vintage Phase 90 not phasing...

[HTH]

Hi all,

I've got a script logo Phase 90 that is passing singal in both bypass and effect modes, but there is no phasing going on at all.

The voltages I've taken seem to suggest a problem with IC's 3, 4 and 5 which are the phase stages according to this schematic... http://www.generalguitargadgets.com/pdf/ggg_p90_sc.pdf

I'm thinking that the IC's (LM741) themselves are unlikely to be the cause, and similarly having all of the FETs (2N5952)being bad is unlikely too. So I'm considering that it's likely the resistors and caps around those three opamps that are the culprits. However, rather than blindly replacing stuff on a vintage pedal, I'd like to troubleshoot this and understand the circuit a bit more so analysis solves the issue.

Any help on this matter is much appreciated. Here's the voltages...

Vs: 9.1v
voltage at zener: 4.1v

FET voltages: gate 1.5v-1.8v, source 4.1v, drain 4.1v
BJT voltages: emitter 4.1v, base 3.9v, collector 1v


pin1 pin2 pin3 pin4 pin5 pin6 pin7 pin8
IC1 0 4.1 3.9 0 0 4.1 9.1 0.1
IC2 0 4.1 4.1 0 0 4.2 9.1 0
IC3 0 6.2 4.1 0 0 8.3 9.1 0
IC4 0 4.9 4.1 0 0 1.8 9.1 0
IC5 0 4.1 4.1 0 0 6.9 9.1 0
IC6 0 3.5-5 4.1 0 0 1.3-8 9.1 0.2

IC6 flips between the two voltages listed on pins 2 and 6.

Any takers?

[jpfamps]

High Ian!

In my experience old 741 are not the most reliable device and can need changing, especially if the pedal has experienced a polarity reversal.

I would concur with your conclusion on the suspect op-amps.

IC6 is the oscillator and seems to be working.

I often remove the offending chips and install IC sockets if you want to check they are OK.

Frank

[Mark Hammer]

Have you attempted to adjust the bias trimpot, to see if that yields any change in phasing effect audibility?

[52 Bill]

Ditto to what Mark said. Always a good place to start.

Phase shifters work by mixing a straight signal with a modulated shifted or delayed signal. If it passes a signal, then you know that input IC1 and the output transistor stage built around Q5 are working.

While it is unlikely that all of the shift ICs have died, if one of them dies then the shift signal dies at that point. The quick test would be to lift one end of R27 which carries the straight signal to the output section. If the unit no longer passes signal, then you know that the problem is in the series string of ICs 2 thru 5.

If the unit still passes signal then the problem is with the modulation part of the circuit consisting of the LFO IC6 and the FETs and the FET bias voltage.

[Mark Hammer]

Ditto to what Mark said. Always a good place to start.

Phase shifters work by mixing a straight signal with a modulated shifted or delayed signal. If it passes a signal, then you know that input IC1 and the output transistor stage built around Q5 are working.

While it is unlikely that all of the shift ICs have died, if one of them dies then the shift signal dies at that point. The quick test would be to lift one end of R27 which carries the straight signal to the output section. If the unit no longer passes signal, then you know that the problem is in the series string of ICs 2 thru 5.

If the unit still passes signal then the problem is with the modulation part of the circuit consisting of the LFO IC6 and the FETs and the FET bias voltage.

Yes and no. The output signal will consist of dry and wet. The dry only passes through one op-amp stage, and then to a mixer/output stage built around a single transistor. What this means is that the wet signal could be held up by one of the other op-amp-based phase-shift stages being fried, OR as you suggest, it could pass through all those stages safely, but nothing is modulating that phase shift. Those two possibilities will not sound particularly different from each other.

If you have a meter that can measure small AC voltages, then the easiest test of which of these two is closer to the true state of affairs would be to measure the AC output of the last op-amp stage (while providing an audio input signal to the pedal, naturally). If that stage provides a (non-modulated) signal, then clearly all previous stages are working just fine, and the problem is with the modulation. That, in turn, could be a mis-set bias, or a problem with the op-amp LFO stage.

If it should happen that there is no audio output from the last op-amp stage, then the challenge becomes one of figuring out what point in the circuit that signal ends its journey. That, in turn, should ideally involve working your way backwards from the last stage, a stage at a time.

