Need help with calibration of DEFX Larry - Mutron Flanger clone

[spacekid71]

Hi everyone,

I have built a DEFX Larry (Mutron Flanger clone) inside a Morley power wah volume pedal. I am powering it with 18V DC as indicated in the build document (https://drive.google.com/file/d/1LDdOgaEvKD7LMD93oQv7EVDiAeX3N072/view). The pedal turns on fine as does the rate LED. The pedal stomp switch also turns on the pedal LED. There is some flanging sound but it seems not completely right, but I have not completed the calibration so that makes sense. I have taken some photos of the fully populated PCB and enclosure in case it will help.

PCB component side



PCB control side



I have started calibrating the pedal and I have been able to calibrate the clock successfully. I did have to change C18 to 68pF for it to work. The next step is biasing the MN3007 BBD IC using a 400 Hz sinewave of 1.25V to 2.5V.



The schematic is provided in the build document and it has a section showing TRIM5 trimmer which does the biasing of the MN3007.



Now for the actual problem that I am having, irrespective of whether I add the 1.5V 400 Hz sine wave to the TP3 test point or not, the output of the MN3007 at TP2, shows the following:

TRIM5 fully turned left



TRIM5 in the middle


TRIM5 full right


From reading the calibration section of the documentation, it seems to suggest that feeding the sine wave in at TP3, would also have the sine wave appear at the output of the MN3007 IC. It might be asymmetrical or distorted but TRIM5 would allow to adjust the sine wave so that it is symmetrical and not distorted. As shown in the above photos though, you can see that there is no sine wave whatsoever.

I would be grateful if the experts could give me a hand with this one.

Cheers,

Mart

[ElectricDruid]

As a first guess, I'd say check that the R63 and R64 100K resistors to +Ve actually have good connections. If one wasn't soldered correctly, you might see what you're looking at.

As a second guess, you need to make sure that Trim5 is set correctly, so that what's going *into* the BBD is where it needs to be. If you don't have  a signal generator,/oscilloscope just play some audio from your phone through the effect, and then tweak the trim until you get the best sound from the other end. You can get pretty close like this, and once you have one end pretty good, you can tweak the other to get that right too.

HTH, good luck!

[duck_arse]

please be measuring the values of R61 and R62. and what voltage appears at their junction?

[spacekid71]

please be measuring the values of R61 and R62. and what voltage appears at their junction?

I just measured the two resistors and here are the results:

R61: 0V
R62: 15V
R61-R62 junction: 1.11V

[spacekid71]

As a first guess, I'd say check that the R63 and R64 100K resistors to +Ve actually have good connections. If one wasn't soldered correctly, you might see what you're looking at.

As a second guess, you need to make sure that Trim5 is set correctly, so that what's going *into* the BBD is where it needs to be. If you don't have  a signal generator,/oscilloscope just play some audio from your phone through the effect, and then tweak the trim until you get the best sound from the other end. You can get pretty close like this, and once you have one end pretty good, you can tweak the other to get that right too.

HTH, good luck!

I measured both R63 and R64 and they are at 15V so they seem to be fine. I looked at this build document from armdnrdy (Larry) who did the original clone I believe, and the sine wave signal should just be coming through normally? I am not sure I understand how the sine wave would go into the MN3007 and would come out not transformed... Maybe someone can educate me?



Aside from that I am still not much clear on what is going on. Why does the wave form at TP2 look so strange with TRIM5 set full left, right, or in the middle?

[Rob Strand]

Now for the actual problem that I am having, irrespective of whether I add the 1.5V 400 Hz sine wave to the TP3 test point or not, the output of the MN3007 at TP2, shows the following:

TRIM5 fully turned left

What's wrong is you are looking at the waveform with a very fast us/dev setting on the oscilloscope so you are seeing samples with glitches and not the 400Hz sine wave cycles.

A 400Hz signal has a period of 2.5ms.   Suppose you want to see 4 or 5 periods on the display, that's 10ms to 12.5ms.   Your oscilloscope has 14 horizontal divisions which means you need 10ms/14 = 0.7ms/div to 12.5ms/14 = 0.9ms/div.   So  you need to set the oscilloscope to 500us/div or 1ms/div, a lot slower that what you have at the moment.

