Phase 90 problems

[Ell]

I'm using this layout http://www.aronnelson.com/gallery/main.php/v/Grooveyard/mxr_phase_90.gif.html

At the moment I do get a kind of swishing sound if I put the speed knob on full, but there's loads of overdrive and it sounds more like a distortion pedal than a Phase 90.
I checked the 2n4125 transistor on a multimeter and I got 127. How do I test the other transistors though? My multimeter has slots for PNP and NPN transistors with ECB inputs for each, but I don't know where to put the DSG lugs for these transistors? Apparently they should be matched and I recon that MAY be my problem?

I have used a 0.05uF capacitor instead of the 0.05pF capacitor that I'm sure is written as a mistake.
No other substitutions. I have added a switch to cut off the top line.

And I have put all the parts into the board exactly as shown in the layout. When I removed the 1N914 diode just now (while it was turned on) and tried out an LED in its place, there was no difference in sound whatsoever when I removed the 1n914 diode. Perhaps that whole side of the circuit is not working?

Thanks. I know it's a lot of effort but I seriously want this pedal to work, it's my Everest.

[R.G.]

Have you read "DEBUGGING: What to do when it doesn't work"?

[Myriad Society]

Ell, go to www.geofex.com and click "Search GEO" on the left hand side of the home page. Next type in "jfet matching" in the search box and do the search. This should yield a list of about 6 results. Click on "fetmatch.htm" about half way down the list.

This article will tell you everything you need to know about matching JFETs as well as how to build a device you can use with your multimeter that will allow you to test and match them for use in your Phase 90. While you are at it, I would bookmark Geofex if you haven't already as that site is a treasure trove of valuable information much like this forum. Hope this helps put you on the right track.

[Ell]

Okay, I tried the Jfet matcher but unfortunately I think I did something wrong and I got 3.93 for all four transistors. Then I tried it without a transistor plugged in and I still for 3.93! When I tested the battery with the multimeter I got about 9.15
For the Jfet matcher I used this layout http://www.aronnelson.com/gallery/main.php/v/Richard-Boop-RLBJR65/album81/album75/Fet_Matcher_1.gif.html

R.G. - here's the information from the Debugging thing. I really hope you can help me.

1.What does it do, not do, and sound like?  It has a modulative effect of some kind when the speed knob is turned to full, it's like a very fast low depth phaser. It also adds tonnes of volume and gain. It does not sound like it's doing full frequency sweeps though, as would be expected with a phaser. If I turn the variable resistor it acts kind of like a wah on one side, where if you increase it quickly it alters the frequencies like a wah does. If I turn it to the other side I get high pitched space ship screaming kind of sounds.

2.Name of the circuit = MXR Phase 90

3.Source of the circuit (URL of schematic or project) = http://www.aronnelson.com/gallery/main.php/v/Grooveyard/mxr_phase_90.gif.html?g2_imageViewsIndex=1

4.Any modifications to the circuit? Yes, I've added a switch after R21 which breaks that the connection to IC3. I also have a switch for C3, with a 50nf cap and a 50pf cap switchable, just to make sure that it is meant to be a 50nf cap.

5.Any parts substitutions? Only C3, 0.05uF cap used instead of 0.05pF cap, which I'm assuming is an error.

6.Positive ground to negative ground conversion? What does this mean?

7.Turn your meter on, set it to the 10V or 20V scale. Remove the battery from the battery clip. Probe the battery terminals with the meter leads before putting it in the clip. What is the out of circuit battery voltage? => 9.16

Now insert the battery into the clip. If your effect is wired so that a plug must be in the input or output jack to turn the battery power on, insert one end of a cord into that jack. Connect the negative/black meter lead to signal ground by clipping the negative/black lead to the outer sleeve of the input or output jack, whichever does not have a plug in it. With the negative lead on signal ground, measure the following:
Voltage at the circuit board end of the red battery lead = 9.12

Voltage at the circuit board end of the black battery lead = 0.00

Q1
D = 3.64
S = 3.70
G = 0.90

Q2
D = 3.67
S = 3.70
G = 0.90

Q3
D = 3.68
S = 3.70
G = 0.90

Q4
D = 3.62
S = 3.70
G = 0.89

Q5
C = 1.95
B = 3.11
E = 3.70

IC1
P1 = 3.70
P2 = 3.70
P3 = 2.50
P4 = 0.00
P5 = 3.27 - 3.33 (flickers)
P6 = 3.99
P7 = Sweeps between 3.84 and 4.10. I'm guessing that's the phase speed.
P8 (top right) = 9.05

IC2
P1 = 3.70
P2 = 3.70
P3 = 3.64
P4 = 0.00
P5 (bottom right) = 3.67
P6 = 3.70
P7 = 3.70
P8 = 9.03

IC3
P1 = 3.70
P2 = 4.19
P3 = 3.68
P4 = 0.00
P5 = 3.60
P6 = 4.02
P7 = 3.67
P8 = 9.03

D1
A = 0.00
K = 9.03

Z1
A = 0.00
K = 3.70

C4
+ = 0.00
- = 0.00

C2
+ = 3.70
- = 0.00

[R.G.]

