Boss RPS-10 troubleshooting (very low output)

[hangnef]

Been lurking here for a while, this is my first post.

I have a Boss RPS-10 which I modded with these changes:

http://www.experimentalistsanonymous.com/board/index.php?topic=1186.0

Mods seemed to go fine, spent a while enjoying the new functionality before
putting it back together.  With each step of putting it all back, I checked
again to make sure I didn't break anything.  After putting the final part of the
case back on, it stopped working. 

I've spent the past week or so reading Boss pedal repair posts here (learning a
TON) and trying to troubleshoot.  After some audio probing and looking with my
scope I thought that maybe the preemphasis op-amp (13a) was the issue based on the output pin
showing pretty much nothing and the volume being extremely quiet.  Replacing it made no change.
I understand very basic op-amp usage, but this one is a bit of a challenge for me, with the
capacitors, etc.

This is the section of the circuit I've been testing.



Full schematic is here

http://www.univertron.com/SAS/manuals/Boss%20(Roland)/Boss%20RPS%2010/rps10-master-schematic-450dpi.gif

One thing that I've picked up from reading is that the biasing of the
non-inverting input should be 4.5V, however, it seems to be pretty close to the input
voltage 9v.  Below are some images from my scope, the input signal (a 1Khz 300mv sine wave)
and the AC and AC+DC @ pins 3,2, and 1 of the op-amp.  I'd greatly appreciate any pointers
on where to look/test or go read.

My next guess would be the diodes, but I honestly have no idea.  I'd also be
interested in understanding what those 2 diodes do (is it protection for the op-amp?).  Any
insight into how the circuit works would be awesome as well.

[Slowpoke101]

Welcome to forum. Hopefully we can help sort out your problem.

The first thing to do is to make sure that you did not damage or break any components or wires. Then make sure that you have not got any stray wires or solder splashes bridging (shorting) anywhere.
After that measure the voltages on IC19b - lower right on the full schematic. This part supplies the Vref (4.5V) power rail. If it is too high or too low, the pedal will not work correctly.
So check those things out and let us know how you go.

[hangnef]

Thanks for the warm welcome! 

I went over the board and could not visually see anything broken, cracked joints, etc.

I also checked the voltage @ pin 7 on IC19 and it is 4.48V. 

[Slowpoke101]

The next step would be to find R25 (1M resistor). One lead goes to IC13 pin 3 and the other lead should connect to pin 7 of IC19. Make sure that those connections are OK. Testing its resistance would be handy but you would need to remove one lead of the resistor from the board to take a meaningful measurement.

Then measure pins 1, 2 and 3 of IC13a. They should be around 4.5V If they are not remove IC13 and measure again. The voltages won't seem to make much sense but it will help with diagnosing what may be the problem.

[antonis]

I'd also be interested in understanding what those 2 diodes do (is it protection for the op-amp?).

IMHO, they set the differential input voltage to 600mV, resulting in non-inverting gain of 2..
(that's why you have on pin1 1.2V_pp with a pin3 input of 0.6V_pp whereas the "theoretical" close-loop gain should be 22.36..)

[hangnef]

The next step would be to find R25 (1M resistor). One lead goes to IC13 pin 3 and the other lead should connect to pin 7 of IC19. Make sure that those connections are OK. Testing its resistance would be handy but you would need to remove one lead of the resistor from the board to take a meaningful measurement.

Then measure pins 1, 2 and 3 of IC13a. They should be around 4.5V If they are not remove IC13 and measure again. The voltages won't seem to make much sense but it will help with diagnosing what may be the problem.

Verified continuity between R25 and pin 7 of IC19.  Also pulled one leg of R25 and it registers 1M.

The voltages of pins 1, 2, and 3 of IC13a in circuit are in the screenshots of my first post, here are the pics with IC13a pulled, again w/ 300mv amplitude 1Khz sine wave.

Pin 1


Pin 2


Pin 3


The bypass volume seemed to go up when I made contact w/ Pin 1.

