My Fuzzface requires a signal stronger than my guitar...

[Seljer]

The transistor test on multimeters is more or less worthless for germaniums. You can however, use the diode test to figure whether it is a PNP or NPN as well as figuring out the pinout

To test the gain and leakage you have to rig up a little test circuit
http://www.geofex.com/article_folders/fuzzface/fftech.htm#Picking%20transistors
https://www.smallbearelec.com/HowTos/FuzzFaceFAQ/FFFAQ.htm (go down to question 26)

I'm usually faster at desoldering small components than soldering them. You should be doing things hastily when soldering in any case. As mentioned above, heat sinking can help, put an alligator clip on the top of your the transistor to soak up the heat.

[joakim]

I've desoldered the transistors, and am trying to test the using my dmm. Since it doesn't appear (to me) to have a dedicated diode check, I'm using the "2000" resistance measure which also has a diode symbol on it (see picture below).

I'm assuming the transistors are PNP, and that the red dot indicates collector (such that the adjoining, "middle" leg should be the base). Here's a matrix of the resistance readings I get
(the top row indicates the leg which has the red, positive dmm lead attached to it):

Q₁:

	+C	+B	+E
-C	--	8.3k	1.0k
-B	6.5k	--	7.1k
-E	1.0k	9.4k	--

Q₂:

	+C	+B	+E
-C	--	10.9k	1.2k
-B	7.7k	--	8.4k
-E	1.2k	11.7k	--

For PNP, I should expect a high resistance when the positive lead is clamped to the base, and the COM lead is touched to the collector and emitter... and this seems to be the case (?)
However, I also get a high resistance when the leads are switched around (negative to base, positive touching C and E)... which would indicate NPN.

If I've gathered the necessary information here, can someone please help me along?

Ps. here's how I'm using my modest dmm:

[duck_arse]

joakim, always always always never assume. the interwebs means you have no excuse for not looking up the datasheets for the parts you are using. collect the whole set! save them on your hard-drive, and draw the pin-outs for each type transistor you are using on all your circuit diagrams. it won't stop you going wrong, but will increase your going rights.

I'm sorry I can't help you w/ multimetering yr parts, I'm hopeless with that particular transistor test. when you got these parts at uni, were they sitting in a drawer unused, or were they soldered into a board of some description?

[joakim]

joakim, always always always never assume. the interwebs means you have no excuse for not looking up the datasheets for the parts you are using. collect the whole set! save them on your hard-drive, and draw the pin-outs for each type transistor you are using on all your circuit diagrams. it won't stop you going wrong, but will increase your going rights.

I'm sorry I can't help you w/ multimetering yr parts, I'm hopeless with that particular transistor test. when you got these parts at uni, were they sitting in a drawer unused, or were they soldered into a board of some description?

I've tried finding the datasheet, but was only partly successful (see first post). These components are quite old.
Also, the point of the measuring is also to identify the pinout...

They were lying in a drawer, yes, unused.

[duck_arse]

orrright. I missed your link in the first post.

the datasheet linked (is correct for yor part) shows the answer to yr questions. the "polarity" colomn shown "p" = pnp or "n" = npn. the "pack outline" colomn shows "TO-1" for your part. any transistor in a "TO-1" package should conform to the pinout and dimensions. in this case, with the pins facing towards you, as shown in the diagram, "1" is emitter, "2" is base, "3" is collector.

the TO-5, 8, 37 etc all show the same pinouts. when the plastic encapsulated silicon transistors came along, this sytem went all to hell.

I didn't know the dot colour indicated anything.

[dwmorrin]

Do you have the manual for that meter?  Might help you interpret the results.
You shouldn't get a similar reading both ways... and the CE readings should be high; they're the lowest in your case.
Only 2 conclusions come to my mind: the test is invalid for some reason, or you've got 2 bad transistors.
Might be time to put these 2 aside for now and try some others in the circuit.
Practice transistor testing on new ones, and then go back to the mystery set.

