Does this fuzz look right? Doesn't work. Spot the obvious?

[Squirrel Murphy]

Maybe you can spot my error? I just want to make sure I didn't make an obvious mistake before I rip it all apart and/or start replacing components.

I bread-boarded this NPN silicon fuzz circuit over at Small Bear, tweaked it up to my liking, and then soldered it up. Now I can't get it to work despite several nights fiddling.

Bypass works. When the circuit is switched on, the LED lights but sound stops going through. There is actually some sound -- very quiet. The fuzz pot seems to sort-of work backwards, and the volume pot does nothing. Circuit is quiet (no hum). I re-soldered several times and replaced suspect wires and connections. I'm confident with continuity. The battery is good too. Checked the transistor leg orientation two or three times and the two polar caps have negative butts out.

Q1 is a 2N4123 and Q2 is a 2N2222A. C1 was reduced to 1uf. I also had 100pf caps across R1 and Q2c-to-b but they are out until I can get it working. The other components are typical for a fuzzface. R4 was biased on the breadboard and hasn't been adjusted again.

Voltages are low or missing at several points in the circuit so something is wrong. Could be a bad component or bad soldering (I'm not great at it). I just want to make sure the layout logic is good before I test further because a rookie mistake is very likely. I don't have voltage readings at the moment. I can do that this week and post if needed. I also have photos of the bread-board version if you want to see it.

I wired the switch, DC power jack, LED, in/out jacks per this drawing. I tested this arrangement successfully on the breadboard.


Here's the board. I'm sparing you the solder side for the moment. 



What do you think?

[dwmorrin]

Voltages are low or missing at several points in the circuit so something is wrong. Could be a bad component or bad soldering (I'm not great at it).

This is logical: take voltages, find a strange voltage, which leads to discovering bad component or soldering.

I just want to make sure the layout logic is good before I test further because a rookie mistake is very likely. I don't have voltage readings at the moment.

What do you think?

Your real rookie mistake was not posting the voltages you observed "low or missing at several points."
You could have the craziest, most illogical layout, and it would probably still work if it agreed with the schematic.

[smallbearelec]

Hi--

I am glad that you breadboarded first and got to hear what the thing should sound like. You started with good parts, so you can be pretty sure that there is at least one error in your layout. As I emphasize in Beginner Dos and Don'ts

https://www.smallbearelec.com/HowTos/BeginnerArticle/BeginnerDos.htm

getting from the breadboard to a workable layout on perf or PCB takes planning. I do it with EAGLE CAD; a lot of people here use DIY Layout Creator. I'm guessing that you worked out your layout as you went. For future builds: You really should know Exactly where everything will go before you pick up a soldering tool.

I do show in another article

https://www.smallbearelec.com/Projects/FurFace/FurFace.htm

a perfboard layout for the complete circuit that was laid out in the breadboard tutorial. But that does not help you since you only built the FF section. Actually, that was a smart choice; the whole thing is too complex for a soldered build for a complete beginner.

If I were debugging this, the first thing I would do is unsolder the leads to the off-board parts. Then I would make a clean copy of the schem and set my meter for continuity. Look at what the schem says is supposed to connect, and test every connection point-to-point. Mark off with a highlighter as you go. You will find things that are not right. When the board is straightened out, connect input jack, output jack, pots and battery. When you get that much working, Then wire the stomper. Happy troubleshooting!

[MaxPower]

Voltage readings are probably be the easiest way to see what the problem is.

[slacker]

If the actual connections on your layout match what you've written and the counter clockwise lugs of both pots go to ground then your layout matches the schematic.

[duck_arse]

and we still need to see the solder side.

[Squirrel Murphy]

I'm guessing that you worked out your layout as you went.

Nailed me.

I drew the circuit several times, and did alot on the breadboard. Got to the point where I understood the layout so I thought "moving" it from BB to a perfboard would be simple enough. I took photos of the breadboard and if I have to, I'll move it all back there again. I thought this go without incident and was confident enough to move forward.

Incidentally, I want to thank you for the tutorial. I've done PCB kits before but it's more like "build by numbers" and I didn't learn much beyond the build process itself. I wanted to learn how these things actually worked so I decided to build from scratch and try swapping parts on the fly. It's definitely been a worthwhile experience.

If the actual connections on your layout match what you've written and the counter clockwise lugs of both pots go to ground then your layout matches the schematic.

They do. Thanks. This was the basic sanity check I was looking for. If I didn't cook a component, then it's probably my soldering.

Voltage readings are probably be the easiest way to see what the problem is.

OK. I took some measurements. I didn't measure any of the the off-board connections.

• 9V on power bus (R1/R3)
• 8V at R3/R4/C3
• 0V at C3/Vol pot CW
• 0V at C1(-) -- the signal side
• 1V at Q1c/R1
• 0.5V at Q1b/R2
• 2V at Q2c/R4
• 1V at Q2e/fuzz pot CW
• 0V at C2(+)/fuzz pot wiper
• everything connected to ground seems to be at 0V

I didn't take measurements when it was breadboarded and man do I wish I had. What I see above doesn't make sense to me.

and we still need to see the solder side.

I was hoping to avoid that embarrassment.

I can't work on this for another week, but I'll come back with some more photos.

Thanks guys. I appreciate the help.

[Keppy]

• 9V on power bus (R1/R3)
• 8V at R3/R4/C3
• 0V at C3/Vol pot CW
• 0V at C1(-) -- the signal side

Fine.