When your monitor the output of the BBD and adjust TRIM5 you should see the waveforms similar to the blue analog scope waveforms you just posted.

Hint: connect channel 1 to the input waveform and trigger off that, then connect channel 2 to the BBD output/test point.    The input signal is clean so the oscilloscope will trigger better.

[Govmnt_Lacky]

@spacekid71

When you power your Mutron Flanger pedal.... are you seeing the Rate LED light up at all? Like, even when the pedal is not activated?? I had to make adjustments to a resistor value in my build to get the Rate LED to be dim enough to not be noticably "on." Otherwise, it was dimly lit constantly as long as power was plugged in.

[spacekid71]

Hi Govmnt_Lacky,

Just had a look at the RATE LED and it seems to be behaving correctly. Will let you know if I notice anything suspicious in the same vein as you are describing.

Cheers,

Mart

[spacekid71]

What's wrong is you are looking at the waveform with a very fast us/dev setting on the oscilloscope so you are seeing samples with glitches and not the 400Hz sine wave cycles.

A 400Hz signal has a period of 2.5ms.   Suppose you want to see 4 or 5 periods on the display, that's 10ms to 12.5ms.   Your oscilloscope has 14 horizontal divisions which means you need 10ms/14 = 0.7ms/div to 12.5ms/14 = 0.9ms/div.   So  you need to set the oscilloscope to 500us/div or 1ms/div, a lot slower that what you have at the moment.

When your monitor the output of the BBD and adjust TRIM5 you should see the waveforms similar to the blue analog scope waveforms you just posted.

Hint: connect channel 1 to the input waveform and trigger off that, then connect channel 2 to the BBD output/test point.    The input signal is clean so the oscilloscope will trigger better.

Thanks for taking the time to explain this to me, I appreciate it. I understand what you are saying so I am applying the 400 Hz sine wave at the input jack and I have set the oscilloscope to 1ms/div and connected it up to TP3 (yellow) and TP2 (purple). The three photos are showing TRIM5 set fully left, middle, and fully right. I must still be doing something wrong because the signal at the BBD output does not seem normal. I am either not measuring correctly still, or there is something wrong with the circuit around the BBD chip.

[Rob Strand]

Thanks for taking the time to explain this to me, I appreciate it. I understand what you are saying so I am applying the 400 Hz sine wave at the input jack and I have set the oscilloscope to 1ms/div and connected it up to TP3 (yellow) and TP2 (purple). The three photos are showing TRIM5 set fully left, middle, and fully right. I must still be doing something wrong because the signal at the BBD output does not seem normal. I am either not measuring correctly still, or there is something wrong with the circuit around the BBD chip.

To be honest I cannot follow the documentation.

The schematic has test points labelled PAD7 etc, which are very hard to find and read, but the instructions talk about test points TP2 etc..    The PAD number and TP number don't match up.    So when you are probing TP2 I cannot work out what point TPx is on the schematic.    That makes it impossible for me to know where you are actually probing.   If the instructions have typos you might actually be probing the wrong point.   I can guess where the test points *should be* on the schematic but that doesn't help if the documentation doesn't match the PCB.

To side step those issues you could try probing  pin 3 and/or pin 1 of IC10.    You should get a sine-wave coming out.

[spacekid71]

To be honest I cannot follow the documentation.

The schematic has test points labelled PAD7 etc, which are very hard to find and read, but the instructions talk about test points TP2 etc..    The PAD number and TP number don't match up.    So when you are probing TP2 I cannot work out what point TPx is on the schematic.    That makes it impossible for me to know where you are actually probing.   If the instructions have typos you might actually be probing the wrong point.   I can guess where the test points *should be* on the schematic but that doesn't help if the documentation doesn't match the PCB.

To side step those issues you could try probing  pin 3 and/or pin 1 of IC10.    You should get a sine-wave coming out.

Yeah it is difficult that TP points and pad numbers don't line up. I have looked at this with the PCB and the schematic and used continuity on my multimeter to figure it out.

TP3    =   PAD3
TP2   =   PAD7

So TP3 is the tip of the input jack and TP2 is at the intersection of TRIM6 and Q3.

[Rob Strand]

Yeah it is difficult that TP points and pad numbers don't line up. I have looked at this with the PCB and the schematic and used continuity on my multimeter to figure it out.

