#1
Hi all,

I'm working on fixing a broken boss tr2. going through the circuit board compoents, it seems there could be 2 blown electrolytic caps. When hooked up to my digital multimeter, they both read about 50K ohms, they should read infinite, right? (The other electro caps on the board do). The caps are 10pF and 470pF.

I guess my main question is should all caps eventually go to infinite resistance? (Multimeter doesn't have a capacitance function)
#2
Irishman, you have the right idea - all caps will look like infinite resistance, when you measure them in isolation. The one thing to watch is that many times a circuit is composed of both a resistor and capacitor in parallel (so when you measure across the capacitor, you measure the resistor path). It might not be a resistor, but some other resistive component (diode, transistor, etc). So you might need to trace the leads on the circuit board to see if they happen to also connect across a resistor.

Most times, if it's something like a diode or transistor in parallel, you'll get a different value (or infinite value) based on which way you place the leads. If you reverse the probe points and get the same value across a resistor, you'll get the same resistance reading.

To really check the cap, if you still think that's the problem - unsolder it from the board and measure it in isolation. Make sure you note which way the cap was in so you can put it back or replace it correctly. There might be a row with "- - - " for the negative side or one or more plus "+" signs on the other lead. To desolder your cap, you can get desolder braid (braided copper mesh) or wire with many copper strands or a solder sucker (best). Use a little flux to pull the solder out into the mesh (or just use the sucker). Try to heat just the solder on the component side and not the cap, and use pliers to get the cap out.

It would be odd to have two electrolytic caps go at the same time. What is the failure mechanism on the TR-2? It doesn't work at all?
#3
Its completely dead, signal through it when "off", but when I stomp the switch its nothing. Looking at the board itself, there's no loose / bad solder joints, so thats why I went to checking each component. That's great info on the parallel resistors etc, as I think there are some parallel components in play here. I hadn't come across that gem! I think I'll take them out and see how they perform anyway, as I can't find anything obviously wrong.

Thanks again!
#4
Do the tops or bottoms of the electrolytic cans exhibit bulging? Have they leaked any electrolyte onto the PCB?

Got any clear detailed photos you can show us?
'It takes 100 guitar players to change a light. One to change the light and 99 to stand around pointing, saying..."Yeah man, look...I can do that too"...'

My gear is in my Profile

Builds & Refurbs
Hondo 780 Deluxe
Gibson Studio
Epi LP100
#5
Just to check, when you turn the pedal on, does the power on LED come no? (that would verify that your battery/wall wart is good and that there is nothing catastrophic shorting out the supply on the board). Also, checking the cans and getting a pic like Phoenix mentioned would be good.
#6
Here's a pic lads, I circled both caps. There's a few shadows that may distort things, but the capacitors aren't too different from others on the board size-wise (not swollen or leaking):



I checked, and both caps are in parallel with resistors, and are showing the same resistance as those resistors. Checked with a working battery and the LED comes on.

Here's a schematic if that helps at all?

Last edited by irishman at Jan 3, 2014,
#7
Nice - schematics and pix!

Well, those are just disk caps, not electrolytics. Disk ones tend to be more reliable. Your schematic shows why they appear resistive (cool!).

Hmmm, so the light comes on so it's trying to power up and do something. You don't have access to an oscilloscope, do you? We might be able to suggest some other way to trace things out. Like if you made a probe with a guitar wire and sent it to a low power practice amp (you could just listen to trace). If you want to try something like we could probably come up with something. Like if you knew that the signal came out on R11, that would tell you something. Then along that path across the top. That shouldn't be too hard and we could talk you through that if you want.

If you (or someone else) did something like hook up a battery adapter backwards, you might blow out something. It has some diodes to seemingly route the power correctly. The IC is probably the weak link that would go if it were hooked up in a bad way.

If you have a simple volt meter, you could try finding those voltages on the bottom 5.5V and 6.5V and see if they're close.
#8
Those arent electrolytic caps. They are ceramics, unpolarised. C23 is part of the tremolo active switch flip flop circuit. C7 just forms part of the low pass filter stage for the output of IC1B.

Key items to check are Q1, Q2, Q3, Q4. Starting with Q1. If isn't working properly, you will get no output no matter if the effect switch is off or on. I'm assuming you understand how to test FETS and BJTs?

Q1 and Q3 switch the pedal into bypass or effect, through the flip flop transistors Q5 and Q7.

Q2 is the final output stage. Its effect is the same as Q1 if it isn't working properly.

First though I'd check if you're getting 5.5V out of Q6 and 4.5v out of IC2B. The lighting of the led is just straight off the 9v battery, which doesn't mean a lot really. The above components must have these volts to work correctly. Need to make sure all the circuit volts are being correctly generated.

Use a process of elimination.
'It takes 100 guitar players to change a light. One to change the light and 99 to stand around pointing, saying..."Yeah man, look...I can do that too"...'

My gear is in my Profile

Builds & Refurbs
Hondo 780 Deluxe
Gibson Studio
Epi LP100
Last edited by Phoenix V at Jan 3, 2014,
#9
timbo:
I don't have an oscilloscope, but I'd be willing to go the route of making a guitar wire probe, sounds fun!

Is it a case of cutting a guitar cable, plugging the jack into an amp (I've a 5W practice amp), and then holding the wire end to the output of the components on the pcb board and seeing if I get any crackle etc. through the amp? Thanks for the help by the way, I'm learning loads here!

