FX90 Analog Delay Troubleshooting and Repair help please

[guitjr]

Recently I posted a topic "DX90 Troubleshooting (and Repair?)". After voicing some initial concerns (having to do with my inexperience with digital multimeters, etc.) I received a kind reply andsering my questons and assuring me that if I went through the work of filling out the instructions specified in http://www.diystompboxes.com/smfforum/index.php?topic=29816.0 that I would soon be receiving some help.

I filled out the instructions and waited patiently but nothing came.

Guess what. Reviewing my post I discovered the probable reason I've heard nothing has been because I was referring to a unit that probably doesn't exist! I was referring to a DX90 and meant to be talking about an FX90. 

So here is all that information related to the DOD FX90 analog delay.

Looking forward to some of your expertise to help me rejuvenate this unit and learn more about troubleshooting electronic equipment.

Thanks in advance.

[guitjr]

Here is the checklist to fill out:
Project: Debugging and repairing a FX90
1.What does it do, not do, and sound like? Originally worked (months ago) Now there is a difference between when the switch is engaged or not (i.e., the switch works!) and when an instrument is plugged in I can cause it to go into some oscillation and feedback (but not like it's supposed to).
2.Name of the circuit = FX90
3.Source of the circuit (URL of schematic or project) = Project: top left corner of http://www.americaspedal.net/fx90/
Schematic: http://experimentalistsanonymous.com/diy/Schematics/Delay%20Echo%20and%20Samplers/DOD%20FX90.jpg (I believe, because the parts diagram accompanying the schematic is accurate)
4.Any modifications to the circuit? N
5.Any parts substitutions? No, working on original unit
6.Positive ground to negative ground conversion? Y or N (Negative ground, I believe)
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? => 9.53v
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 = 9.53v
Voltage at the circuit board end of the black battery lead =  .1 mv

Now, using the original schematic as a reference for which part is which (that is, which transistor is Q1, Q2, etc. and which IC is IC1, IC2, C1, and so on) measure and list the voltage on each pin of every transistor and IC. Just keep the black lead on ground, and touch the pointed end of the red probe to each one in turn. Report the voltages:



Because I cannot verify that the schematic (I found which claims to be accurate) is exactly the one corresponding to this unit I have taken a photo of the parts layout and done my best to measure and accurately list the voltages of all diodes, transistors, and ICs. Because I cannot associate each diode and transistor with the schematic, I numbered these components in top > bottom; left > right fashion; labeled the photo and used that nomenclature to provide the following listings.

I assume all the measurements that fell within the mv range (many of which fluctuated) are irrelevant. For that reason, I didn't indicate the fluctuations (which may, one time, have been fluctuating up; and another, have been fluctuating down).

Diodes         
d1:    A= .4 mv   K=5.96   
d2:    A= 8.66   K=8.44   
d3:    A= .3 mv   K=9.5   
d4:    A= 200 mv   K=7.98   
         
Transistors         
t1   E=5.3   B=.5mv   C=8.8
t2   E=84 mv   B=.546   C=5.67
t3   E=.3 mv   B=.2 mv   C=1.23
t4   E=4.41   B=4.41   C=200 mv
t5   E=.2 mv   B=.2 mv   C=1.81
t6   E=3.74   B=4.28   C=8.77
t7   E=199 mv   B=.71   C= 8.78
t8   E=.72   B=1.12   C=8.78
         
ICs         
MN3101         
1   8.26      
2   4.15      
3   ~ .6      
4   4.14      
5   8.1      
6   4.2      
7   5.75      
8   0.56      
         
MN3005         
1   8.88      
2   4.13      
3   5.98      
4   5.98      
5   .5mv      
6   4.14      
7   5.29      
8   5.57      
         
NE570N         
1   1.21      
2   1.82      
3   1.82      
4   .4mv      
5   3.38      
6   3.38      
7   - 23mv      
8   1.82      
9   1.82      
10   4.37      
11   3mv      
12   3.5mv      
13   8.95      
14   52mv      
15   1.818      
16   12mv      
         
HCF4007VBE         
1   8.84      
2   8.85      
3   .2mv      
4   .2mv      
5   8.84      
6   8.84      
7   .2mv      
8   .2mv      
9   4.02      
10   8.43      
11   .3mv      
12   8.03      
13   .3mv      
14   8.82      
         
TLO22CP         
1   4.45      
2   4.43      
3   4.43      
4   .3mv      
5   4.43      
6   4.58      
7   4.43      
8   8.82      

[Fender3D]

TR8 (your Tr3) is shorting NE570's pin 1 to GND, thus blocking wet signal. It may be TR8 or 4007 forcing it...
Check your voltages with pedal on AND off.