• 1V at Q1c/R1

1V is kind of low. Q2b should be connected here as well, so low voltage here might keep Q2 from turning on.

• 0.5V at Q1b/R2

Fine.

• 2V at Q2c/R4

Usually, R4 is used to trim this to 4.5v.

• 1V at Q2e/fuzz pot CW

This should be one diode drop (.5-.7v) below the Q1c/Q2b junction. Instead, it is exactly the same. Q2 is not turned on.

• 0V at C2(+)/fuzz pot wiper

Fine, if the fuzz pot is turned all the way down. At any other setting this will should read somewhere between the other two lugs of the pot (0-1v). If that isn't the case, your wiper is shorted to ground.

• everything connected to ground seems to be at 0V

Fine.

It looks to me like you have a problem with Q2. Check that:

Q2 is oriented correctly
Q2b connects to Q1c
There is no short between Q2b and Q2e

Good luck!

[LucifersTrip]

For voltages, you really only need to report Q1/2 E,B,C

"I bread-boarded this NPN silicon fuzz circuit over at Small Bear, tweaked it up to my liking, and then soldered it up."

Can you compare the voltages you just listed with the ones when it worked after you "tweaked it up"?

for reference, here's working voltages from ggg

Q1
Collector
1.4v
Base
0.6v
Emitter
0.0v

Q2
Collector
4.5v (without trimming, you should be kind of close, ie, 3.5 -6V)
Base
1.4v
Emitter
0.8v

[smallbearelec]

I drew the circuit several times, and did alot on the breadboard.

That's a right approach.

I thought "moving" it from BB to a perfboard would be simple enough.

NOT! As I think you now realize, board layout is HARD. R. G. Keen wrote a whole book about it. One does not "move" a layout from breadboard to perf. They are completely different media, and the board layout has to be thought out in terms of the enclosure in which it is to go. This is why I suggest that beginners always work from a layout that has been vetted by experienced builders.

All of that said, you probably can get your board working if you debug systematically in the way I suggested. At that point, if you want to continue doing your own layouts, back up and go through the pain of learning to use layout software. There's a long learning curve to achieving some independence, there are numerous steep hills and potholes along the way, and you don't always know how steep they are or where they fall. You have done some things right but gotten tripped up. Let us know how things go.

[PRR]

>> see the solder side.
> I was hoping to avoid that embarrassment.

Electrons "can" move through some very embarrassing joints.

But when they don't move the way you want them to, "embarrassing" really means "too messy to de-bug efficiently". Which seems to be where you are at.

I don't know what you have. But soldering IS a skill as much as string-plucking or chord-fingering. Practice where you won't be embarrassed. Tear apart an old VCR, de-solder parts from its PCB, then solder them back as neat as the factory did. (The factory hates bad joints; and if they are bad they need to find the fault fast or drown in rejects.)

Yes, soldering leads backside of perf is different from PCB work. But if you can make a GOOD joint EVERY time on PCB, then going to perf just adds the problem of getting the right this-goes-to-that connections (while peering through the perf with X-ray eyes).

[MaxPower]

I ran a simulation in Tina-ti for the hell of it. Pretty much what LT posted:

Q1
Base: .616v
Emitter: 0v
Collector: 1.65v

Q2
Base: 1.65v
Emitter: .999v
Collector: 4.69v (with Collector resistor set to 4k Ohms)

[Squirrel Murphy]

NOT! As I think you now realize, board layout is HARD.

Yeah. 

I'm ok learning things this way though. Stumbling is ok sometimes. But y'all make it look so easy. And I picked the simplest circuit for a reason.

you probably can get your board working

Let us know how things go.

I most certainly will. This wont beat me. I'm out of town at the moment, but I'll attack this when I get back. Great group of people here... Really boosts the confidence...

[Squirrel Murphy]

It looks to me like you have a problem with Q2. Check that:

Q2 is oriented correctly
Q2b connects to Q1c
There is no short between Q2b and Q2e

I think you nailed it. I either had Q2 in wrong, the legs were shorted, or my soldering was inadequate (or a little of all three). Q2b needs to be twisted under to fit the layout and I suspect a short. I did a removal and careful replacement and clean-up -- the circuit came alive. I added 100pf filtering caps across Q2b and Q2c and across R1. Also added a 1N4002 diode for power protection. Still works after further mods  . Had to tweak the bias back up to 4.5V from 4.0 (maybe I bumped it?).

I still have an intermittent connection somewhere and I get some crazy squealing sometimes (esp at max fuzz). Maybe  shouldn't' have used solid-core hook-up wire. Def need to shorten the off-board wiring when boxing up...

New measurements:

• Q1c = 1.6V
• Q1b/Q2b = 0.5V
• Q2c = 4.5V
• Q2e = 0.9V

I hope that using a 1.1uf input cap wasn't a mistake given that I don't have a tone control. I liked it better than the 2.2 in the breadboard.

I ran a simulation in Tina-ti for the hell of it. Pretty much what LT posted:

Looks like I'm pretty much right there.

Thanks all. Learning alot and appreciate the support...

[MaxPower]

Q1b/Q2b=0.5v

Is that a typo? Because I believe Q2b is supposed to be connected to Q1c, not Q1b. Q2b should be 1.6v (same as Q1c).

[Keppy]

Glad you got it working!

[Squirrel Murphy]

Is that a typo? Because I believe Q2b is supposed to be connected to Q1c, not Q1b. Q2b should be 1.6v (same as Q1c).

You are quite right. It's a typo. It's what I meant to type and not what I did type.