Phase 90 LFO problem (I think)

[jdoolin]

I got two used re-issue Block Phase 90's cheap and figured it would be a good opportunity to try out some fun mods.  One of them is working fine while the other is not.

First, if this belongs in another forum, forgive me and move it to the appropriate place.  Second, I'm perfectly willing to provide all the information suggested by the DEBUGGING topic if that would be of assistance in this situation.

Now, the problem:  it actually does phase but with a very strange sweep pattern and not at all positions on the time knob.  If I understand what I've read, it seems like a problem with the LFO.  But maybe not.  I'm just barely understanding how the JFETs work as variable resistors, so maybe that could be it too.

Anyway, up to about 12:00 there is no noticeable phase at all.  Almost like it's stuck in one spot.  Then when you get *JUST* past 12:00 or so, it starts, but it's not an even sweep.  I understand the LFO produces a triangle wave.  This sounds more like a sawtooth, where it's very asymmetrical sounding.  Here is a clip:

http://theemptyset.net/jdoolin/Phase90.mp3

Here is the circuit:

http://vidal.hugo.perso.sfr.fr/Photos/guitare/MXR.gif

Here's what I've tested:

-Adjusting the trim pot.  I've tried several times with VERY fine adjustments, tediously trying the whole range and found no sweet spot
-Testing the trim pot with a multimeter - it checks out and the values are consistent with the working P90
-Testing the timing pot - again, it checks out
-Replacing U1 - I did this just taking a totally amateur pot-shot at fixing the problem.  If anything, it was good practice with removing ICs, but didn't fix my problem
-Replacing C7 - Again, just a barely educated guess, "Gee, I wonder if this would work" easy thing that did nothing.

Any ideas?  I'd be happy to provide any additional information (voltages at certain components, etc).

Thanks!

[R.G.]

Hmmm.
Did it have this problem before the mods?
What mods did you do?

This is one of those problems where I'd go get out the oscilloscope. Do you have a 'scope?

Since one works, the other doesn't, how do they compare?

[jdoolin]

Hmmm.
Did it have this problem before the mods?
What mods did you do?

This is one of those problems where I'd go get out the oscilloscope. Do you have a 'scope?

Since one works, the other doesn't, how do they compare?

Ooops, I totally neglected to mention that I haven't done any of the mods other than adding a switch for R28 in the operational P90.  Otherwise, they are exactly as I acquired them.

I don't have an oscilloscope, but I'm starting to think that it may be in my best interest to acquire one. 

Anyway, I just looked again and happened to notice something I hadn't before.  It *looks* like the JFETS may have been replaced or tampered with at some point.  I took the best photo I could manage of the solder side:

[GGBB]

The JFETs in a phase 90 have to be matched. If some have been replaced, it's possible they weren't all matched and that may be why it doesn't sound right.

[jdoolin]

Now I'm not sure about the JFETs being tampered with.  I happened to see another photo of the solder side of this particular P90 model and it has the same discoloration.  Now, unless this happened to somehow be the exact same pedal, I suppose they all look like this:

http://stinkfoot.se/archives/975

It even has the same 4 band resistor in R22.  So again, unless I somehow acquired this exact pedal through some mystery of the universe, this is normal.

[jdoolin]

So at this point would it be in my best interest to break out the multimeter and gather Voltage values for various component leads on the board?   And how does that work for the ones that will have varying values?  Just a min/max value?

[R.G.]

So at this point would it be in my best interest to break out the multimeter and gather Voltage values for various component leads on the board?   And how does that work for the ones that will have varying values?  Just a min/max value?

Just do the best you can for a first pass. It may be something obvious.

[jdoolin]

Thanks, I'll do it as soon as I can.  But it's daddy/daughter evening.  Might have to wait until tomorrow morning.

[jdoolin]

Here goes.  Digital multimeter, set to DC 20V, one probe connected to ground.  I triple, quadruple, sometimes quintuple checked these.  This is every transistor (x1), JFET (x4), IC (x2) and diode (x2) and they correspond to those in the circuit I linked to in the first post.

http://vidal.hugo.perso.sfr.fr/Photos/guitare/MXR.gif

Q1
E = 4.91
B = 4.3
C = 3.1

Q2
G = 1.40
S = 4.92
D = 4.82

Q3
G = 1.40
S = 4.92
D = 4.82

Q4
G = 1.4
S = 4.92
D = 4.82

Q5
G = 1.32
S = 4.92
D = 4.82

U1
p1 = 4.93
p2 = 4.93
p3 = 2.44
p4 = 0
p5 = 3.70 minimum/5.15 maximum, jumps back and forth between values within that range
p6 = 4.71
p7 = 8.24
p8 = 8.91

U2
p1 = 4.93
p2 = 4.93
p3 = 4.83
p4 = 8.91
p5 = 4.93
p6 = 4.93
p7 = 4.93
p8 = 4.93
p9 = 4.93
p10 = 4.83
p11 = 0
p12 = 4.82
p13 = 4.93
p14 = 4.93

D1
A = 0
K = 4.92

Z1
A = 0
K = 8.91

If I'm missing anything or more information would help, just ask.  Thanks to anyone who takes a look.

[jdoolin]

Bump, I suppose?  The only anomaly that my untrained eye sees is that 1.32 V on the one JFET where the others are 1.4.  Is that enough of a difference to cause the problem I'm hearing?

[mcknib]

Your gate sorry drain voltages are too low try adjusting them via the trimmer to about 2.2 to 2.4v.

Your U1 voltages look wrong especially pins 3 non inverted input and pin 7 which is an output on the LFO part.

