Troubleshooting a Marshall Supervibe SV1

[potul]

Hello everyone

I recently got a used Marshall Supervibe chorus and I received a defective one. The seller didn't accept the return... so I decided to transform my anger into a challenge and try to repair it.
I attach the schematic for reference. It's a BBD based chorus that uses an MN3007
The failure is that only the dry signal comes out, both in bypass and non-bypass mode.
I have used an audio probe to trace the wet signal, and it goes into the MN3007 input correctly, but I can't hear anything coming out from the outputs OP1 and OP2.
Using a oscilloscope I've checked the clock inputs, and they seem to be ok. I'm seeing a clock signal around 10Vpp with a frequency that oscillates from 60 to 100Khz and reacts to the "speed" knob. (higher speed=faster oscillation of the clock frequency)
Bias voltage of the MN3007 is 632mV, vs the 560mV shown in the schematic

Does this seem like a bad MN3007? Anything else I can measure before trying to replace it?

BTW, there is a trimmer in the pcb, but I can't find it in the schematic... I want to try to play with it and see if it fixes something, but I would like to first know what it does.... Any clue on what it could be? Bias adjustment?

[potul]

Here you have a link to the PDF of the schematic if you prefer it.

https://files.fm/u/86xarp3pq

[ElectricDruid]

I'd say bias adjustment is most likely for the trimmer, yes. If it's not working now, you won't harm anything by twiddling it, so give it a try. It's certainly possible to *stop* the signal coming through a BBD by putting the bias way out, so perhaps someone has mucked with it before and thought they broke the pedal. You never know.

[Fender3D]

What's labeled "bias" on schematic it's what datasheet usually calls "VGG".
You wouldn't want to change it... For MN30xx is usually 1 diode drop above GND or 14/15 Vcc.
Bias voltage everybody refers to, when talking about BBD, is the voltage applied @ pin 3 (input).
It's usually just above half Vcc.
You should check if the trimmer you say is on PCB connects to "ACTIVEREF" somehow, if it does, then it's bias setting.
If trimmer do not connect to ACTIVEREF, it may not be bias setting, try reverse engineering following tracks...
So, first check voltage @ BBD's pin 3 is about 1/2Vcc, slightly higher the better, if trimmer is NOT bias setting, you may try and change it (the voltage) by adding one trimmer 470K or higher over R36, then check if you can get some signal on BBD's out pins.

[potul]

oK, So I did some more investigation and measured some DC.

First of all, the trimmer regulates ACTIVEREF voltage. It's sitting right above R36. But looking at the schematic, it looks like this is only used for the biasing of the buffer/bypass section. I don't see how this would impact the MN3007. I have tested the whole range of the trimmer without any improvement, and no change in the biasing of the MN section.

Now for the weird things. MN3007 input DC reading is very low, 0.16v, which is way below Vcc/2. Here you have all voltages in the MN3007 IC:

PIN1: 9.8V
PIN2: 4.8V
PIN3: 0.16V
PIN4: 0.63V
PIN5: 0V
PIN6: 4.8V
PIN7: 4.8V
PIN8: 4.8V

I have also measured some points in the schematic in the section before the MN3007

Point B: 7.1v
Point C: 9.24v
Point D: 0.16v

Do you think there is something wrong in the MN3007 input bias?

Thanks in advance

Mat

EDIT: Now I have seen that ACTIVEREF is attached to TL072 pin3, so I guess it might have an influence to the bias of this section as well. Although I don't notice any difference when moving the trimmer. ACTIVEREF is reading 5V at the moment

EDIT2: I measured wrongly point C. I repeated and the voltage reads 9.24v now.

[ElectricDruid]

What's labeled "bias" on schematic it's what datasheet usually calls "VGG".
You wouldn't want to change it... For MN30xx is usually 1 diode drop above GND or 14/15 Vcc.

Yes! Agree! That wasn't what I was talking about.

Bias voltage everybody refers to, when talking about BBD, is the voltage applied @ pin 3 (input).
It's usually just above half Vcc.

Yes again! That's the one I was talking about, and that's the one that's more commonly trimmable.

[Fender3D]

Sorry Tom I didn't read your post 'cause I was writing mine while you posted...

EDIT2: I measured wrongly point C. I repeated and the voltage reads 9.24v now.

Check IC4 TL072 voltages, but I'm almost sure IC4 is dead...

[potul]

Check IC4 TL072 voltages, but I'm almost sure IC4 is dead...

Ok, I will do it tomorrow afternoon. I too think there is something funky in IC4, pins 6 and seven were around 0.17v, but I will measure and report all pins tomorrow. Thanks for the help.

[Rob Strand]

Check IC4 TL072 voltages, but I'm almost sure IC4 is dead...

Ok, I will do it tomorrow afternoon. I too think there is something funky in IC4, pins 6 and seven were around 0.17v, but I will measure and report all pins tomorrow. Thanks for the help.

On the right track for sure.

