including the Burr Brown OPA2134PA. There is sound on the output and it's louder than the input, but not distorted at all. I've checked the voltages:

V = 9.36
1 = 4.26
2 = 3.67
3 = 1.72
4 = 0
5 = 0
6 = 0.09
7 = 0.09
8 = 8.53

As you can see, the 2nd and the 3rd pins are off. 2 should be equal to 1 (how can that be? There is a 1M resistor between them?) and 3 should be 2V higher.

There is also a lot of oscillating going on when I turn up the drive pot past about 75%. I've first soldered it to a piecce of perfboard, with the above results. I'ts now on a prototype board and the result is exactly the same! The only parts I've re-used are the opamp, the pots and the diodes. I've measured the pots, they are fine. I have replaced the diodes with a triplet of 1N34A's, which does increase the distortion a little bit.

So the only weak part would be the opamp, right?

The sound is not too bad, there's a nice boost but it sounds tinny, too much high end. Sustain fizzes out a little bit. When pushing against oscillation this turns bad. Very tinny, no sustain to speak of.

Any hints or remarks would be greatly appreciated!

[johngreene]

Check the junction of the 3 1Meg resistors. It should be 1/2 VCC. If it is, then the non-inverting input of the opamp is drawing current for some reason and you'll need to figure out why. With that much difference in voltage between pins 2 and 3, I would expect the output to be really close to 0V.

--john

[remmelt]

Voltage @ junction of 3 x 1M is 2.86V.
Current consumption of the entire thing seems pretty high: 8.6~8.9 mA depending on the amount of Drive.

So the junction voltage is not 0.5V. Could you elaborate on what "the non-inverting input of the opamp is drawing current for some reason" means?

For what it's worth, the output DC voltage is around minus 0.8V... That's negative! I'm confused!

[johngreene]

So the junction voltage is not 0.5V. Could you elaborate on what "the non-inverting input of the opamp is drawing current for some reason" means?

Sure, the OPA2134PA is spec'd to have less than 10pA of input bias current. The junction of the two 1 Meg resistors should be 1/2 of VCC or 4.265V given your pin 8 measurement. You only measure 2.86V. This means that -something- is drawing current from this junction in order to drop the additional 1.405 volts. This equates to 1.405/1Meg = 1.405 uA of additional current draw. You are measuring 1.72 Volts at pin 3 of the opamp and 1.72 + 1.405 = 3.125V which implies that it isn't the input of the opamp that is drawing all the current. Possibly the 1uF cap may be leaky. Try removing it and then measuring the DC voltage. If the DC voltage changes, replace the cap. Make sure everthing is clean of flux and stuff because we're talking uA of current draw here and that's pretty easy to get through a dirty board.

8.6 mA is not an excessive amount of current draw for this device. It is spec'd to have a quiesent current draw of 4-5 mA per amp. So that unused opamp is drawing half your power.

I think your whole problem is with the bias tree connected to pin 3. You can try changing the two 1 Meg resistors connected between VCC and ground to 10K to swamp out any leakage currents but leave the 1Meg connected to pin 3. The bias tree will draw 100 times more current but that is small compared to the 8.5mA the opamp is drawing (8.53/ 20k = .4265 mA).

--john