1.What does it do, not do, and sound like?  It has a modulative effect of some kind when the speed knob is turned to full, it's like a very fast low depth phaser. It also adds tonnes of volume and gain. It does not sound like it's doing full frequency sweeps though, as would be expected with a phaser. If I turn the variable resistor it acts kind of like a wah on one side, where if you increase it quickly it alters the frequencies like a wah does. If I turn it to the other side I get high pitched space ship screaming kind of sounds.

4.Any modifications to the circuit? Yes, I've added a switch after R21 which breaks that the connection to IC3. I also have a switch for C3, with a 50nf cap and a 50pf cap switchable, just to make sure that it is meant to be a 50nf cap.

It's 50nF, like C7/8/9/10. Solder that one in and get rid of the switch.

5.Any parts substitutions? Only C3, 0.05uF cap used instead of 0.05pF cap, which I'm assuming is an error.

You're right, it is a typo. 0.05pF is the size of the capacitance between earth and Saturn.

6.Positive ground to negative ground conversion? What does this mean?

There is a common mod to some pedals which have an inconvenient grounding arrangement. The P90 does not. Ignore it.

Q1
D = 3.64
S = 3.70
G = 0.90

Q2
D = 3.67
S = 3.70
G = 0.90

Q3
D = 3.68
S = 3.70
G = 0.90

Q4
D = 3.62
S = 3.70
G = 0.89

JFETs are OK, excepting perhaps that the trimmer pot is too far toward ground. Did you use specifically 2N5952s, not a substitute?

Q5
C = 1.95
B = 3.11
E = 3.70

Looks OK.

IC1
P1 = 3.70
P2 = 3.70
P3 = 2.50
P4 = 0.00
P5 = 3.27 - 3.33 (flickers)
P6 = 3.99
P7 = Sweeps between 3.84 and 4.10. I'm guessing that's the phase speed.
P8 (top right) = 9.05

IC2
P1 = 3.70
P2 = 3.70
P3 = 3.64
P4 = 0.00
P5 (bottom right) = 3.67
P6 = 3.70
P7 = 3.70
P8 = 9.03

IC3
P1 = 3.70
P2 = 4.19
P3 = 3.68
P4 = 0.00
P5 = 3.60
P6 = 4.02
P7 = 3.67
P8 = 9.03

D1
A = 0.00
K = 9.03

Z1
A = 0.00
K = 3.70

Again, looks OK.

C4
+ = 0.00
- = 0.00

This is a problem. The voltage on this cap should rise and fall with the LFO, affected by the speed knob.

C2
+ = 3.70
- = 0.00

Probably OK.

I think you may have C4 in backwards; the information on the layout is very poor for how to orient it, and it is a polarized cap. C2 and C4 should have their negative (-) ends both on that single strip. The + ends should face away from each other.

The "tons of volume and gain" makes me think there is a soldering problem in one of the feedback resistors, R 17/20/24/27, or that the connection between pin 1 and 2 of IC 1 may be open. Check that out and let us know what you find.

[Ell]

Okay. I checked all the resistors that you said and I found that R20 was loose, so I soldered that back. Still a problem though. C4 was already the right way round, so unfortunately that's not the problem. I took it out and resoldered it anyway and now I am getting a reading which alternates with the speed, like you said it should. Strange.
I tested the 1n914 and it looks like it had blown, so I've replaced that.

Yes, they are all 2n5952's.
The connection between pins 1 and 2 of IC1 looks alright.

All of the solder joints that you've mentioned seem pretty strong as well, in that I can't move the component at all, and I can't move the leg on the copper side.


Do you think it would be worth my time buying this set of 4 matched transistors for this pedal? http://banzaieffects.de/2N5952-matched-transistors-pr-16283.html

[Ell]

NEW PROBLEM.
Now the distortion is gone, as is the phase. All I get is a slightly quieter version of the clean sound.

