Yet another fuzz face question

[memphis88]

Greetings to all DIYers here :-p
I'm new in making stompboxes (and new to this forum as it comes) and I'd like to ask for some help with my first project. So, as the topic says, I finished a few days ago one of the various clones of Fuzz Face. Just to let you know, I'm not any expert in electronics. I understand terms like "current" or "voltage" and I barely understand ohm's law. So about my problem, I've followed this schematic bellow to make the FF:

The only differences about my build is:
1)instead of the MPSA18 transistors, I used BC108C on both Q1 and Q2.
2)instead of 43k resistor I used 1 33k and 1 10k and out them in series (to get 43k) cause I couldn't find in market

So the issue is that when the pedal is working, the output volume drops down to lower than half the pedal is off (true bypass).
I've been really curious what the problem really is, if it's transistors or some connectivity (which I doubt so far) or maybe my power supply so I've written down the following data just in case they are handy:
My power supply isn't of the best qualities so it's output voltage is 9.8V
With this power supply I've measured the transistors' voltages and here they are:
Q1
C:0.99V
B:0.57V
E:0.00V
Q2
C:9.18V
B:0.99V
E:0.59V

I'm sorry if such a question is already answered but I've searched this forum lots of times and didn't find an answer (this place is huge actually). Any help is highly appreciated! Thank you for your time.

[brett]

Hi
those are exactly the data we need to dignose a problem.
For some reason transistor 2 isn't turned "on". (the base voltage is 0.6 to 0.7 V higher than the emitter in a working silicon transistor).
The base voltage of Q2 is loo low (it should be 1.3 V).
Check out that 43k resistor again. It's not 330k+10k or 33k+100k is it?
good luck

[memphis88]

Hi
those are exactly the data we need to dignose a problem.
For some reason transistor 2 isn't turned "on". (the base voltage is 0.6 to 0.7 V higher than the emitter in a working silicon transistor).
The base voltage of Q2 is loo low (it should be 1.3 V).
Check out that 43k resistor again. It's not 330k+10k or 33k+100k is it?
good luck

Well I did check the resistors again and they are correct :/ (checked both on multimeter and on color code). Maybe 43k is too much resistance for Q1? Cause Q1's collector voltage is Q2's base voltage actually. On the other hand, seeing this schematic again and again on internet makes me wonder if it's wrong, I don't think that's the case.

P.S. Thank you for the quick reply !

[smallbearelec]

Once you get this working, (and you will,) you will likely find the tone kind of harsh. That's because the MPSA18 and the BC108C are very high-gain devices, and the silicon Fuzz Face sounds waaay better with lower-gain parts that are correctly biased. But you should still be able to get it working as shown.

The collector voltage on Q2 is much too high; it should be closer to half of the supply voltage. Most likely this is because of a wiring error. Set your multimeter to measure continuity and test to make sure that every point that the schematic says should be connected actually Is. Mark off with a highlighter as you work. Two of my articles that might help:

http://www.smallbearelec.com/HowTos/BreadboardSiFF/BreadboardSiFF.htm
http://www.smallbearelec.com/Projects/FurFace/FurFace.htm

Let us know how it goes.

Regards
Steve Daniels

[Joe Hart]

The transistors seem misbiased. Change the boas resistors to trimpots (or full-sized pots) and try to dial up voltages that are closer to "ideal." Do a search. There is a TON of info on how to do this. Good luck and welcome to the forum!
-Joe Hart

[memphis88]

Big thanks for all the assistance. I'm assuming that the connectivity is ok cause I made the circuit on PCB on my own. Here's a screenshot of the layout from eagle (the values are missing but the positioning is just like the schematic from my first post w/ the change on the 43k resistor)


As for the transistors, I already know that the sound may be harsh but it's not up to me. I don't even play guitar actually, I made this for my brother, that's the pedal he asked for.
So as a summary, what I understand so far is that transistor's voltages are wrong. Could you please tell me approximately what voltages should be more like it for the transistors?

P.S. Great work on those articles Mr. Daniels!

[AnalogPackrat]

Q2 E and Q1 B are almost the same voltage...perhaps your 68k feedback resistor is actually a much lower value (or perhaps there's a wiring error there).  Also, did you double check your transistor pinouts?  They may not match the 2N5089s on your layout.  If they are TO-18 metal cans, the tab is closest to the emitter.

