PLEASE HELP - JMK PCBs Super Phaser (no signal coming through)

[jfrabat]

The next thing to test is to see if the LFO is oscillating. This is not hard to do but it is boring. Set both the rate and depth controls to minimum (fully counter clockwise ) and set the bias trim-pot to mid range. Apply power and measure the voltage on pin 7 of IC1. You are not actually worrying about an absolute voltage measurement, you are merely going to see if the voltage varies over the period of a few minutes. We are interested in the minimum and maximum voltages and how long it takes to go from one extreme to the other and back again.

I started with the Speed and Width controls fully counter-clockwise; no oscillation...  Tried it fully clockwise...  Same!  Depending on pot position, I get from 1.388 to 1.413V, but no oscillation!  Mind you, there was nothing connected to the input, but I tried touching the cable to give it some signal, and had NO effect on voltage.

[Slowpoke101]

OK. LFO appears to be not working. Confirm that the rate pot is a 500K type (it does not yet matter what taper it is ).
Check the circuit board traces going to the rate pot. Repair any broken traces.
Generally check any of the components connected to IC1 pins 5, 6 and 7.
Pin 7 should vary from about 1.45V to 8.2V

[jfrabat]

OK. LFO appears to be not working. Confirm that the rate pot is a 500K type (it does not yet matter what taper it is ).

Pots are correct

Check the circuit board traces going to the rate pot. Repair any broken traces.
Generally check any of the components connected to IC1 pins 5, 6 and 7.
Pin 7 should vary from about 1.45V to 8.2V

What is the best way to check the traces?  Visual inspection?  I will review visually and with the meter for now, but if there is a specific way, please let me know.

[antonis]

Before tracing, remove all flux residuals with a medium softnes brush (an old toothbrush is the ultimate tool..)

[jfrabat]

Did the tracing, and I think I found the problem...  I found no continuity from IC1 5 to R21 or R22.  Digging deeper, I DID have continuity there from the solder point, but not from the leg of the IC.  I did a test, taking out the IC, and I get no continuity from the top of the socket to the solder joint.  My best guess is that, since I used one of those cheap sockets I used to use at the time, the leg bent and did not go into the hole.  Although, taking a picture, looks like the contact is bent (I may have bent it trying to look for continuity, though).



So, anyone got a good way to remove a socket from a board without destroying a board?

[antonis]

A good desoldering pump and care with patience.. 

[duck_arse]

try putting the IC back in, and prodding a cut-off lead dwon by the pin/socket wipe. see if you get readings then - you don't care about the socket now, so it can't really hurt.

that's pretty good detecting, by the way.

What is the best way to check the traces?

Luke, use the meter!

[jfrabat]

try putting the IC back in, and prodding a cut-off lead dwon by the pin/socket wipe. see if you get readings then - you don't care about the socket now, so it can't really hurt.

I tried something similar; I used a lead in the socket and measured continuity from the top to the solder joint.  Nothing.

I think this shows that the leg did not go through the hole...

[jfrabat]

I was thinking of using my micro torch to try to get the socket out.  I use it to recover parts from old electronics and to get pots back from ruined boards.  It roasts a bit the board, though, as you can see in this old iRobot board from which I got some transistors:



Traces look intact, though.  Obviously, on the board above I was not being careful not to damage the board; I can try to fan the flame a bit to avoid damage, but I am sure some scorching will be unavoidable.  Also, looking at the board (the phaser), I do not see any traces under the IC1, so that is also a positive (see pic in post above), just the ground, so any damage would not be critical.

Should I give it a go?

[antonis]

OK.. try to bend IC pin 5 outward (90^o or more), put IC in place and solder a thin wire between bended pin and respective pad on the other side of PCB..

[jfrabat]

OK.. try to bend IC pin 5 outward (90^o or more), put IC in place and solder a thin wire between bended pin and respective pad on the other side of PCB..

You mean the socket?  Impossible, I can't reach it...  I can't even SEE it!

Decided to go the micro torch route...  I think it went well.  Here is the board after taking out the socket:



An here is the culprit:



"Naw their's yar prablem right there!" (use think hillbilly accent when reading)

I will now clean the holes and use a good bracket for the IC...

[antonis]

You mean the socket? 

No I meant the IC ..

But all god now..

[jfrabat]

Fix one thing, and 3 more errors pop their ugly head... UUURGH!

OK, replaced the socket, but this is what I get:

Taking voltage measurements now...

[jfrabat]

Here is the new socket:



And this is how the board looks on the back:



Here are the new voltages:

Q1:
G: 0.616~0.705V
S: 0.594~0.715V
D: 4.32V

Q2:
G: 4.12V
S: 4.11V
D: 0.577~0.715V

Q3:
G: 4.11V
S: 4.12V
D: 0.575~0.715V

Q4:
G: 4.11V
S: 4.13V
D: 0.574~0.714V

Q5:
E: 2.30~1.32V
B: 3.58~3.64V
C: 4.12V

IC1
1: 4.27~4.92V
2: 4.25~4.9V
3: 0V
4: 0V
5: 4.08~4.26V
6: 3.27~3.90V
7: 4.27~4.95V
8: 8.65V

IC2
1: 3.98~3.4V
2: 4.11V
3: 4.11V
4: 8.66V
5: 4.10V
6: 4.12V
7: 4.09~4.56V
8: 4.08~3.67V
9: 4.03~4.10V
10: 4.11V
11: 0V
12: 4.24~4.33V
13: 4.42V
14: 4.25~4.73V

[Slowpoke101]

This little phaser is taking a lot more to get going than what the OCD pedal did...But you are making good progress.
The bent pin on the IC socket was interesting but you got it sorted.
Now the voltages;

IC1 pin 3 is shown as 0V. Measure it again as it should be about 4.2V
Q1 Gate and Source are both shown as close to 0.6V. Check this again as Source should be a similar voltage as Drain.
Please test the voltages on all the FETs as something is a bit odd.