[pinkjimiphoton]

i tell ya, sockets. ya won't hurt anything.

you can't always tell with a ge which is which... cuz truth is, in some cases, it doesn't matter. seriously.

you can reverse beta them (which may or may not be reversed) and they'll work just fine.

the red dot from what i've read is collector. cool. ya know what c is.

now all ya have to do is see which way sounds better.... if it sounds farty one way, and good the other, the good orientation is the one to go with.

power down between swaps, and you're fine. yes, some people will say that you can damage the transistor, but i doubt 9v will hurt them

even looking up the data sheets doesn't always help,  sometimes the parts are opposite what the sheet says.

this may prove helpful to ya bro:

http://www.rfcafe.com/references/popular-electronics/understanding-transistor-circuits-aug-1959-popular-electronics.htm

seriously.

even sil sometimes is backwards pinout, or sounds better with the reverse beta

[joakim]

Refering to the manual:


I set the meter to the "200k" range, not the "2000" range, to get the readings above. I though the values would be in ohms, but reading the manual (cough cough), it seems to me that the values are actually in mV? Even though the dmm is set to the "resistance range"   I guess voltage drop is what to measure when testing a transistor, but how come I shouldn't use the "DCV" range?

Always another puzzle, which is what I love about this hobby  

pinkjimiphoton: Yes, I've thought about using sockets. The PCB actually got a bit damaged when I desoldered the transistors... The plan now is to test the transistors, then test the circuit on a breadboard (which I should have done in the first place, before committing to the PCB -- I just got eager), and then fix the PCB or make a new one. Maybe with sockets! Very odd how the transistors can be rotated   But I did have some success with the circuit as it was, where the transistors apparently worked in reverse beta.

[dwmorrin]

I set the meter to the "200k" range, not the "2000" range, to get the readings above. I though the values would be in ohms, but reading the manual (cough cough), it seems to me that the values are actually in mV? Even though the dmm is set to the "resistance range"  I guess voltage drop is what to measure when testing a transistor, but how come I shouldn't use the "DCV" range?

So retest with the 2000 range... that's the range with the diode symbol next to it, right?
The meter applies a small voltage for both resistance and diode checks.
The voltage setting applies no voltage from the meter.
Diode readings are always read as voltage.

That is good news.  Hopefully, your transistors may be OK after all.

[pinkjimiphoton]

dude, i'm confused...

you need to be reading the voltage i think, not trying to read the current (reading across a resistor)


what we need is voltages to ground, probably the 20v range on your meter.

stick your black lead to battery ground, and your red probe, read the voltages at e/b/c of each transistor.

if you can post the voltages it will be very helpful. generally, if the pedal works with the transistor flipped 180 degrees, that's the problem.

collector voltage should be highest, base must be slightly lowe, emitter should be lower still.

[joakim]

Here are the readings again, using the "2000" range:

Q₁:

	+C	+B	+E
-C	--	"1"	*
-B	145	--	148
-E	"1"	"1"	--

Q₂:

	+C	+B	+E
-C	--	"1"	*
-B	143	--	146
-E	"1"	"1"	--

"1" is what the dmm shows when the reading is off the chart (relative to the 2000ohm setting, which is also apparently the diode check). This is why I increased to the 20k setting in the earlier post.
* means the reading started at some high-ish value (somewhere between 1000 and 2000) and started dropping steadily. I lost patience after around two minutes, when it had reached below 200.

pinkjimiphoton: I'm not sure what you're asking. I tried to read voltage (using the 20 DCV range of the dmm), with the COM probe connected to a 9V battery negative, but the dmm gave no reading for any of the lugs (just reported "1"). I'm not sure how I would expect to get any reading like this though, since I understand a voltmeter should be connected in parallel to what's being read, right? I must have misunderstood your question, since this has the voltmeter (dmm) in series with the transistor lug... 
I did a reading of the transistors in-circuit earlier, under a ~9V supply:

Q₂
C: 0.67V
B: 8.61V
E: 8.63V

Q₁
C: 8.60V
B: 8.64V
E: 8.72V = which is also VCC (using a standard 9V battery)

[dwmorrin]

That's great.  Your transistors are OK.
Your getting 140-150mV diode drops which is good for Ge transistors.
So it's probably not the transistors.

And your pinout appears to be correct.  The mV for the BE should be higher than the BC.  Everything looks right.

[pinkjimiphoton]

forget the resistance reading bro.

the voltages tell much. q2 is installed backwards, and the bias of the transistor is way too high.

q1  is also backwards. you need to flip both 180 degrees. you have the emitters and collectors switched on both units. your emitter should be at or near ground potential, not close to B+. not only are the transistors in backwards, they are biased close to or at cutoff...that's why only loud signals pass thru.

so... first things first,... reverse the transistors  by turning each 180 degrees.

then check your voltages again, and repost your voltages.

then we'll adjust the resistors from b+ (or negative, in this case) to collectors.

your voltage ratios look ok, but the overall voltage for a fuzzface should be more like 4.5-6.5 volts, not 8.72.

hollah back bro. we'll finger this out.

