Build Your Own Clone Message Board

In most cases you cannot measure a resistor’s value while it is in the circuit. The circuitry in parallel with the resistor will throw your meter’s reading off. At the very least you need to un-solder one end of it to get an accurate reading.

_________________
"It’s your soldering."

I've heard this, but every other resistor and capacitor tested just fine on the board. It was just this one section, and as it leads into that troublesome 3rd pin, I figured it was related. But apparently not.

It was hard to understand what you meant with that marked up schematic, and the comment about a troublesome 3rd pin. Are you saying the output of one of the opamps is shorted to one of the power supply rails?

Sorry, let me try to break it down further.

1) The markup was something we noticed when investigating the schematic. We couldn't rectify the "proper" values we should be getting on the pins based on the way it was written. We noticed, eventually, that by switching those two sections, that suddenly everything from other peoples' working pedal values made sense. The markup is not a comment on what's going on with my board, but rather pointing out an error in the schematic.

2) With my pedal not working, the major issue seems to be that pin 3 is at a way higher voltage than it's supposed to be. The rest of the values are close, but not quite. I'm trying to chase down what the issue is. All the joints look good (I've reflowed them a few times, even enlisted one of my buddies who teaches a soldering class to clean and inspect the board). The components all test fine. I'm trying to understand how I can end up with too much voltage flowing to that 3rd pin and see if I can work my way up the schematic from there, but I'm at a loss.

Has anybody modeled this circuit? Digitally tinkering with values might be easier than trying to work them out on the board.

I would certainly inspect both sides of the PCB to check for any solder bridging or other shorting around the IC. But if that shows nothing & based on what you've already done/determined, I'd be suspecting that the IC itself may be bad, i.e. the power short may be internal.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

Still not sure if we are on the same page or not. Don't know how much you know about troubleshooting circuits, so I will go through possibly some extra detail to try to make sure we are talking about the same things:

The way to count pins is start at pin one, the solder pad on the bottom of the board should be square rather than round like all the other pins. Or from the top is it the one with a dot next to it, or if there is a mark straight on the end, it is the first pin counterclockwise from the mark.

Starting from pin 1, count clockwise from the bottom, or counterclockwise from the top to enumerate the other pins.

According to the schematic, pin-3 of IC-1 is connected to pin-7, so they should be at the same voltage. Is that what you get?
Also, the chip pinout can be seen here: http://www.me.umn.edu/labs/hmd/lab/elec ... /lm324.jpg
Pin-4 is the + power (from a battery or external 9v power supply) which should be around 9 volts or so. And Pin-11 should be at ground.

Also, from the schematic you can see that a few of the opamp pins are connected to something called 1/2V, which should be the same for each opamp connected to that, and which should be right around 1/2 of your 9volts power supply voltage. In addition, Pin-10 should be at a little above 1/2V, and Pin-9 should be at pretty close to the voltage on Pin-10.

The way opamps work, and there are 4 opamps in IC1 with each shown as a little triangle, is that both the + and - inputs should be at about the same voltage as each other, within a few millivolts. The output pins at the pointy end of the triangle should go to whatever voltage is necessary to make the input pins at equal voltages (within a few millivolts or each other, that is).

Hopefully, by now this is starting to make more sense. If you are counting pins correctly, then you should report the voltages you measure. And if you find something that doesn't make sense according to the information you have so far, then that should be investigated until an explanation is found.

EDIT: It sounds like from what you found as shown on the marked up schematic is that you think the pin numbers may be shown wrong on the schematic? If so, it would be good to verify that by checking the traces on the bottom of the PCB to see if the pin wiring goes where is shown on the schematic or if the wiring goes where your meter readings would suggest.

Hey Mark, thanks for the help! I am definitely new, so yeah, simplicity is KEY for me.

Here are my pin readouts:

1) 5.25
2) 5.13
3) 6.86
4) 8.53
5) 4.45
6) 5.2
7) 5.25

4.84
9) 4.54
10) 5.28
11) ground
12) 4.45
13) 4.45
14) 4.45

So no, 3 and 7 are not the same. But they also are not in Stephen's readings from his working Mimosa Jr. Here are Stephen's so you don't have to dig around to find them:



As you can see, the biggest difference seems to be on that 3rd pin. Other things are confusing too, but I'm trying to focus my efforts on fixing that one, big, glaring issue in hopes that the other differences may be cascading from there.

That pin reading is definitely helpful. Thanks for posting it!

Here is a picture of the component layout from the bottom: viewtopic.php?f=19&t=39389

There is also a picture of the bottom of the PCB without the labels on page 7 of the instructions. In that picture it looks like pin-7 connects to pin-10. So, I would say the schematic pin numbers are wrong.

Pin 9 appears to be the inverting input of the opamp that connects to R8 and R9. You should be able to disconnect the power and check this theory out with an ohmmeter. It should measure 100k ohms from Pin-9 to Pin-8, if so. That would be the resistance of R9. If you can confirm whether or not this is the case, that would be the next step. Confirm the schematic is wrong and that we need to relabel it before continuing. Either that or confirm that the schematic is correct.