I've been working on one with no phasing which is now phasing after changing U1 and U2 with a TL072 and TL074 you don't mention what you used for U1 if it's not pin compatible it may be causing your strange voltages.

With mine the owner had snipped out the 1N914 and not replaced it so I assumed he had fried the IC's the initial readings on the TL064 showed correct voltages on non inverted pins and 2.7 to 2.9V on the inverted inputs and the outputs telling me it was faulty.

Looking at your U1 voltages on pin 3 and 7 may suggest it's either duff or not pin compatible.

You can also check if you are getting fluctuating voltages on the pot lugs to see if the LFO is pulsing

My readings on U1 are:

1. 4.51
2. 4.52
3. 4.09
4. 0
5. 4.08 - 4.85
6. 4.28 - 4.25
7. 3.98 - 5.25 these eventually settle similar to pin's 5 and 6
8. 9.10

I am not saying it's at the sweetest spot but it phases throughout the speed pot's rotation.

[jdoolin]

Your gate voltages are too low try adjusting them via the trimmer to about 2.2 to 2.4v.

I just tried that and got no phasing at all.

Your U1 voltages look wrong especially pins 3 non inverted input and pin 7 which is an output on the LFO part.

Looking at your U1 voltages on pin 3 and 7 may suggest it's either duff or not compatible.

I had actually replaced U1 suspecting the original may have been the problem.  I replaced it with the same thing that came factory: TL062.  I ordered it from Small Bear.  But alas, the same problem existed before and after replacing U1.

You can also check if you are getting fluctuating voltages on the pot lugs to see if the LFO is pulsing

Which pot?

My readings on U1 are:

1. 4.51
2. 4.52
3. 4.09
4. 0
5. 4.08 - 4.85
6. 4.28 - 4.25
7. 3.98 - 5.25 these eventually settle similar to pin's 5 and 6
8. 9.10

Awesome, thanks for the reference.

[mcknib]

The speed pot lugs you should get fluctuating voltages there, I'll try and take Q readings and U2 readings for you with this one but as I said your U1 pins 3 and 7 make me think this may be what's causing your problems. Your U2 readings look ok to me.

Check your U1 socket does not have a pin bent upwards easy enough to do just make sure there's a pin sticking through each hole on the solder side you should be able to feel it with your finger and have a close look on the socket side if you see a wee thin silver sliver (bent socket pin.) up it's side you've found your problem.

[mcknib]

Measured U2 all pins read 4.52v except pin 4 - 9.11v power supply and pin 11 - 0v ground

Q1 2N4126
C. 4.52
B. 4.52
E. 2.74

Q2 to Q5 2N5952
All
G. 4.52
S. 4.52
D. 2.38 to 2.43

Apologies got my drain and gate mixed up earlier
I'm still thinking bent socket pin especially with pin 7 at 8.24v maybe bridging with pin 8 somewhere or possibly a dry solder joint on your U1 socket

[mcknib]

I thought it actually said you'd used a socket for U1 did you? if you didn't obviously ignore what I said re: socket pins hopefully you did because IC's are especially easy to damage with heat from soldering unless you're a solder guru! and if you did use a socket take voltage readings with U1 out of socket to see if it's any different if they're more like mine that'll tell you U1's probably duff

[jdoolin]

I thought it actually said you'd used a socket for U1 did you? if you didn't obviously ignore what I said re: socket pins hopefully you did because IC's are especially easy to damage with heat from soldering unless you're a solder guru! and if you did use a socket take voltage readings with U1 out of socket to see if it's any different if they're more like mine that'll tell you U1's probably duff

Hmmm... I didn't use a socket.  I soldered it directly and while I'm pretty handy with it, I'm certainly not a solder guru.  I could place an order for another IC and a socket and see if that works.

[mcknib]

Might be ok depends on how long you left the soldering iron on the pad and IC pins while dabbing the solder on 2 to 3 seconds and it should be ok.

Not that I'm a guru far from it, but I never solder IC's direct to the board I always use sockets and let the socket cool down before I put the IC in the other benefit being you can try different IC's, Op Amps etc to see what you like best. With the Phase 90 there's a few dual op amps you could use in U1 albeit they're more or less the same and wouldn't make much difference aside from perhaps in the input buffer stage.

If it were a distortion circuit then socketing IC's, transistors, diodes etc gives you the ability to experiment with it and of course with the research you'd do helps you to understand circuits better.

FYI

If you look at this muff circuit for example you can see loads of parts you could socket to try different values etc to get your preferred sounds
http://www.kitrae.net/music/Images_Secret_Music_Page/BIG%20MUFF%20CIRCUIT%20GUIDE.pdf

Mind you as long as you don't go too mad I've found myself socketing entire circuits before and you can end up in a bit of a mess.

RG's done a phaser here in his technology of series which I use often (sorry RG but in my book although I know you'll disagree you are indded a guru haha) http://www.geofex.com/article_folders/phasers/phase.html

There's also a good Phase 90 circuit analysis here + a few other common pedals
http://www.electrosmash.com/mxr-phase90

A note just in case you don't know but you'll have to flatten the pins on the IC by lying it on it's pin sides on a flat surface and gently pressing to make it fit easily into the socket and obviously be super careful when de-soldering your U1 IC the pads can come off fairly easily

Looking at a schematic U1a input buffer would be pins 2 and 3 as inputs and 1 as output and U1b would be the LFO part with pins 5 and 6 as inputs and 7 as the output and reading your voltages in comparison to mine it does look like an LFO problem but it could also be parts of the circuit around it causing the incorrect voltages.