Here are the new ratings

Battery = 8.95

Q1
D = 3.70
S = 3.70
G = 1.11 - 1.30

Q2
D = 3.70
S = 3.70
G = 1.11 - 1.30

Q3
D = 3.70
S = 3.70
G = 1.11 - 1.30

Q4
D = 3.70
S = 3.70
G = 1.11 - 1.30

Q5
C = 1.91
B = 3.11
E = 3.70

IC1
1 = 3.70
2 = 3.70
3 = 2.51
4 = 0.00
5 = 3.88 - 4.30
6 = 3.20 - 4.83
7 = 1.28 - 8.32 (only when speed control is set to minimum, when set to maximum I get about 3.80 - 4.10)
8 = 8.95

IC2
1 = 3.70
2 = 3.70
3 = 3.70
4 = 0.00
5 = 3.70
6 = 3.70
7 = 3.70
8 = 8.95

IC3
1 = 3.70
2 = 3.99 - 4.15
3 = 3.70
4 = 0.00
5 = 3.70
6 = 3.99 - 4.19
7 = 3.70
8 = 8.94

D1
A = 0.00
K = 8.94

Z1
A = 0.00
K = 3.70

C4
+ = 3.10 - 4.80
- = 0.00

C2
+ = 3.70
- = 0.00


So, C4 is now giving a reading, but for some reason the phasing has stopped altogether and so has the distortion.

[R O Tiree]

Silly question, perhaps, but have you tried tweaking the trim pot?

[Ell]

Silly question, perhaps, but have you tried tweaking the trim pot?

Yeah. Is there a specific point on it where it HAS to be for the pedal to operate though? I was assuming I could just tweak it to taste after the pedal is working, like with other trimpot pedals like the Orange Squeezer.

[frequencycentral]

There is an 'optimal point' for the trimpot, where the effect will be most pronounced. You can tweak it either side of this 'optimal point' and still get an effect, but tweak too far from it in either direction and the effect will be lost. Exactly where on the trimpot is this 'optimal point' is entirely dependent on the Vgs of your matched FETs. If your FETs are not matched, each FET will have a slightly different 'optimal point'.

[gena_p1]

what is Vgs of your FETs?

"G = 1.11 - 1.30" - i think it's too low, use trim pot

[Ell]

http://www.aronnelson.com/gallery/main.php/v/Richard-Boop-RLBJR65/album81/album75/Fet_Matcher_1.gif.html

Is this layout perfect?

I just get the same reading for every transistor, 3.84. I even get that reading when there isn't a transistor in the circuit? Which setting should I have my multimeter set to? Am I doing something very wrong?! Please answer this as soon as possible as I have to order some parts off of Banzai very soon and if I can get this circuit working then I'll buy 20 or so 5952 transistors and match them myself. Otherwise I'll have to buy a set of four prematched ones for the more expensive price which is over 5 euros. Thanks!

[R.G.]

Please accept my apologies if this sounds like I am being critical of you. It is not intended to be.

I think you have some problems with soldering reliably. As such, it is not surprising that you would have problems with making a JFET matcher as well as the P90. I also think that you may be having some problems with Vero board construction.

I personally do not like Vero board. It is OK when used by someone already skilled in construction, but it is both confusing to beginners and easy to make mistakes on, both of which signal to me that using veroboard is not something which should be promoted for beginners. Yet beginners seem to pick up Vero board layouts and have many problems. This is very much a personal opinion, but it is one which I held strongly enough to avoid mentioning stripboard for years when I was trying to get more people involved in making effects.

[Ell]

Please accept my apologies if this sounds like I am being critical of you. It is not intended to be.

I think you have some problems with soldering reliably. As such, it is not surprising that you would have problems with making a JFET matcher as well as the P90. I also think that you may be having some problems with Vero board construction.

I personally do not like Vero board. It is OK when used by someone already skilled in construction, but it is both confusing to beginners and easy to make mistakes on, both of which signal to me that using veroboard is not something which should be promoted for beginners. Yet beginners seem to pick up Vero board layouts and have many problems. This is very much a personal opinion, but it is one which I held strongly enough to avoid mentioning stripboard for years when I was trying to get more people involved in making effects.

Yeah, there probably is an unwanted link somewhere, or something is not quite connected, or a cut isn't quite cut - but it's not like this is my first time with veroboard. Unfortunately it's my first big problem, but I've had a few (about 10) successes with it. I just can't afford to buy the circuit boards off of tonepad or make my own. Also, I have a lot of layouts designed for veroboard so that's kind of convenient.


EDIT:

what is Vgs of your FETs?

"G = 1.11 - 1.30" - i think it's too low, use trim pot

Right, the Jfet matcher is working.

The Vgs of my FETs are -1.75, -1.35, -1.64 and -1.40.

These values are perhaps unreliable to ratings of the same 5952's using different test equipment, because I used 12K resistors instead of 10K's, just because I happened to have spare 12K's lying around. It just means that I can compare them to each other, and see how close they are to each other.