I don't think I hooked up a battery connector backwards, do you mean I might have had the polarity wrong on the tip? The pedal worked fine and then one day with the same supply I'd always used decided enough was enough.

phoenix:
I checked the different diodes there and they all appear to be working fine. One thing, Q2 seems to be part of the long black part on the bottom of my picture, right? Seems it's a IC semicoductor from google?

It'll take me a while to get my head around figuring out how to measure the voltage out of the IC's, but I'll report back with the results.

Thanks again!
Last edited by irishman at Jan 3, 2014,
#10
Yeah, I thought this might be fun. First, you don't really need to cut the wire. This is a pretty poor-man's approach, but it might work. Run something into the input jack via the guitar wire there. Now if you had a signal generator, that would be nice (you could run a continuous tone in). You could also just run your guitar in, but then you'll need to keep strumming it. That's option #1. If you have some other sound source like a radio, then you could use an adapter and run some headphone output into the guitar pedal. Keep the level really low and test it with the pedal in bypass to make sure things are working.

Then make your guitar wire "probe". If you have some wires and tape (!!) connect the outer wire of your probe to the ground of the effect. You might be able to get the wire touching both your guitar wire and the other wire to the shiny outer part of where the guitar cord plugs into the effect (either input or output side - doesn't matter).

Then tape another long wire to the middle part of your probe. Make it long because when you move it around, you don't want to pull on the tape (OK, I admit doing stuff like this myself). If the wire is long, you can move it around more easily. Put a book on it as a strain relief to keep it from moving around. Plug the other end of your probe into the practice amp.

Now - before you start probing... Realize that you don't want to really send the 9V supply into your practice amp. I don't *think* it would do anything other than a loud pop, but who knows. They might have a capacitor on the input just to remove any DC like that. But I'd make it a point to carefully pick your probe points. Expect some crackle when you touch *anything* so keep the levels low.

The first thing is to check the "probe". Try probing the input at R15 where the sound source comes in. That should be the "proof of concept" that you're all grounded and such and it should be easy - no special stuff required.

Then start down the signal path. I'd say: R11 (non ground side), C3 (where it touches R5), and R4 (where it touches R1). I'm trying to pick places that if you probe the wrong side you don't get the supply.

You should be able to see the audio signal is there, it's still there, and then some point it's gone. The problem is between those two spots.

It looks pretty straightfoward - there's an oscillator wiggling the gain in IC1B. But you'd think if the wiggling thing isn't working, they you just get a tremalo pedal that passes the sound through without anything going on. If you get nothing out, then it's in the signal path.

You might find the signal is there all the way through. Then the problem would be the output jack (maybe a solder joint or something, but that's easiest to fix.

I had a compressor rack mount (I think) that went and I later found it was a bad solder joint. If you scan the solder side of the board, you may find a place where something doesn't have full shiny solder around the pin and there's some "air" around it. You might push on the bad pin and get the sound to come out. That's an easy fix.

Let us know how it comes out.
#11
Great man, that sounds definitely doable! It'll take me a while to get all the bits and pieces together, so I'll probably attempt it with a clear head tomorrow. I'll check back in with the results.

#12
I've only done troubleshooting on one Boss pedal before, but I had issues with the protection diode (D1) and a couple of opamps (I think they were the M5218AL's). I'd definitely check voltages off IC2B.
#13
Before you decided there aren't any bad solder joints, have you scrupulously examined EVERY joint with a flashlight and high-power magnification? Just because a solder joint looks ok doesn't mean it actually is ok.

Also, don't assume that the problem is a transistor or IC just because it is the most expensive component. Most of the time fairly old (or even fairly new) Boss pedals get bad solder joints due to years of thermal cycles of normal operation and environmental effects.

The vast majority of problems with electronic gear can be solved with simple observation and are something simple. Most of the techies will bust out the DMM and start testing and replacing parts first thing. It's kind of like not seeing the forest for the trees, so to speak.

Just something to think about.
#14
Quote by irishman
timbo:

phoenix:
I checked the different diodes there and they all appear to be working fine. One thing, Q2 seems to be part of the long black part on the bottom of my picture, right? Seems it's a IC semicoductor from google?

It'll take me a while to get my head around figuring out how to measure the voltage out of the IC's, but I'll report back with the results.

Thanks again!


Hi. Don't worry about diodes. First check if Q6 and IC2B are outputting the correct volts as nominated on the schematic, with a good 9v battery installed. This should be a very quick test.

Q2 is a transistor. The long black pinned devices are IC's as designated on the PCB silkscreened text. They aren't the same thing.
'It takes 100 guitar players to change a light. One to change the light and 99 to stand around pointing, saying..."Yeah man, look...I can do that too"...'

My gear is in my Profile

Builds & Refurbs
Hondo 780 Deluxe
Gibson Studio
Epi LP100
#15
Hi all.

So to update, it's alive! The problem was a [drumroll...] bad solder joint!

But this process was more about the journey than the destination. I learned an absolute TON of useful information from this. Seriously, you guys really did teach me a lot about electronics, reading circuits and diagnosing problems. Its something I always wanted to be able to do, up until now if I didn't see an obviously broken connection I'd hit a wall, I feel like I can take on the world now (definitely can't, but still!).

Thanks again!
#16
Cool - I'm glad it worked out! Thanks for letting us know what it really was. Yeah, that sums it up nicely - these things are often more about the journey and being ready for the next one that comes along. These things can be a lot of fun to solve and pretty rewarding when you save the cost of a new piece of equipment!