[Electric Warrior]

the voltages tell much. q2 is installed backwards, and the bias of the transistor is way too high.

q1  is also backwards. you need to flip both 180 degrees. you have the emitters and collectors switched on both units. your emitter should be at or near ground potential, not close to B+. not only are the transistors in backwards, they are biased close to or at cutoff...that's why only loud signals pass thru.

so... first things first,... reverse the transistors  by turning each 180 degrees.

No. Q1 e is not supposed to be at ground potential. Supply voltage is what he needs to measure there. Look at the schematic he's using. And turning the transistors around would certainly not fix this. He identified the collectors correctly.

[LucifersTrip]

op, if you're spending this much time, you may as well do the classic test with just a few components
http://www.geofex.com/Article_Folders/ffselect.htm

there are many other simple testers that you can experiment with, also
http://tinyurl.com/lf9cnbp

[pinkjimiphoton]

the voltages tell much. q2 is installed backwards, and the bias of the transistor is way too high.

q1  is also backwards. you need to flip both 180 degrees. you have the emitters and collectors switched on both units. your emitter should be at or near ground potential, not close to B+. not only are the transistors in backwards, they are biased close to or at cutoff...that's why only loud signals pass thru.

so... first things first,... reverse the transistors  by turning each 180 degrees.

i said at or near. being the emitter should be lower voltage than the collector, not higher.

No. Q1 e is not supposed to be at ground potential. Supply voltage is what he needs to measure there. Look at the schematic he's using. And turning the transistors around would certainly not fix this. He identified the collectors correctly.

[dwmorrin]

Going back to the original info,
Q2 in circuit only had a 20mV BE drop... as you just verified out-of-circuit that the BE of this transistor should be ~140mV, so Q2 is definitely cutoff in-circuit.

The 100k feedback resistor only has 10mV drop across it.  This doesn't look right either... the "good" voltages posted earlier were in the 100s of mV.

In your pictures, the 1k pot is just drawn in.  Can you verify it is hooked up correctly?  Maybe a pic of it actually connected?  Verify with resistance check that is actually 1k, and that Q2's emitter is 1k from your +9?  (Would be ground on other schematics)

[Electric Warrior]

i said at or near. being the emitter should be lower voltage than the collector, not higher.

No way. You're still not getting the mod. This is not a regular fuzz face schematic.

[pinkjimiphoton]

you're right, just looked at it. weird.

[joakim]

Going back to the original info,
Q2 in circuit only had a 20mV BE drop... as you just verified out-of-circuit that the BE of this transistor should be ~140mV, so Q2 is definitely cutoff in-circuit.

The 100k feedback resistor only has 10mV drop across it.  This doesn't look right either... the "good" voltages posted earlier were in the 100s of mV.

In your pictures, the 1k pot is just drawn in.  Can you verify it is hooked up correctly?  Maybe a pic of it actually connected?  Verify with resistance check that is actually 1k, and that Q2's emitter is 1k from your +9?  (Would be ground on other schematics)

Here's my project; a generalised stompbox, where I've extracted everything but the "effect specific" components out into the casing. The effect can then be placed on a PCB, and inserted in the slot following the pinout indicated in pic in my initial post. I.e., port 3, 4, 5 go to pot 2 (1k linear), and 6, 7, 8 go to pot 1 (500k log), as seen here

(click for larger)

The big idea is (was) to be able to make simple effect cards, without having to re-do the big job of making a new box. This was a gift to a friend of mine, christmas of 2006... But (as we now know too well) it did not work when he tried it. I'm picking it up again now, because it bugs me that I never finished it

I have checked with the dmm that pots 1 and 2 are indeed 500k and 1k, respectively, and that their lugs are connected to the correct ports in the slot. Also, finally, that everything actually connects from the port and onto the traces on the PCB -- this including VCC, GND, input and output. All that should be fine.

The DPDT is configured in a "Millenium Bypass", with an indicator LED.

However, since that small "sub-circuit" is indeed physically disconnected from the input -> effect -> output, when the effect is stomped in, I thought I'd simplify the op by not mentioning all this ;-) But, since it came up ...

Anyway. Electric Warrior's questions: I have checked that the Fuzz pot is indeed 1k, and connects to the PCB as intended (and shown in the schematic, although the schematic says "100k"). Similar for the 500k volume pot.
But how do I measure/calculate the voltage at Q2's emitter? Ie., should tghe fuzz pot be turned fully on (to 1k)?