If the numbers are indeed swapped then Pin-3 would presumably be connected to R4 and R5, so it might be good to check they are the correct resistors. They form a voltage divider that should give 0.547 times V+. Since your + voltage is 8.53, we would expect to see about 4.66 volts at pin-3. Since you are not seeing that, there would appear to a problem in that part of the circuit. You didn't say if you are measuring voltages from the top or the bottom of the board. It is possible to have a bad socket with the voltage at the IC pin on top different from the voltage on the bottom side of the board. So, it might make sense to try to verify what is and what is not correct. It should be easy to figure out between measuring and eyeballing circuit traces. It's much harder for me to go much further sitting here at a computer. So, please let us know what you find, of if you find I made a mistake somewhere in what I wrote (which is always possible too!). Check everything out thoroughly and you will find something wrong somewhere.

EDIT: If you do find Pin-3 is connected to R4 and R5 and they are the correct values, you could try disconnecting the power, remove the IC, and reconnect the power with the IC still removed from the circuit. At that point you should see around 4.66 volts at pin 3 of the IC socket. If the voltage is correct with IC removed, and incorrect with the IC in the circuit, then the IC must be causing the voltage to go too high (we would say pin-3 of IC-1 is "pulling up" the voltage at that node, which shouldn't be).

Dude, awesome! thanks for the insights. I typically measure from the top, because it's easier for me, but I did a quick run from the bottom of the board too, just to check. My values were the same either way,

R8 and R9 do, in fact, show 100k ohms of resistance between them.

I checked those two resistors on the board, and I was getting wrong values. Knowing that it could be different, I pulled them off the board and tested again, and they both are, in fact, working and resisting at the right amount.

I was not able to check the voltage at R3 with the IC out because I damaged one of my resistors while pulling it out, so I'm going to have to wait until tomorrow when I can get a replacement. I will report back soon!

Okay, but let's not confuse Pin 8 and Pin 9 of the IC socket for R8 and R9 which are two resistors. It's hard for me to tell if you are using the terms interchangeably, but just to be certain, pins and resistors are different types of things. I was suggesting measuring between pins 8 and 9 of the IC socket.

On the other hand, the concept of measuring between two resistors has ambiguous meaning. So, I'm not sure what you did.

I'm just really dumb. You'll have to forgive me. I meant the pins.

Okay, no problem. We seem to be communicating well enough to make some good progress.

Have you had a chance to check the circuitry around Pin 3? (which we think is incorrectly labeled as pin-10 on the schematic) That may be where the problem is, or right around there somewhere.

Yes. I checked R4 and R5. Off the board, they tested fine, but I broke a leg off R5 and I have to replace it.

Tomorrow, when I do, I'll see what happens to the voltage at pin 3's empty spot when I remove the IC.

If there is a little tiny stub of wire left on R5, you could solder on a longer wire. But, if the resistor body itself is cracked and broken then replacement would be needed.

However, if you look at the schematic the junction or node where R4 and R5 connect together has only two other connections, to C3 and to one of the IC pins. An ohmmeter can charge up a capacitor, but the capacitor still blocks DC current after it is charged up. Also, modern ohmmeters don't put out enough voltage to bias on a silicon diode. That's to make it easier and safer to measure resistance without removing components to do it. What it all means is that R4 and R5 could have had their resistances measured in place, without removal of anything. Just connect the ohmmeter across one of the resistors and wait until the display stops changing. Once it stops that means the capacitor is charged up and the meter is then showing the correct resistance. The only time it is necessary to remove a resistor to measure it is if there is a CD current path parallel with it. That's because an ohmmeter runs a small current though a resistor to measure the resistance. If there is another DC current path then current from the ohmmeter will flow there to and cause in incorrect reading.

What I sometimes do if I have to measure a resistor and it has to be disconnected from the circuit to do it, I just cut one lead of the resistor on top of the circuit board, then lift up that end of the resistor or lift the lead a little to make sure there is no connection, and then just measure the resistor with the other end still soldered in place. With one lead cut, there is no way for DC current from the ohmmeter to flow anywhere but through the resistor, so no need to remove both leads. After measuring, I solder the cut lead back together. It usually goes pretty quick, certainly more quickly than removing the whole resistor. It's also easier on the circuit board, since the holes don't have to be heated and cleaned out to reinstall the resistor.

Will do next time!

On the board, both R4 and R5 were testing at a significantly lower resistance than they were supposed to. Around 200k instead of 390k and 470k. The numbers weren't changing, they were solidly measuring at this low reading, so I removed them. Upon retesting, off the board, they were just fine.

In the future, I'll try the one leg method. I thought that might work, but didn't want to be accused of cutting corners later. Haha.

Okay, there was probably return path through the power supply. I should have thought about that first. Oh, well.

Anyway, since the resistors are correct, the voltage should be correct with the IC removed. If the voltage changes more than a few millivolts with the IC back in the socket, then the IC would appear to be bad.