Check further back to IC4 pin 1.  IC4 pin 1 derives its voltage from ACTIVEREF.   The IC4 pin 1 voltage feeds through to IC4 pin 7 then that goes to IC1 pin 3.    The whole chain of voltages should sit at ACTIVEREF.

[potul]

Ok, these are the readings at IC4

PIN1:9.16V
PIN2:8.3V
PIN3:5.2V
PIN4:0V
PIN5:8.7V
PIN6:0.12V
PIN7:0.12V
PIN8:9.75V

Somethins seems wrong. Shouldn't we have almost the same voltage in pins 2/3  and 5/6? 

[antonis]

Shouldn't we have almost the same voltage in pins 2/3  and 5/6?

In case of a healthy IC, yes..

[Rob Strand]

PIN1:9.16V
PIN2:8.3V
PIN3:5.2V

PIN5:8.7V
PIN6:0.12V
PIN7:0.12V

All those voltages look totally whacked.

Pin 1 and Pin 3 should sit at 5.2V (=ACTIVEVREF).   That probably could be caused by IC4 faulty, or, a short/bad caps around the IC.

BUT, Pin 5 doesn't even have the same voltage as pin 1.   That would happen if pin 5 was pulling current.

And to make matters worse pin 6 and pin 7 (which are wired together, so they have the same voltage) doesn't follow pin 5.   Pin 5 is near 9V and pin 6/pin7 near zero doesn't make sense either.

Pretty much everything is muddle up.   So a faulty IC4 would make a lot of sense.   Just the same, check soldering around IC4, even though it's unlikely to be the cause.

[potul]

Ok, this confirms my suspects then. At least we have narrowed the issue around IC4. I will first check all solderings around, and try to spot if any short or damaged cap around the area. If I can't spot anything, I will go and desolder IC4 and cross fingers not to damage anything during the process... at least it's a cheap component.

[Rob Strand]

Ok, this confirms my suspects then. At least we have narrowed the issue around IC4. I will first check all solderings around, and try to spot if any short or damaged cap around the area. If I can't spot anything, I will go and desolder IC4 and cross fingers not to damage anything during the process... at least it's a cheap component.

When you are certain an IC is dead like that it's often easier to cut the pins off the top of the package and remove the pins one by one.

[potul]

Ok, this confirms my suspects then. At least we have narrowed the issue around IC4. I will first check all solderings around, and try to spot if any short or damaged cap around the area. If I can't spot anything, I will go and desolder IC4 and cross fingers not to damage anything during the process... at least it's a cheap component.

When you are certain an IC is dead like that it's often easier to cut the pins off the top of the package and remove the pins one by one.

Thanks for the advice. As it is a cheap and easy to replace IC, I will do this. I would probably think twice with the MN3007 

[antonis]

As it is a cheap and easy to replace IC, I will do this.

It should be wise to solder an IC socket and check voltages without IC in place..
(you should get nothing on pins 1, 2, 5, 6 & 7..)

[potul]

Good day!

I made some progress today. I managed to replace the IC4, and now the voltages are back to what was expected. Nice biasing of the signal, and the MN3007 has come back to life, providing a nice vibrato in its output. Good news, as the BBD chip is working fine.
But.... the vibrato signal is not reaching to the end of the path. I took my audio probe and followed the signal. It's reaching pin 5 of the IC6 (another TL072), but not getting out. Pin 7 is dead silent.
So I checked DC voltages in IC6, and they seem to be off as well:

pin1: 0v
pin2: 0v
pin3: 5.2v
pin4: 0v
pin5: 3.7v
pin6: 0v
pin7: 0v
pin8: 9.8v

So again, it seems like a defective chip. What's the probability of having 2 defective TL072  in the same board? Maybe somebody used the wrong power supply and both got damaged?

Any other thing I should check before starting to desolder IC6?

[ElectricDruid]

Any other thing I should check before starting to desolder IC6?

No, not really. That looks like another convincingly-dead TL072!


PS: Your guess of "wrong power supply" is certainly the most likely, though we'll never know for sure.

[Rob Strand]

PS: Your guess of "wrong power supply" is certainly the most likely, though we'll never know for sure.

Seems like the only explanation.   Makes you wonder what happened and why the (shunt) reverse protection diode didn't save it.  It's a 1N4004 not a wimpy 1N4148.  Also remaining TL072's still good.

Two opamps gone in the main signal chain and the MN3007 in the middle of the destruction survive - so lucky.  Must have some Supervibe powers.

[potul]

Ok, quick update from today.

I removed IC6, added a socket and plug in a brand new TL072 and....... IT'S ALIVE !

Now this thing works as it should, nice chorus that teleports me to the 80's. I still have to test it extensively, but so far it's a nice sounding chorus. A little noisy, but with the tone control you can remove most of the hiss.

Thanks everyone for your help, it has been a fun troubleshooting. I learned some new things and restored my faith in the universe. I was really lucky that only two TL072 were damaged.