As far as DC, this is what I measured, all of them went down once I attached my meter (some sort of loading?)

Pin 1 2.85v
Pin 2 2.85v
Pin 3 2.85v

I also measured 1.5v across r25.

[hangnef]

I'd also be interested in understanding what those 2 diodes do (is it protection for the op-amp?).

IMHO, they set the differential input voltage to 600mV, resulting in non-inverting gain of 2..
(that's why you have on pin1 1.2V_pp with a pin3 input of 0.6V_pp whereas the "theoretical" close-loop gain should be 22.36..)

Thanks for that info, makes sense given the test point pictures on the schematic.  I'm not sure how to calculate the gain of this particular circuit (I only know how to calculate for very simple stuff like a single feedback resistor, etc).  Does the .6v come from the forward bias'ed drop of the diodes?  And having 1 going each way accounts for both directions of the incoming AC signal?

[Tony Forestiere]

After putting the final part of the case back on, it stopped working. 

Hmmmm.
Welcome. 

[hangnef]

After putting the final part of the case back on, it stopped working. 

Hmmmm.
Welcome. 

Thanks!

And, hehe, yeah.  I thought maybe when trying to align all of the input jacks on the back to the holes maybe something cracked etc.  I was pretty gentle.  So far nothing in that area has seemed to pan out.

[antonis]

I'm not sure how to calculate the gain of this particular circuit (I only know how to calculate for very simple stuff like a single feedback resistor, etc).  Does the .6v come from the forward bias'ed drop of the diodes?  And having 1 going each way accounts for both directions of the incoming AC signal?

Quite right..!! 

Signals higher than 600mV are shunted to GND (both via diode/R8/C9 and diode/R12 - you may alternatively consider it as one diode always facing to virtual ground) so they aren't "included" in NFB loop..
For signals lower than 600mV, diode pair sets the differential input voltage to 600mV (it actually offsets the inverting "vitrual" ground to 600mV, effectivelly seting R8 out of game).
That sets the feedback factor to unity resulting in a gain of 2 due to non-inverting arrangement..

P.S.
Oversimplified, of course, but I think you get into the point..

[hangnef]

Yup, I think I get it. 

With that info, the voltages I measured don't make much sense to me (and probably why it isn't working).  Would it be worth changing the 2 diodes?  Or would it make more sense to pull them and test them?  Or is it possible the issue is somewhere else?

Wondering what to try/test next now that I've pulled the op-amp out of circuit.

[antonis]

<As far as DC, this is what I measured, all of them went down once I attached my meter (some sort of loading?)
Pin 1 2.85v
Pin 2 2.85v
Pin 3 2.85v
I also measured 1.5v across r25.>

IC3a seems to work OK by itself..

Considering no DC voltage difference between pins 3 & 2 (and, of course, no current flowing into op-amp inputs) tell us that the 1.25μA flowing though R25 (for a 1.25V drop) find some way to a lower voltage point..

First suspect should be shorted C9 cap..
(current flows from Vref through R25, D2 & R8 to GND..)

BUT

in such a case, you shouldn't measure 2.85VDC on pin 1 'cause with C9 shorted R8 is effectivelly connected to GND resulting in DC gain of 1 + R12/R8..

You have to find the "leaky" point around pin3 so lift up diodes to see if IC3a works well as "ordinary" AC amplifier with gain of about 22, preferably somewhere in the mid-band of R8/C9 & R12/C7 cut-off points..

[hangnef]

Ok, some measurement progress, still not sure what to make of it all.

Put R25 back in circuit, put in a socket for the op-amp chip and put that back in circuit.  Pulled the 2 diodes out of circuit.
If I'm understanding correctly, with the 2 diodes out of the picture, this small circuit is in effect around pin 3.



So no current goes into the op-amp so these should be the only components between the 4.5 v @ R25 for biasing the op-amp and the input signal.
What I'm seeing is pretty much rail voltage at the other end of R25.  I don't understand how that is happening given the 4.5V and the 300mv input signal. 