Other voltages are weird [if you have 5.98V on MN's pins 3-4, you should have 5.98 on TR3's (your T1) base]

[guitjr]

WHEW!

Here are all the voltages redone in both on and off (footswitch) mode.

Before taking these measurements I made sure to turn the three main pots (Delay, Mix, Repeat) to mid-position. This adjustment probably accounts for the fact most of the readings are slightly lower than they were originally.

All the measurements (most of which varied a bit each time) were (at least) double-checked. Unless otherwise noted all readings are in Volts. One surprise was that the act of testing two of the pins on the 4007 turned the footswitch ON!

Here are the figures (I did my best to get them to line up.)

Diodes      Original      On    Off   
d1:    
     A= .4 mv       0.518   0.514   
     K=5.96          5.57   5.52   
d2:    
     A= 8.66        8.13   7.4   
     K=8.44        7.5   7.4   
d3:    
     A= .3 mv        0.3   0.4   
     K=9.5        8.6   8.6   
d4:    
     A= 200 mv       100mv   -31mv   
     K=7.98        7.51   .3mv   
            
Transistors            
t1   
        E=5.3        4.9   5.1   
        B=.5mv        5.5   5.4   
        C=8.8        8.1   8.1   
t2   
        E=84 mv        219mv   215mv   
        B=.546        .7          .69   
        C=5.67        5.1   5.1   
t3   
        E=.3 mv        .2mv   .3mv   
        B=.2 mv        .1mv   .61v   
        C=1.23        1.2v   33mv   
t4   
        E=4.41        4.1   3.4   
        B=4.41        4.1   1.9   
        C=200 mv        300mv   .18mv   
t5   
        E=.2 mv        .2          .2   
        B=.2 mv        .2          .7   
        C=1.81        1.8   27mv   
t6   
        E=3.74        3.4   3.4   
        B=4.28        3.9   3.9   
        C=8.77        8.1   8.1   
t7   
        E=199 mv        240mv   245mv   
        B=.71        .75   .76   
        C= 8.78        8.2   8.1   
t8   
        E=.72        .77   .78   
        B=1.12        1.16   1.19   
        C=8.78        8.2   8.13
   
ICs            
MN3101            
1   8.26        7.67   7.46   
2   4.15        3.87   3.76   
3   ~ .6        .8mv   .5mv   
4   4.14        3.88   3.77   
5   8.1        7.35   7.15   
6   4.2        .525   .516   
7   5.75        5.35   .517   
8   0.56        .512   .504   
            
MN3005            
1   8.88        8.29   8.08   
2   4.13        3.87   3.77   
3   5.98        5.6   5.53   
4   5.98        5.65   5.5   
5   .5mv        .3mv   .5mv   
6   4.14        3.87   3.77   
7   5.29        4.95   4.83   
8   5.57        .517v   .5v   
            
NE570N            
1   1.21        1.2v   34mv   
2   1.82        1.82   1.81   
3   1.82        1.82   1.81   
4   .4mv        .2mv   .3mv   
5   3.38        1.82   1.82   
6   3.38        3.37   3.37   
7   - 23mv   3.37   3.37   
8   1.82        1.82   1.82   
9   1.82        1.81   1.81   
10   4.37        4.38   4.38   
11   3mv        1.81   1.81   
12   3.5mv     1.82   1.82   
13   8.95        8.38   8.22   
14   52mv        1.82   1.82   
15   1.818     1.82   1.82   
16   12mv        1.08   1.17   
            
HCF4007VBE            
1   8.84        8.38   0   
2   8.85        8.37   8.28   
3   .2mv        .1mv   0   touching this pin with the V lead turned on footswitch!
4   .2mv        .1mv   0   
5   8.84        8.37   .2mv   
6   8.84        8.37   .1mv   
7   .2mv      0          .1mv   
8   .2mv        0           8.04   
9   4.02        3.73v   3.3mv   
10   8.43        8.03   7.97   
11   .3mv        0          .2mv   touching this pin with the V lead turned on footswitch!
12   8.03        3.1mv   3.9mv   
13   .3mv        8.07v   8.01v   
14   8.82        8.37   8.31   
            
TLO22CP            
1   4.45        4.22v   4.17v   
2   4.43        4.21   4.17   
3   4.43        4.21   4.16   
4   .3mv        0   0   
5   4.43        4.2   4.16   
6   4.58        4.24   4.16   
7   4.43        4.2   4.16   
8   8.82        8.36   8.28   


I've also included a photo of the hardware-to-board connections—just to make sure they're not the problem.