Yeah, that makes sense. Those two resistors in parallel would have a nominal resistance of 213K, right in the ballpark of what our OP observed.


Agreed.

BTW, as was suggested, two of the quad opamp stages in the schematic are switched. I followed the traces from the PCB diagrams in the instructions and the labeled PCB HERE, and here's what it should be:

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

Does anyone have any tips for desoldering the IC? I've been at it for about 2 hours, and it still won't budge. I can see clean through the holes next to the pins, but it seems to really be stuck.

Removing a 14-pin IC that is directly soldered to the PCB is a tough job even for a seasoned builder. Done incorrectly, you can easily rip out an eyelet or sever a trace on the board. My advice would be to just clip all the legs with cutting dikes and remove them from the board individually, and replace it with a new one. It's not like this is a rare or expensive NOS IC--the TL074 is a widely available, inexpensive quad op amp. BYOC sells 'em for $1, but you can certainly find them elsewhere.

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

Man, there is not much room to operate in there. So far I've been unable to cut it loose. I do have another IC on the way, but I can't seem to get this old one out.

Small sharp diagonal cutters for use with soft copper only are available for fine detail electronics work. Usually, they can get between IC pins from the top. Investing in a pair can be wise if doing this type of work. Beyond that, some other information of possible help is outlined below:

If you have good quality solder wick, such as Chemwick, you can sometimes remove enough solder from pins from the bottom of the board to free them up completely. If not all the solder comes off, sometimes adding fresh solder helps add new flux and improves the quality of any remaining original solder. Once as much solder has been removed as possible, the pin can be checked by trying to wiggle it with a small screwdriver. If it wiggles, it is free and you can move onto the next one. If only barely still soldered on, sometimes it can be popped loose with the screwdriver. Once enough pins are unsoldered, sometimes one side of the IC can be lifted up which can help gain further access.

In some cases it may help to destroy the plastic IC package with diagonal cutters or a Dremel tool grinder. Once the IC is broken up into pieces, sometimes the remnants can be more easily removed.

If the pins have been cut off, there are at least a couple of good ways to clean solder and remaining bits of pins out of the PCB holes. The idea is to do it quickly and with minimal heat to avoid damage to the board. Added fresh solder can help if old solder is oxidized. Heating the hole and slamming the board down against the workbench often throws most of the remaining solder out of the hole. A possibly better and less violent method is to use a small locking pliers or locking tweezers to clamp onto the shaft of a stainless steel sewing needle of a diameter that just fits through a new, clean hole. Hold the sharp tip of the needle against the hole while heating the needle and hole at the same time with the solder iron tip. As soon as the solder starts to soften the needle will push through the hole. Then quickly remove the solder iron. At that point the needle can be twisted and wiggled out of the hole, since solder doesn't stick to stainless steel. Using this method clears out more room in the hole than any other method I know. It also causes less damage to the board.

Exactly. Here's what BYOC offers, but you can find these at any electronics supply store, whether online or "b&m": https://buildyourownclone.com/collectio ... al-cutters


These techniques will certainly work, but I think a more efficient means of removing residual solder from PCB eyelets would be with a "solder sucker" like this: https://www.amazon.com/WEmake-WM-SP4-So ... B0002KRAAG Use your iron to melt the solder from one side of the board, while placing the tip of this tool over the eyelet from the other side. Push the button to release the spring-loaded plunger, and the suction it creates sucks the molten solder from the eyelet very thoroughly. I do some pedal repair work on the side, and this is the standard tool that I use for component removal and board clean-up. Very handy!

_________________
“My favorite programming language is SOLDER” - Bob Pease (RIP)

My Website * My Musical Gear * My DIY Pedals: Pg.1 - Pg.2

I am picky about tools, so for small copper-only wire cutters I prefer Lindstrom 8142: https://www.amazon.com/Lindstrom-8142-8 ... B000C79I7W

Regarding solder suckers, I have tried most of them and don't really like any. But the best might have been Precista. Don't know if they are still around.

A temperature controlled solder station with a variety of tips is very helpful.

Maybe the best solder wick: https://www.amazon.com/Chemtronics-Deso ... B0019V5MAC

Unfortunately, having good tools can be expensive. The ones I have were acquired over a lifetime, and I try to take very good care of them since they can be hard to replace.

Still use leaded solder for my own work. It's easier to remove if needed and works very well at a lower temperature than the newer unleaded solders.

First of all, shout out to Mark for the "destroy the IC" suggestion. That was both fun AND effective. Also, incredibly therapeutic, considering all the trouble this thing has given me.

With the IC out, I am getting around 5 volts at pin 3, MUCH MUCH closer to the ideal. Once the new one comes in the next couple of days, I'll try it out and see.

Are bad ICs common? How can I test them before putting them in boards in the future? I'm hoping my next pedal goes MUCH more smoothly.

Bad ICs aren't common, but overheating them is. If you can, use sockets instead of soldering them directly to the board.


Sent from my iPhone using Tapatalk

_________________
Pedal building is like the opposite of sex. All the fun stuff happens before you get in the box.