TP3    =   PAD3
TP2   =   PAD7

So TP3 is the tip of the input jack and TP2 is at the intersection of TRIM6 and Q3.

OK got it, so it is like that.

The TP2 signal obviously isn't correct.  There will be some glitches at that point but the waveform should sort of follow a sine wave.


At this point I'd do some basic checks:

1) Check the supply voltages: BBD power IC9 pin 1 and BBD bias IC9 pin 4.
    Also check ground is getting to the BBD IC9 pin 5 using continuity - measure at the IC pin not the PCB.

2) With 400hz sine wave signal applied to the pedal input, check the sine wave is present on
    IC9 pin 3 (BBD analog input).

3) Make sure the both clocks are getting to the BBD chip.
   With the oscilloscope on a fast sweep and no analog input, and using both channels:
   Make sure you are seeing clocks on both pins IC9 pins 2 and 6 and  that
   the signals are complementary.  You can also check the same signals at IC6.
   Take note of the clock voltages, they should swing zero to +VE.

Ignore the voltages on the pic:


If you get that far then there's no real reason not to get a glitchy sine wave at TP2 (TRIM6).  If you don't the BBD chip might have a problem.   What to do from there is a little tricky without replacing the BBD chip, a costly experiment.

Another angle is to run the pedal at a lower voltage like 12V or 15V to see if it comes to life.  Perhaps the modern BBD chip doesn't like 18V.  18V is the absolute maximum voltage for that chip and it's usually unwise to operate at absolute maximum!

You could also try soldering another 1k in parallel with R61 (1k2) to see if it does anything.  You could even try a very low value like 100 ohm.   I don't think it's going to help but if the BBD chip has an issue it might kick it in the right direction.

[Govmnt_Lacky]

Another angle is to run the pedal at a lower voltage like 12V or 15V to see if it comes to life.  Perhaps the modern BBD chip doesn't like 18V.  18V is the absolute maximum voltage for that chip and it's usually unwise to operate at absolute maximum!

Rob,

This pedal runs with a 12 or 15V regulator. As long as input voltage (AC or DC) does not exceed the regulator's requirement then the BBD should see nothing more than the regulator's output voltage.

[Rob Strand]

Another angle is to run the pedal at a lower voltage like 12V or 15V to see if it comes to life.  Perhaps the modern BBD chip doesn't like 18V.  18V is the absolute maximum voltage for that chip and it's usually unwise to operate at absolute maximum!

Rob,

This pedal runs with a 12 or 15V regulator. As long as input voltage (AC or DC) does not exceed the regulator's requirement then the BBD should see nothing more than the regulator's output voltage.

Thanks, found it.  I must have been zoomed in and scrolled past that section.  Before I could only find the Vref voltage divider stuff.

Looks like 15V on the parts list.

[spacekid71]

OK got it, so it is like that.

The TP2 signal obviously isn't correct.  There will be some glitches at that point but the waveform should sort of follow a sine wave.


At this point I'd do some basic checks:

1) Check the supply voltages: BBD power IC9 pin 1 and BBD bias IC9 pin 4.
    Also check ground is getting to the BBD IC9 pin 5 using continuity - measure at the IC pin not the PCB.

2) With 400hz sine wave signal applied to the pedal input, check the sine wave is present on
    IC9 pin 3 (BBD analog input).

3) Make sure the both clocks are getting to the BBD chip.
   With the oscilloscope on a fast sweep and no analog input, and using both channels:
   Make sure you are seeing clocks on both pins IC9 pins 2 and 6 and  that
   the signals are complementary.  You can also check the same signals at IC6.
   Take note of the clock voltages, they should swing zero to +VE.

Ignore the voltages on the pic:


If you get that far then there's no real reason not to get a glitchy sine wave at TP2 (TRIM6).  If you don't the BBD chip might have a problem.   What to do from there is a little tricky without replacing the BBD chip, a costly experiment.

Another angle is to run the pedal at a lower voltage like 12V or 15V to see if it comes to life.  Perhaps the modern BBD chip doesn't like 18V.  18V is the absolute maximum voltage for that chip and it's usually unwise to operate at absolute maximum!