[guitjr]

OK, looking at my figures and referring that information to the comments I received from Fender3D I have the following to report:

Other voltages are weird [if you have 5.98V on MN's pins 3-4, you should have 5.98 on TR3's (your T1) base]

Rechecking the voltages I notice that I must have made an error in my initial readings. The new readings confirm there is a voltage of 5.5 V on both the base of TR3 (My T1) and pins 3 and 4 of the MN3005.

TR8 (your Tr3) is shorting NE570's pin 1 to GND, thus blocking wet signal. It may be TR8 or 4007 forcing it...
Check your voltages with pedal on AND off.

I've checked (all) the voltages. These are the voltages that relate toTR8 and pin 1 of the NE570

t3   (TR8)
            Before         On      Off   
        E=.3 mv        .2mv   .3mv   
        B=.2 mv        .1mv   .61v   
        C=1.23        1.2v      33mv   

NE570
           Before   On      Off   
Pin1   1.21        1.2v   34mv   

One more thing:
Looking at this one particular (Google) image of a transistor



  I'm confused about telling which is the Emitter, Base, and Connector of the NPN Transistors in the circuit. I thought that one ought be able to tell the configuration with a transistor in hand—and assumed (from looking at most of the pictures in Google Images that the Emitter would be on the left when viewing the flattened head of the transistor. Now (aside from turning the board around and tracing the schematic, which I didn't do) I don't know.

How should I proceed now?

[armdnrdy]

  I'm confused about telling which is the Emitter, Base, and Connector of the NPN Transistors in the circuit. I thought that one ought be able to tell the configuration with a transistor in hand—and assumed (from looking at most of the pictures in Google Images that the Emitter would be on the left when viewing the flattened head of the transistor. Now (aside from turning the board around and tracing the schematic, which I didn't do) I don't know.

How should I proceed now?

You look up the data sheet for the transistor(s) in question to find the pinout.

[PRR]

The two lower-right pictures show two *different* pinouts.

50+ years ago, most US transistors had one pinout, and for some reason most Japanese transistors had a different pinout. Also packages have changed. And sometimes some BIG customer will order a part made with yet another pinout, and it happens to come into common use.

So while Larry's recent post shows two the same, not all are the same.

I've lost track of what exact transistors you used, but since you have them handy, you can probably find the exact type's data sheet.

[armdnrdy]

I posted the pinouts for the trannys listed in the schematic that guitjr posted.

I forgot the JFET:

[Fender3D]

How should I proceed now?

Well, sadly it's  signal probing-time..

Better if you have a signal generator and a o-scope...
but for a quick overview, a jack cable, your finger and an amp might be enough...

In substance, follow the rabb.. cough, signal, from input and check each active stage (according to schematic), up to output and report where it fails

[duck_arse]

if I can suggest: go back to the original board layout you linked to, and adjust your part numbering on the photo posted above to match, they are the same as far as I can tell. then, paying particular attention to the pinouts your datasheets provide, retest your volts with all the right board refs and pin ID's in place. things may then become more readilly apparent. to someone.

there is a thread somewhere hereabouts on making and using an audio probe.

[guitjr]

(Normally I would follow duck_arse's suggestion that I correct all the graphics I'd already posted before going on, but since I think we are beyond the stage where that information is necessary and since I spend an awful lot of my life properly preparing for eventualities that never happen because all I did was prepare I decided to (using all the kind advice I've been given)carry on in the audio probe direction.

I fashioned an audio probe ala http://www.diystompboxes.com/pedals/debug.html , plugged it into my amp and 'drove' the FX90 with the headphone output from a clock radio I had lying around... So I got to do the audio probe by listening to music!

I also:
a)    "corresponded" all my transistor designations to those on the schematic (TR1=T8; TR2=T2; TR3=T1; TR4=T8; TR5=T7; TR6=TT5; TR7=T6; TR8=T3)
b)   Looked up the transistor specs to make sure I knew where their EBCs (and GSDs!) were so that I culd accurately report audio probe points on the schematic.

And I have the following to report. (Well, "report" is maybe overstating the case, since I don't know what I am talking about.)

Since I don't know what is going on in the circuit I really have no way of knowing how to actually "trace" the signal. So I opted for checking all the IC and transistor points.

I figure the most telling place to do this report is directly on the schematic. That way someone might be able to tell me what I need to re-probe,  further probe or report on.

The red dots indicate places where the audio probe returned music. I have not indicated where the music might have been louder or less loud (hard to tell); where hum was present; etc. Although the stompbox was activated, at no point did the sound-trace I heard have any Delay or Repeat properties.

click to enlarge 

What have I learned (about the FX90. later I'll have a bunch of questions about the process) and what do I do next?