I had also measured every pin on the op-amp and saw the same amount of DC, not seeing the expected gain probably due to no headroom.  Given the 4.5 V bias, I would expect the V+ for the op-amp to be the 9v and the V- to be ground (0v).  Pin 4 is -V and I wouldn't expect that to be the rail voltage.  It's also extremely likely that I'm mistaken.   

Here are some screenshots:

4.5V at one side of R25


8+V + input signal @ other side (pin 3 of op amp as well)


Input signal @ other side of C8 (didn't change the vertical scaling, but it is ~280mv amplitude 1khz input signal)

[antonis]

@hangnef: I'm confused a bit about which of your scope measurements are DC coupled and which are AC coupled.. 

BUT

you shouldn't measure DC higher than +4.5V on pin 3 in any case..!!

[hangnef]

Hi antonis,

Sorry about that... those last 3 were all DC coupled.  I tried to make sure each sshot showed in the lower left the coupling, I was probably not consistent.   
I'll make sure with all pics I have that going forward.   

And I agree, > 4.5 seems to be wrong.  Given the circuit I think is in play, I'm unsure where that would come from.  My understanding is the op-amp input is both high impedance and (in theory) does not allow any current flow into it.  I'm going to poke around to see why all the pins on the op amp have rail voltage.  I had swapped op-amps and I got the same behavior.  I don't know enough about the inner workings of an op-amp to know if the V+ and V- are, let's say 9v what I would see on pin 3 in a totally working circuit.

If you have any other suggestions, that would be great as well!

thanks everyone for the help so far, learning a ton and it seems getting closer to the culprit.

[antonis]

I don't know enough about the inner workings of an op-amp to know if the V+ and V- are, let's say 9v what I would see on pin 3 in a totally working circuit.

With both Vcc & Vee powered on the same voltage (zero voltage difference), there shouldn't be any kind of totally working circuit..!! 

Check step by step what's happening between Power Supply (+9V), Vref (+4.5V) voltage divider and 1M non-inerting bias resistor..

edit: If by V+ & V- you mean non-inverting & inverting inputs, plz ignore my first sentence comment..

When both op-amp inputs are set/biased on exactly the same voltage, output is set on op-amps inherent input offset voltage times close-loop gain..

[PRR]

> I would expect the V+ for the op-amp to be the 9v and the V- to be ground (0v).  Pin 4 is -V and I wouldn't expect that to be the rail voltage.

One pin should be near +9V, another near ZERO. The rest would typically be near half-way.

I sure suspect you have lost the power return from chip to ground. That would explain "all voltages high".

[hangnef]

Thanks for all the ideas everyone!

I made some good progress over the past few days.  I was thinking about it, since I couldn't see where the 9v might come from, I wondered if there is something wrong with ground in that area, and it seems that the end of C9 and pin 4 of the op-amp which are connected to ground, were not.  I've looked at the trace that connects a point that does pin out to ground and also to those points and I see no signs of a break in the trace, even with a magnifying lens.  I jumped them and I have volume back!

However, no wet signal at all.   

I need to put the 2 diodes back in circuit, but I can't imagine that will change the no wet situation.

Any idea on what could be going on with that ground (not so great picture below)?  I'll post back after I put the diodes back in, but any pointers on where to start for the wet signal would be great.

[Slowpoke101]

The break in the track is shown here;



Have a close look at the other board mounted jacks and their solder points.
This sort of damage is not uncommon.

[hangnef]

Ok, I bridged that trace and also put the diodes back in.  Dry sound is good, verified the gain is 2 (600mv p2p -vs- 1.2v p2p).

When I'm not using a signal generator w/ a fixed sine wave, like audio for example, I do hear some wet signal when the mix is 100% wet.  It is pretty faint, and sounds distorted.  All the different functions seem to work.  When I put it on reverse, it sounds like it is in reverse, etc.

I wonder if an opamp or transistor blew.

It also makes a little sense to why that last step of putting it back together may have caused this.  To get the last piece on, you have to align the connectors to the holes which means moving them a bit.  Moving them probably caused the trace to break.