Help with squealing MXR Phase 90

[52 Bill]

I tried what you suggest as turning the rate knob does change the sound of the squeal slightly.

Are you saying the rate control does work? That is, it changes the speed of the control voltage oscillator.

[Gypsey Eyes]

yeah it definately does something - but the squeal is still horrible.

[52 Bill]

Try this.

First you don't need to plug the pedal into the amp for this test, so don't worry about the screech. Plug a cord into the input jack to turn on the battery. Take your voltmeter and set it to a range that will handle 9 volts dc. Ground the black meter lead to any convenient ground point like the metal case. I will usually stick the probe into one of the four screw holes for the bottom plate. To test the ground connection of your meter, touch the red lead to the battery terminals and see if you can read 9 volts dc.

Now take the red meter lead and touch it to one of the terminals of the rate pot and see if there is a voltage there. If the oscillator is working, there should be a voltage that rises and lowers at a rate set by the pot. If this voltage is there, then you know that the oscillator is working.

If the oscillator is working, and the pedal will pass signal when it is on, then you may assume that the TL062 dual op amp is ok, as this chip controls the input buffer and the oscillator circuit.

If the oscillator is not working, then you will need to troubleshoot the circuit around the TL062 chip.

Whenever I have to work on anything that has been worked on by a novice, I will inspect every possible thing that the hacker could have worked on. One problem with modern pc boards is that they tend to be fragile. Often the pad where the part is inserted will break off from the board and the part will no longer connect to the circuit. Then of course there are always bridged solder traces and broken part leads. Be sure to check everything on the board, even though the original owner said he only did one mod he may have done other things as well.

[Gypsey Eyes]

Hey thanks for the tip - I checked the voltage and it looks like the oscillator and TL062 chip are okay. It must be a short or bad part somewhere further down the circuit.

[52 Bill]

The only other things are the FETs, a transistor and the quad opamp (TL064).

Based on the symptoms, the TL064 would be my first guess. Using the meter the same way as before, read the voltages on the pins of the TL064. Because of the single voltage power supply, there will be voltages on all of the outputs. Look for odd or missing voltages. Or just post your findings here and we can help you figure if there is a problem.

I guess the easiest thing to do would be to just replace the chip and see what happens.

In my experience, the newer Dunlop boards are very fragile with tiny traces that can be damaged easily. Carefully inspect the previous owners work to see if there are any signs of damage to the traces, pads, etc.

[Gypsey Eyes]

ok here's what I found but one question first -

should the voltage of all three prongs for the rate pot change? I notice that only two of the three seem to change - one stays the same.


now onto the TL604:

these are the readings
14 6.74
13 6.72
12 6.57
11 8.98
10 6.48
9 6.74
8 6.72
7 6.74
6 6.72
5 6.58
4 0
3 6.47
2 6.73
1 6.72

Some of the readings vary every testing by up to .10v

[Gypsey Eyes]

I've also checked the board all over for other damage. the only thing I keep wondering about is a mark where the orig 'modder' snipped the R28. Looks like his wire cutters cut into the board when he did it but upon magnification it looks only like surface damage to a grounded part.

[52 Bill]

Two of the three terminals of the rate pot are connected together, and it really doesn't matter, as we were only trying to see if the oscillator was working.

I'm guessing that you have counted the pins wrong on the TL064, as pin #4 should have 9vdc and pin#11 should be at 0. But in any case, none of the voltages seem to be abnormal.

You should test the board with your meter and not just your eyes. If your meter has a continuity setting use it to follow the traces around r28. Sometimes something might look ok, but in fact isn't. Use the ohm reading setting if you don't have a continuity setting.

Has the trim pot r27, been messed with? Older versions of this pedal would have the trim pot either soldered or painted to lock in the adjustment spot.

[Gypsey Eyes]

yes, I was looking at the chip from the backside rather than the front so all my numbers are reversed.

All the connections coming from r28 appear to test fine with my meter - I also see no sign of the trim pot being replaced or any setting mark on it.

[52 Bill]

What are the values of C11 and C12? Have these been changed? Apparently these are a couple of other parts that are often messed with.

When the pedal is screeching, is the screech a solid tone or does it change with the low freq. osc.?

[Gypsey Eyes]

both caps look original - c11 says 103; c12 is hard to see but it looks like 681.

I uploaded a video that shows what happens when I turn the oscillator pot and then the rate pot.

YouTube - bkplant's Channel

[52 Bill]

You know, this video thing might just catch on.
The sound is really cool, maybe someone would want it as a signal generator.

I sounds like too much regeneration. Again I assume that this video was shot with the original 24K value for R28 in circuit.

For lack of anything else to try, tweak the bias pot R27. Mark it so you know where it was originally set before you start turning it. You may be able to reduce the effect enough to kill the feedback.