You could also try soldering another 1k in parallel with R61 (1k2) to see if it does anything.  You could even try a very low value like 100 ohm.   I don't think it's going to help but if the BBD chip has an issue it might kick it in the right direction.

Thank you for your detailed explanation on how to investigate. I have some preliminary results from the investigation:

IC9 (MN3007)   -   pin1:   14.98V
IC9 (MN3007)   -   pin4:    1.11V
IC9 (MN3007)   -   pin5:   GND verified using continuity

IC8 (NE571)   -   pin7:   400 Hz sine wave available

It looks like I lose the 400 Hz sine wave after it passes through R57. I have taken part of the schematic and indicated where the signal is okay still and where it is lost. Maybe the 1N34a and 1N914 diodes have something to do with it?

[Rob Strand]

It looks like I lose the 400 Hz sine wave after it passes through R57. I have taken part of the schematic and indicated where the signal is okay still and where it is lost. Maybe the 1N34a and 1N914 diodes have something to do with it

That's good progress.   Highly likely to be the 1N34A's.   You might have some leaky or damaged devices.  If you just want to keep going on the BBD verification what you can do is replace one or both of the 1N34A's with 1N914's.   Deal with the 1N34A issue later later.

Just in case there's a soldering or PCB issue, make sure you are getting continuity from diode side of R57 (2k2) to C35 (47n).

There's a good chance things will start working once you sort out the loss of signal issue.

[Govmnt_Lacky]

It looks like I lose the 400 Hz sine wave after it passes through R57. I have taken part of the schematic and indicated where the signal is okay still and where it is lost. Maybe the 1N34a and 1N914 diodes have something to do with it?

You can always use your multimeter in diode mode to verify that you are getting the proper voltage drop across the diodes. 1N914s should be about 0.6V and the 1N34As should be around 0.35V. It could be something as simple as needing a solder re-flow on the contacts.

Good Luck! 

[Rob Strand]

Probably shouldn't overlook the possibility of (solder) shorts on the PCB around that area of the circuit and around the BBD pins.

[spacekid71]

So I had another look at things just now. I measured the 1N34as and the 1N914s and they came in at around 0.27V and 0.62V respectively. I did reflow all four diodes but that hasn't make any difference. The left side of R57 (on the schematic) gives me 400 Hz but the right side give me about ten times as much, around 4.0 kHz.

I decided to remove C35 (47 nF) just to see if the 400 Hz would appear on the right side of R57 and on the left side of C35, and it did...

One other thing I noticed is that the BOM says to use 1N4148 for D9, D10, D11, D12, but the schematic shows 1N914. I replaced the 1N4148s that I had put just in case, but that hasn't changed anything.

I am wondering what could increase the frequency from 400 to 4000 Hz. Any ideas? I can do some measurements if that would help.

[Rob Strand]

So I had another look at things just now. I measured the 1N34as and the 1N914s and they came in at around 0.27V and 0.62V respectively. I did reflow all four diodes but that hasn't make any difference. The left side of R57 (on the schematic) gives me 400 Hz but the right side give me about ten times as much, around 4.0 kHz.

I decided to remove C35 (47 nF) just to see if the 400 Hz would appear on the right side of R57 and on the left side of C35, and it did...

...
I am wondering what could increase the frequency from 400 to 4000 Hz. Any ideas? I can do some measurements if that would help.

The short/problem is probably on the BBD side of C35.   Perhaps check for shorts around the BBD pins, or even check for shorts to ground on the BBD input using the DMM contnuity.

The 4kHz is probably an artifact of the short.  The signal should be very low.  (A pic of the waveform might help but really the problem is around the BBD.)

It doesn't make 100% sense to me as the cap would short low frequencies less as far as the diodes are concerned.

One other thing I noticed is that the BOM says to use 1N4148 for D9, D10, D11, D12, but the schematic shows 1N914. I replaced the 1N4148s that I had put just in case, but that hasn't changed anything.

So it doesn't look like the diodes are the issue but for the sake of correctness:

The schematic has a bug for sure.   Each pair of diodes should be a 1N4148 in series with a 1N34A.

Here's a PCB pic of the real unit around the diodes:

Top


Bottom, flipped left/right for x-ray view to match PCB top view.


Here's the connections I see on the PCB: