Tweet
Here are your voltages:
Can you confirm the voltage on U68-1 (=U3-11). Is it 3.5V or -3.5V? (I guess it is negative as U68-2 is positive).
If U77-1 is 0V, then the left end of R620 must be 0V (unless there is a broken track!). The right end of U620 has a small negative voltage on it. This point is known as a virtual earth. Virtual earths are one of the cool things about op-amps. A bit too complex to explain here. A virtual earth point is not connected to earth, but behaves electrically as though it is. The fact that you have a small negative voltage there is a big clue. As U80-1 has an almost full-rail positive voltage, the only place the negative voltage can come from is out of U3-12.
This still looks like a faulty U80, but as that has been changed, I think we must have a duff U3.
Can you check all the voltages on U3 please including the ones used by the other channels.
This is what I would expect:
1 should be at or near +15V
16 should be +15V
9 should be -15V
3, 6, 11 and 14 are the control voltage pins for the 1st 4 channels. Voltage will be signal and Threshold dependant. Worth comparing the channels though. We know 11 will look strange but that is because of the fault, not the cause of it.
All other pins should be 0V - Clearly pins 10 and 12 are not, but they should be.
The datasheet is here: http://www.milton.arachsys.com/nj71/pdf/ssm2164.pdf
Mackie have pretty well built the circuit on page 8. All they have done is to loop the output back to the input. I think that by putting the VCA in the feedback loop of the op-amp they achieve a lower distortion at low levels and a higher distortion at high levels when you wouldn't hear it as it is hidden by the loud signal.
Any clever clogs reading this have a better explanation?
Do you have an oscilloscope? If so, can you look at the VCA outputs to see if anything is oscillating? Sometimes, when something oscillates, the signal can get rectified by a diode or transistor inside a chip and so manifest itself as a DC voltage. C608 is 20pF in Mackie's design but shown as 100pF in the datasheet. If we do have oscillation, you could try bridging a larger value across the 20pF to see if it changes. It is there for stability. Too large a value will only roll off a bit of the treble. The chip can operate up to 500kHz so no worry there. It does also mean that it could be hooting way out of the audio range.
If you don't have a scope, put your multimeter on AC volts or millivolts and attach a capacitor (anything you have, say 1uF) between the black probe and ground. If you still see a signal where you were seeing DC, it must be oscillating as the cap is blocking any DC from reaching the meter.
If you do take U3 off the board, I am pretty sure that the whole mixer will run quite happily without it - apart from having no compression of course. If U80-1 reverts to 0V with no U3, then you have found the problem.
Martin
Can you confirm the voltage on U68-1 (=U3-11). Is it 3.5V or -3.5V? (I guess it is negative as U68-2 is positive).
If U77-1 is 0V, then the left end of R620 must be 0V (unless there is a broken track!). The right end of U620 has a small negative voltage on it. This point is known as a virtual earth. Virtual earths are one of the cool things about op-amps. A bit too complex to explain here. A virtual earth point is not connected to earth, but behaves electrically as though it is. The fact that you have a small negative voltage there is a big clue. As U80-1 has an almost full-rail positive voltage, the only place the negative voltage can come from is out of U3-12.
This still looks like a faulty U80, but as that has been changed, I think we must have a duff U3.
Can you check all the voltages on U3 please including the ones used by the other channels.
This is what I would expect:
1 should be at or near +15V
16 should be +15V
9 should be -15V
3, 6, 11 and 14 are the control voltage pins for the 1st 4 channels. Voltage will be signal and Threshold dependant. Worth comparing the channels though. We know 11 will look strange but that is because of the fault, not the cause of it.
All other pins should be 0V - Clearly pins 10 and 12 are not, but they should be.
The datasheet is here: http://www.milton.arachsys.com/nj71/pdf/ssm2164.pdf
Mackie have pretty well built the circuit on page 8. All they have done is to loop the output back to the input. I think that by putting the VCA in the feedback loop of the op-amp they achieve a lower distortion at low levels and a higher distortion at high levels when you wouldn't hear it as it is hidden by the loud signal.
Any clever clogs reading this have a better explanation?
Do you have an oscilloscope? If so, can you look at the VCA outputs to see if anything is oscillating? Sometimes, when something oscillates, the signal can get rectified by a diode or transistor inside a chip and so manifest itself as a DC voltage. C608 is 20pF in Mackie's design but shown as 100pF in the datasheet. If we do have oscillation, you could try bridging a larger value across the 20pF to see if it changes. It is there for stability. Too large a value will only roll off a bit of the treble. The chip can operate up to 500kHz so no worry there. It does also mean that it could be hooting way out of the audio range.
If you don't have a scope, put your multimeter on AC volts or millivolts and attach a capacitor (anything you have, say 1uF) between the black probe and ground. If you still see a signal where you were seeing DC, it must be oscillating as the cap is blocking any DC from reaching the meter.
If you do take U3 off the board, I am pretty sure that the whole mixer will run quite happily without it - apart from having no compression of course. If U80-1 reverts to 0V with no U3, then you have found the problem.
Martin
Attached Files
All channels work well and are identical now.
Comment