[smallbearelec]

Typical voltages from the Fur Face:

    Collector Base Emitter
Q1  1.56     .65       0
Q2  4.93    1.56     .93

You Can't presume that continuity is there simply because you made a board! It's very common for a self-made board to have a trace that looks solid and really isn't. Measure!

[memphis88]

I've been checking for an hour now and can't find anything wrong with the connectivity. Transistors have exactly the expected values of resistance on their tips, the continuity is also ok. I really dunno what more to say, I'm starting to give up my hopes. The most annoying thing is that the circuit is actually working but it has low volume...

[Joe Hart]

Transistors have exactly the expected values of resistance on their tips

"Resistance"? You cannot measure resistance in circuit. Did you mean "voltage"? And if so, you didn't originally have the correct values. Have you sorted that out?
-Joe Hart

[AnalogPackrat]

Did you socket your transistors or solder them in?  If they're socketed you can do some simple tests.  Pull the power and unplug input and output cords.  Pull the transistors.  Now look carefully at the schematic and imagine that each of the transistors is now three unterminated points in the schematic.  You can measure resistance between many points in the circuit to look for unintentional shorts or opens.  With capacitors on some nodes you will have to wait a bit for the measurement to stabilize (your meter will be charging up the cap(s) in these nodes).

On Q1 there is no DC path between any of the pins.  Measure resisitance between each--they should all be open circuit.  No look at Q2.  There is a DC path between C and B which is 10k + 330R + 43k = ~53k.  Do you get about 53k between C & B on Q2?  C-E and B-E should be open circuit on Q2.  Now measure between Q1 B and Q2 E--it should be 68k.  Check the path from the 9V supply input to Q1 E (should be 470R + 43k--will take a while to stabilize as you have that big filter cap on the supply rail).

You can also use a magnifying glass or loupe to look for solder shorts--I've been burned by this one many times when working on perf or PCBs with no solder mask.  Just remember, physics doesn't lie.  Whatever is happening is happening for a logical reason, you just have to figure out why the result is not correct.

AP

[memphis88]

Did you socket your transistors or solder them in?  If they're socketed you can do some simple tests.  Pull the power and unplug input and output cords.  Pull the transistors.  Now look carefully at the schematic and imagine that each of the transistors is now three unterminated points in the schematic.  You can measure resistance between many points in the circuit to look for unintentional shorts or opens.  With capacitors on some nodes you will have to wait a bit for the measurement to stabilize (your meter will be charging up the cap(s) in these nodes).

On Q1 there is no DC path between any of the pins.  Measure resisitance between each--they should all be open circuit.  No look at Q2.  There is a DC path between C and B which is 10k + 330R + 43k = ~53k.  Do you get about 53k between C & B on Q2?  C-E and B-E should be open circuit on Q2.  Now measure between Q1 B and Q2 E--it should be 68k.  Check the path from the 9V supply input to Q1 E (should be 470R + 43k--will take a while to stabilize as you have that big filter cap on the supply rail).

You can also use a magnifying glass or loupe to look for solder shorts--I've been burned by this one many times when working on perf or PCBs with no solder mask.  Just remember, physics doesn't lie.  Whatever is happening is happening for a logical reason, you just have to figure out why the result is not correct.

AP

Ok that was really helpfull there. I pulled out the transistors to make the tests, although they were soldered (desoldered ofc). I noticed 2 wierd things while measuring what you told me above:
1) on Q1 there's a resistance between B and E about 69k
2) when tried to measure resistance between +9 pin to Q1 E at the scale of 200k the multimeter started to count until it reached it's limit so I changed the scale to 200M and it stopped at 1.7M
The rest of the results was as you said.

I'm not sure about the 1st measurement, in the beggining I thought it was correct but according to your post it's not. Mind explaining to me? As for the second one, I can't tell really if it's correct or not so I'll trust your words. I'm starting to feel that the problem is at the schematic I made on eagle to make the layout after, if someone wants to give it a look:

[AnalogPackrat]

I screwed up on the Q1 e-b thing.  There's a path through ground which goes through the Q2 emitter resistor (1k) and the feedback resistor (68k) so your 69k is right on.

I had a typo on the other one (apparently I need more coffee)..I meant to say +9V to Q1 c (not e as I had typed).