[52 Bill]

What this means is that the wet signal could be held up by one of the other op-amp-based phase-shift stages being fried, OR as you suggest, it could pass through all those stages safely, but nothing is modulating that phase shift. Those two possibilities will not sound particularly different from each other.

Now I'm confused here. What two possibilities will not sound different?

If the mix resistor R27 is pulled, we should only be listening to the wet signal, correct? If there is no sound, then the signal is dying in the phase shift line. If there is sound then the string of phase shift stages is working, but not being modulated by the LFO and related parts, correct?

[Mark Hammer]

That's what I get for reading forums in small print on my tablet! My bad. I missed that little, but very important, detail.

You are correct. LIfting the dry path such that there is ONLY wet signal will most assuredly identify the problem as being somewhere within the 2nd op-amp stage (since the first is required to split the signal to dry and wet paths) to 5th stage, OR the LFO.

If the LFO is fried for some reason, but the remainder of the wet path intact, then tweaking the trimpot back and forth will yield some rudimentary phasing. Of course if the dry path is still lifted, you'll only get vibrato, rather than phasing. Unfortunately, where phasing can be audible at very slow rates (of the sort one would produce while tweaking the trimpot), vibrato generally requires much faster sweep rates to be audible, so it will be difficult to determine that the phase-shift stages are all fine, and the problem relegated to the LFO by tweaking the trimpot without the dry signal added in.

Kinda makes one want a scope, eh?

[cjlectronics]

A quick check of C7 (15uf) may fix the LFO if it is not oscillating. On more than one repair of a Phase 90, I have found C7 (tantalum cap) to be bad. Try measuring across C7 with an ohmmeter. If it is shorted to GND... BOOM, there is your problem... (or at least one of the problems??)

Oh, and check if it is in backward... that'll make it not oscillate as well. (I've seen this before, too)

[HTH]

Thanks for the input everyone - much appreciated. I have tried tweaking the trimpot, but no phasing at any setting.

I'll try lifting the mix resistor for the dry signal and see where that takes me.

Also got some LM741s coming (just in case).

Anyone got ideas about the funky voltages on IC's 3, 4 and 5? (pins 2 and 6).

[52 Bill]

Anyone got ideas about the funky voltages on IC's 3, 4 and 5? (pins 2 and 6).

Yes they are odd, but double check them to see if they are misreadings or real. I usually only measure the voltages on the outputs (pin 6) to see if they are higher than the expected bias voltage.

[HTH]

Thanks for the input everyone - much appreciated. I have tried tweaking the trimpot, but no phasing at any setting.

I'll try lifting the mix resistor for the dry signal and see where that takes me.

Also got some LM741s coming (just in case).

Anyone got ideas about the funky voltages on IC's 3, 4 and 5? (pins 2 and 6).

According to the voltages on this page... http://www.generalguitargadgets.com/...0_instruct.pdf the output on pin 6 should be around 2v with a bias voltage around 5v.

I'm getting 4.1v bias on pin 2 and about the same on pin6, except for IC3 where the output on pin 6 is 8.3v, IC4 has the output at 1.8v and IC5 has the output at 6.9v. Also the bias on IC3 is 6.9v.

I'm really loathed to have to pull ICs on this one - could leaky caps cause the voltage weirdness, or do you think bad IC's themselves are causing the weird voltages?

[52 Bill]

I'm getting 4.1v bias on pin 2 and about the same on pin6, except for IC3 where the output on pin 6 is 8.3v, IC4 has the output at 1.8v and IC5 has the output at 6.9v. Also the bias on IC3 is 6.9v.

I'm really loathed to have to pull ICs on this one - could leaky caps cause the voltage weirdness, or do you think bad IC's themselves are causing the weird voltages?

Where you have 7 volts on the outputs is a red flag. If there is 7 volts on an input pin and 7 volts on the output the output could just be following the input voltage. If there is 7 volts on the output and lower voltages on the inputs, then that could be an indication of a shorted chip.

[HTH]

Just a bit feedback for future reference and also to thank everyone who gave suggestions and help (much appreciated btw).

In the end one of the phase stage chips was bad, got some old stock ones off eBay to match the vintage of the pedal and its sounding great again.

I'm glad I measured everything and tried to understand what was going on (with a good bit of help from you guys), always great to take something away from a repair job and learn new tricks.

[Mark Hammer]

Thanks for the positive feedback. Always nice to find out things worked out in the end and that people came away from a consult a little more self-sufficient.