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

[jfrabat]

I had this board laying around for some years now.  It was populated, but I never could get it to work.  Decided to give it a go once again.  I checked all the component values, and they are right (well, I will admit I have not finished checking the resistors), but still, no signal coming through.  PLEASE HELP!

This is the build doc for that.  But for simplicity, here's the Schematic and the PCB parts list:





Here is my board:





Measurements (all pots at ~50%)

Battery: 9.1V

Q1:
G: 1.109V
S: 1.185V
D: 1.184V

Q2:
G: 1.111V
S: 1.182V
D: 1.182V

Q3:
G: 1.113V
S: 1.181V
D: 1.180V

Q4:
G: 1.124V
S: 1.180V
D: 1.180V

Q5:
E: 1.179V
B: 1.043V
C: 3.6mV

IC1
1: 1.368V
2: 1.368V
3: 1.129V
4: 0V
5: 1.448V
6: 1.448V
7: 2.36V
8: 8.93V

IC2
1: 1.489V
2: 1.427V
3: 1.176V
4: 8.91V
5: 1.175V
6: 1.494V
7: 1.502V
8: 1.486V
9: 1.494V
10: 1.172V
11: 0V
12: 1.171V
13: 1.488V
14: 1.489V

Speed Pot:
2.35V
1.156V
1.156V

width pot
1.139V
1.137V
0V

Bias Trimpot
0V
0V
1.169V

[antonis]

At afirst glance, issue is located on Q5..

[Slowpoke101]

Your Vref rail is very low. It should be above 4.5V not 1.169V
Check that the Bias trim-pot is actually a 250K unit - measure it, do not rely on its markings.
D1 is supposed to be a 5.1V device. Make certain that it is and is working.
C2 may be leaky so check it too.

A silly question would be - did you match the JFETs? This doesn't matter at the moment but could be an issue when the no signal at all issue is resolved.

[jfrabat]

At a first glance, issue is located on Q5..

Remeasured to avoid human error (prior to changing trim pot):

Q5:
E: 1.193V
B: 1.052V
C: 0.286V

Your Vref rail is very low. It should be above 4.5V not 1.169V
Check that the Bias trim-pot is actually a 250K unit - measure it, do not rely on its markings.

Tried to measure the previous one, but could not get a proper reading.  Replaced it with a measured on (measured 247, so we are good).

D1 is supposed to be a 5.1V device. Make certain that it is and is working.

D1 Voltages:
C: 1.189V
A: 0V

Should the cathode be 5.1V (or there abouts)?  Seems to me (in my limited understanding of electronics!) that this diode controls the voltage for the Vf Rail, right?

C2 may be leaky so check it too.

C2 Voltages (prior to trimpot change):
1.189V at +
0.1mV at -

Is this how you check for leaks?

A silly question would be - did you match the JFETs? This doesn't matter at the moment but could be an issue when the no signal at all issue is resolved.

Yes, they are balanced.  I bought them balanced from JMK...  1.670, 1.671, 1.671, 1.672 measured.

After changing the trimpot, now I hear static (like when you hold the guitar cable in your hand), and I can hear the strings being strummed (faintly) now.  But it is still not working.

I checked the zenner, and I can get from 2.7mV to 1.166V (on +, still 0V at -) depending on the trimpot setting.

[Slowpoke101]

Trim-pot is 247 ?  Is that 247K ?

To test C2 - remove it and see if Vref comes up. If it does then C2 is faulty.
Remove D1 (lift one leg of the diode free from the board ) and see what voltage reading you get on Vref.

Let us know what you find.

[jfrabat]

Trim-pot is 247 ?  Is that 247K ?

Yes, sorry.  Actually measured 0.247M (same thing, but to be exact)

To test C2 - remove it and see if Vref comes up. If it does then C2 is faulty.
Remove D1 (lift one leg of the diode free from the board ) and see what voltage reading you get on Vref.

Let us know what you find.

\

Will try it and report back

[jfrabat]

To test C2 - remove it and see if Vref comes up. If it does then C2 is faulty.

Removed C2.  Vref is still 1.180V (measured at the zenner)

Remove D1 (lift one leg of the diode free from the board ) and see what voltage reading you get on Vref.

Removed Zenner (and C2 above).  Vref is now 7.21V.  Resoldered C2, 7.16V.