[guitjr]

Oh no Mr Bill!! (Offline I'd asked one of you guys how to structure a graphic so that it could come in small and click to normal size. Apparently it didn't work.)

So here's the results of the initial audio probe in full size.

[armdnrdy]

[Fender3D]

Ok,
you've got sound @ MN's outputs, just we don't know WHAT signal is it...
If you're lucky, it's input signal with a very short delay...
Hard to tell without an oscilloscope, though.
If you have a DMM with frequency counter, check MN3101's pins 2 and 4, if you have a clock signal (maybe some 100KHz)

Otherwise try "fingering" the MN3101 right side (pins 4-8 and neighbor parts) and check if sound changes.
Whether it changes, then you've spot where looking for.
If nothing happens, one of MNs is not working properly, but for a reliable diagnosis you'll need the proper tools...

[guitjr]

Seems my DMM can "Measure Frequency and Duty Cycle". Does this mean the DMM has a frequency counter? And if it does how do I check (MN3101's pins 2 and 4) to see if I have a clock signal.

The instructions that go with this functionality are:
"Measuring Frequency and Duty Cycle"
1)   Connect the black test lead to the CO< jack and the red test lead to the V/omega/F/Hx jack.
2)   Set the function switch tat the Hz/DUTY position and connect test leads across the source or load under measurement.
3)   Push Hz/DUTY selecting button when you need to test the duty cycle of the frequency.
(Note: The input voltage should be between 1V and 10Vrms ac. If the voltage is more than 10V rms, reading may be out of the accuracy range."

[armdnrdy]

Connect the black lead to ground (any ground) and the red lead to pins 2 or 4 of the MN3101.

[guitjr]

I hooked the meter up as indicated (I did not Push Hz/DUTY selecting button when you need to test the duty cycle of the frequency.) and took readings.

Pins 2 and 4 of MN3101 both read 22.88 (don't know what units).

Using a guitar as input (outputting to an amp) fingering pins 4-8 resulted in sound changes when I fingered pin 6 but none of the sound changes read as "Delay" or "Repeat".

1) Would just replacing the MN3101 guarantee a fix?

[guitjr]

Well I'm disappointed. Seems that repairing a 5-IC device isn't as straightforward as I'd hoped it would be.

But given that I don't have a oscilloscope I guess I'll have to give up on the project (or at least delay it till I run up against someone who does). I really appreciate the help some of you have provided but I'm having a hard time correlating my work to where things have thus far gotten.

To keep this venture from being a total failure, I'd like to ask some general debugging questions.:
   

1)   Without an oscilloscope , could I have done more?
2)   Was the problem definitively traced to the MN3101 based on my voltage measurements? If so, what particular measurements led someone to suspect the MN3101 as the culprit chip. If not, then why was the MN3191 suspected?
3)   Would just replacing the MN3101 guarantee a fix to the unit?
4)   Could I have gotten further by judicious use of the data/spec sheets that come with each of the ICs? (i.e., Is the answer in these, or...
5)   If the situation were such that someone had a working FX90, set the three controls (pots) to the same position as mine (say max resistance), and compared their voltage measurements to those I took (at all the IC terminals) would that definitively disclose the source of the problem?


.

[Fender3D]

1) It depends on how much stubborn you are...
2) No
2b) Who did suspect it?
3) No
4) Yes, but you should, at least, know what unit you're reading...

About question #2,
MN3101 is the clock generator (q 4), if it doesn't feed clock to BBD, BBD can't work.
It's the first check, after signal flow, when troubleshooting BBD pedals

My fingering suggestion:
fingering around pins  5,6,7 you'll add tiny capacitance to the clock capacitor, thus altering clock freq., thus varying BBD's delay time.

If, as it seems, you have clock from 3101 to BBD, you should hear a signal detuning at output.
You said:

...Using a guitar as input (outputting to an amp) fingering pins 4-8 resulted in sound changes...

But you didn't say HOW sound changes, does it become noise? Or it detunes? Or both?

5) I have no FX90 at hand... but I wouldn't rely only on voltages for debugging

[Quackzed]

that cap under t3 looks bad... i think its the 10uf from btween 22k and 22k (off of the ne570 pin12)...  i know, it might be nothing...
as a last ditch, try plugging it in normally and poke around with feedback all the way up, move caps wiggle resistors turn knobs back and forth fast...  see if you can get it to do anything, buzz click and remember where and what did it... electro caps can go bad, could replace em all...