So far you have eliminated several potential problems.  Your feedback resistor is correct as is the Q2 emitter resistor (pot).  If you other measurements were good then maybe you've got a bum Q2 or have the pinout wrong.  If you can find sockets I would use those for the transistors so you can swap at will.  I use the inline machine pin type and snip them to length.  Looks like your layout has inline pads for the transistors with 0.1" spacing so you could do the same.  I don't know where you are located, but for reference go to the Futurlec web page and search for part # "SIL30" for a picture.

A P

[memphis88]

I screwed up on the Q1 e-b thing.  There's a path through ground which goes through the Q2 emitter resistor (1k) and the feedback resistor (68k) so your 69k is right on.

I had a typo on the other one (apparently I need more coffee)..I meant to say +9V to Q1 c (not e as I had typed).

So far you have eliminated several potential problems.  Your feedback resistor is correct as is the Q2 emitter resistor (pot).  If you other measurements were good then maybe you've got a bum Q2 or have the pinout wrong.  If you can find sockets I would use those for the transistors so you can swap at will.  I use the inline machine pin type and snip them to length.  Looks like your layout has inline pads for the transistors with 0.1" spacing so you could do the same.  I don't know where you are located, but for reference go to the Futurlec web page and search for part # "SIL30" for a picture.

A P

Yeap the resistance between +9 and Q1 C seems correct also (~43k). Think I'll put another transistor on Q2 since I have some spare. Btw something else that keeps bothering me for some time is that my first build was with BC547. With those the pedal didn't sound at all but when I changed to 108C then I got a nice output but on low volume. I'll make the changes and update. Oh about the pins, I've looked the datasheet several times to confirm and it has to be correct (I'm getting sound after all). If nothing changes I may consider swapping the pins after all, I don't think it will get damaged.
P.S. I'm from Greece, Athens

[brett]

Hi
modern transistors give some gain when reversed (ie swapping emitter for collector). Not a lot (e.g. hFE=5 instead of 500)
cheers

[memphis88]

Unfortunately, I only have bad news. I did some experiments with the transistors:
1) Tried at first again with BC547 just in case -> No sound at all (didn't have the courage to measure voltages)
2) Tried 2 new BC108C -> same effect as before, cool output sound but with low volume
3) Switched the pins just in case I read the datasheet wrong and placed them wrong from the very beginning -> no sound at all, just a little bit with the amp on max volume and it didn't sound fuzzed actually

The worst (or maybe the best?) part from all of this is that the board got damage on the transistors' place cause of all this soldering-desoldering and I can't experiment on it any more (it's not worth the time spending to make the connections). I said maybe the best cause I'm thinking of giving up this schematic and try a simpler fuzz face, dunno which one yet.

I only have 2 last questions if someone doesn't mind answering.
1) Is there any chance that the original schematic was wrong?
2) Is there any chance that the original schematic would only work with MPSA18 transistors? (I'm mostly asking this because I got output with the BC108C which have higher gain from BC547 and AFAIK MPSA18 has even more gain)

I'm sorry for all the trouble, I'm really gratefull for the help.

[AnalogPackrat]

First, no need to apologize.  Everyone has gone through similar experiences.  Bit of free wisdom from a 45 year old engineer: you usually learn the most from your mistakes.  So, take a deep breath, maybe step away from electronics for a day or two, then come back and think about what happened and what you can learn from it.  For instance, next time would you start out with a board layout/etch and solder before even hearing what the circuit sounds like?  If not, what are the alternatives?  Rather than trusting that the pinout is what you think it is, is there a way to test the transistor (non-destructively) to verify it?

As for your questions:

1) it looks OK to me.  Most likely it was a problem with your board, your soldering, a component, or some combination.
2) other than the pinout differences, the circuit should basically work with any NPN Si transistors.  It might not sound very good with some types, but it should work.

My advice would be to invest in a decent solderless breadboard and a few feet of different colored 22AWG solid core wire.  With this set up you can do a lot more experimentation before committing to a specific circuit.  Changes are essentially free.  I don't know what sources you have in Greece for components and equipment, but I usually find that its cheaper to buy multiples of something if its a common part.  Transistors in quantity 10 are cheaper than 1 and 100 are even cheaper.  Same with caps, resistors, etc.  That's assuming you think you're into this hobby for the long term, of course.

Whatever you do, don't give up.  Reset your expectations (rarely do things like this work the first time, especially when you're new at it) and give it another go.

A P