Seems like the Zenner is the issue.  Replace it?  That is a 1N4733, so it is 5.1V...  I got more if I need to replace it, but I checked it was correct (you can just make it out in the pic below - took the pic after lifting one leg off the board).

[Slowpoke101]

Yes, replace the zener.
Also make certain that R2 is actually 10K.

[jfrabat]

Also make certain that R2 is actually 10K.

What about 3PO?  (sorry, could not resist!)

[duck_arse]

a few observations - doc shows CLR as 4k7 - you have 470R
doc shows R26 as 150k - you have .... I dunno, looks like an orange band on one end
doc shows R17 as 3M9 - you have 3k9.

[jfrabat]

Yes, replace the zener.
Also make certain that R2 is actually 10K (Brown, Black, Orange)

a few observations - doc shows CLR as 4k7 - you have 470R
doc shows R26 as 150k - you have .... I dunno, looks like an orange band on one end
doc shows R17 as 3M9 - you have 3k9.

OK, replaced Zener.  Now I got 4.16V at Vref.  Still not up to 5.1V, though.

Checked R2; it is actually 10K.   

So I checked ALL R color codes...
R16 was suppossed to be 4.6K, and I have 4.7K.  That I will leave like that (not even sure I got 4.6K).
R17 is 3.9K, and should be 3.9M
CLR says 4.7K, and I have 470, but that is what I need for my application (I am using a ring LED, and it requires 470 Ohms).
The rest were OK.
R26 is OK (Brown, Green, Yellow)

Anyway, I will get to work on R17...

[duck_arse]

R26 is OK (Brown, Green, Yellow)

use the meter Luke, measure it. I say this because the last colour band doesn't make standard resistor sense. that resistor also doesn't match any of the other types on board. suspicious.

[jfrabat]

R26 is OK (Brown, Green, Yellow)

use the meter Luke, measure it. I say this because the last colour band doesn't make standard resistor sense. that resistor also doesn't match any of the other types on board. suspicious.

147.7K...  The meter is strong with this one!

The issue I have is that I do not have any 3.9M resistors.  I got 3.3M, and the next upper value I got is 4.7M.  I could put a 1.5M and 2.2M in series (I have those values in the small 1/8W? resistors).  That should be equivalent to 3.7M, right?
Series adds up, right?  Would that be close enough?

[antonis]

Sure it adds on, but I'm with Stephen..!! 
(R17 indeed IS 3k9 - Orange/White/Red..)

Put a real 3M3 resistor (don't bother too much with its precise value - order of magnitude is what counts..) and see what happens..

[jfrabat]

Sure it adds on, but I'm with Stephen..!! 
(R17 indeed IS 3k9 - Orange/White/Red..)

Put a real 3M3 resistor (don't bother too much with its precise value - order of magnitude is what counts..) and see what happens..

I put 3M8 (in series, measured resistance).  I now have sound coming through.  Now the issue is that I cannot hear any phasing, regardless of pot position.  The only difference comes from the Bias trimpot, and anything other than the minimum will cause clipping in the signal when strumming (when signal is loud).  But at least we have progress, right?

[antonis]

But at least we have progress, right?

One at a time.. Nothing better than this..

[jfrabat]

So, what do I try now?

[Slowpoke101]

Progress is good 

Silly question time - You have matched FETs (which is good, very good ). What exact type are they ?
We need to confirm that they are installed in the board correctly. Not all FETs have the same pin-out even from the same group types.

You have a tantalum for C12 (15uF ). Is it installed with the correct polarity ?

[jfrabat]

Silly question time - You have matched FETs (which is good, very good ).

Yeah, I bought them from JMK when I bought the board.  I wish I had used sockets, but a little late for that (I built this board when I started building pedals; maybe 3 years ago?).  They all came labeled.

What exact type are they ?
We need to confirm that they are installed in the board correctly. Not all FETs have the same pin-out even from the same group types.

Here is one I took the measurement off..  They are 2N5952

You have a tantalum for C12 (15uF ). Is it installed with the correct polarity ?

I checked yesterday with the others I have in stock; polarity is correct. 





I CAN change it to this, though...

[Slowpoke101]

All looks good so far. Yes, it is a pity that the FETs aren't in sockets but not to worry.
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.