One other thing to make sure of is that the bias trim-pot is not shorting to Q1 gate.

[jfrabat]

IC1 pin 3 is shown as 0V. Measure it again as it should be about 4.2V

This one struck me as odd too...  But it is 0V.  Tripple checked.  I am getting 4.11 on one side of R3, but 0V on the other side.  Time to replace R3?

Q1 Gate and Source are both shown as close to 0.6V. Check this again as Source should be a similar voltage as Drain.
Please test the voltages on all the FETs as something is a bit odd.

OK, remeasured voltages:

Q1:
G: 3.12V~3.24V
S: 4.11V
D: 4.29V

Q2:
G: 0.580~0.711V
S: 4.11V
D: 4.12V

Q3:
G: 0.580~0.714V
S: 4.13V
D: 4.14V

Q4:
G: 0.576~0.712V
S: 4.13V
D: 4.09V

Sorry about the previous misreadings...

One other thing to make sure of is that the bias trim-pot is not shorting to Q1 gate.

I checked continuity, and there is no short.

[Slowpoke101]

With IC1 pin 3 do a continuity test between pin 3 and ground (pin 4 ). If there is no continuity then replace the 470K resistor. I think that this is not really a problem as pin 1 shows the correct voltage. Besides, if there was no voltage on pin 3 there would be no signal passing through the effect at all.

Of the most interest are the voltages on the FETs. Q1 now looks better on the Source but now Gate is very odd. All the FETs should have similar voltages to each other. Time to confirm that they are connected to each other correctly. Set your meter for continuity test again (remember that no power is applied when doing a continuity or resistance tests ). Place one test lead onto the leg of R17 that is closest to Q4 and apply the other test lead to the Gate legs of each FET. There must be continuity between between all the Gates and R17. Also make certain that the Source (middle pin ) of each FET is connected to each other and the cathode of D1 (zener ). If there are any disconnects (breaks ), repair them.

The next test is to apply power and measure the voltage on pin 1 of the Width pot. Adjust the Bias trim-pot so that it reads about 3.4V DC.

Let us know how you go.

[jfrabat]

With IC1 pin 3 do a continuity test between pin 3 and ground (pin 4 ). If there is no continuity then replace the 470K resistor. I think that this is not really a problem as pin 1 shows the correct voltage. Besides, if there was no voltage on pin 3 there would be no signal passing through the effect at all.

There is continuity between IC1 3 & 4.

Place one test lead onto the leg of R17 that is closest to Q4 and apply the other test lead to the Gate legs of each FET. There must be continuity between between all the Gates and R17.

Good connectivity on all FETs

Also make certain that the Source (middle pin ) of each FET is connected to each other and the cathode of D1 (zener ). If there are any disconnects (breaks ), repair them.

Good connectivity on all 4 FET Sources and D1

The next test is to apply power and measure the voltage on pin 1 of the Width pot. Adjust the Bias trim-pot so that it reads about 3.4V DC.

Width pin 1 3.50V at almost 12 o clock position (11:30?).  3.52V with Bias at 12 o clock position (midway of trimpot).  With trimpot in this last position:


Q1:
G: 0.557V
S: 4.13V
D: 4.22V

Q2:
G: 0.558V
S: 4.12V
D: 4.13V

Q3:
G: 0.557V
S: 4.13V
D: 4.12V

Q4:
G: 0.557V
S: 4.13V
D: 4.12V

Q1 D still looks odd...

[Slowpoke101]

There is continuity between IC1 3 & 4.

OK. This is now a problem. Remove IC1 and test for continuity between pins 3 and 4 of the IC socket. If the short has disappeared replace IC1 and retest. If the short is still there, find out what is causing it.

As suggested by antonis, clean up as much of the flux residue as possible - it can hide problems such as fine slivers of solder shorting things out.

The voltages on the FETs are now much closer to each other. Q1 appears OK. But the Gate voltages are too low. Remove power and set your meter to resistance test and measure the resistance between any Gate and ground. Set the width pot to maximum (fully clockwise ) when doing this test. Also test the width pot and make sure that it is OK (1M and no shorts to ground on its pins ).

Good luck.

[jfrabat]

Remove IC1 and test for continuity between pins 3 and 4 of the IC socket. If the short has disappeared replace IC1 and retest. If the short is still there, find out what is causing it.

I do have continuity in the socket.  I tried looking for the reason, but cannot find it.  Eyes are tired, so will continue tomorrow...

Remove power and set your meter to resistance test and measure the resistance between any Gate and ground. Set the width pot to maximum (fully clockwise ) when doing this test.

Gate to GND is 1.9 Ohms (if I turn the pot the other way, it is 677K Ohms).

Also test the width pot and make sure that it is OK (1M and no shorts to ground on its pins ).

Pot measures 0.764M, and I DO have continuity from pin 3 to GND

Anyway, tomorrow I will clean it again and try to look for the short...  I am inclined to think it is somewhere behind the socket, as I did not have that short before changing the socket (I had actual voltage in pin 3 of IC1).  Maybe tomorrow I will desolder it again and see...