germanium transistor orientation

[SJON]

Hello.

I am new here and I am new at building pedals. I'm currently building a 69-type Fuzz from a kit. It all seems to be wired up correctly and works fine in bypass mode. When the circuit is activated all I get is a very faint version of the bypass sound, I'm talking like 5% of the bypassed volume.

Now this may well be, of course, because I don't have the two germanium transistors connected and the circuit is presumably therefore incomplete. My problem is that I don't know how to orient them. As per the instructions I have mounted on the PCB two little 3-pin sockets like this:



Here is the PCB showing the markings for the transistors at Q1 and Q2, holes for 3-pin sockets visible:



And here is the representation of the AC122 transistors in terms of which leg does what:



So, my question is how do I know which leg of the transistor to put in which socket hole in each case?

I would be immensely grateful for any help.

Best regards,

Simon

[GGBB]

The obvious answer is by function. C, B, E - collector, base, emitter. The key is in knowing which is which. For the transistor, you should always refer to the manufacturer's data sheet - your drawing shows that the tab marks the emitter. On that PCB, the only indication is the tab. If you have written instructions for that PCB, refer to those, but otherwise assume that the tab marks the emitter on the PCB as well, and that base is in the middle. You could verify that by checking the traces on the PCB to see which other components connect to the transistor pads and comparing that to a schematic. But most likely here - tab to tab.

[SJON]

Many thanks, GGBB. I will try that.

Alas, despite an otherwise very well-presented kit, there are no instructions regarding this point.

The other thing that has me a little mystified is that the kit provides a ceramic 100pF cap (what we call a "lentil" in Spain) for the C5 position on the PCB, but on the actual assembly image it is not present. I put it in anyway.

Best regards and thanks again,

Simon

[PRR]

Yes, on old Ge parts you can bet the tab/dot is Emitter and be pretty sure the Base is in the middle.

> the kit provides a ceramic 100pF cap (what we call a "lentil" in Spain)

"Bean". In American we call modern $1/10 transistors "jelly-bean", which is a different food (a candy) but the same joke. However where a friend used to work, ceramic caps were called "crunchy-caps" because they broke with a "crunch" if you handled them too rough (or stepped on them).

The cap is probably "optional". Leave it out. If the sound is too bright or shrill, put it in. (Just a guess without knowing the specific plan you are working from.)

[SJON]

Hello again.

Well, The germanium fuzz is built but has certain issues. It works, but there is a considerable volume decay, especially on higher strings, after a couple of seconds of playing a note or chord. Also very low output when compared to a Fulltone 69.

The guy who checked it over for me (an engineer, electricity not his speciality, but a knowledgable guy nevertheless pedal-wise) suggests that Q1 is receiving insufficient voltage and should perhaps be preceded by a 1K resistor and a variable resistor in series  in order to adjust voltage to the particular characteristics of the transistor. Either that or the transistors are defective, in which case I can see myself trying to track down a guaranteed matched pair or converting the whole thing to silicon.

I did e-mail the supplier, but they are on holiday and their forum is down.

Once again, any advice would be humbly appreciated.

Best regards,

Simon

[GGBB]

We would typically want to see a schematic of the circuit, and ask you to post the voltages at each of the pins of both transistors as well as your supply voltage. If you can, please do. Additionally, the PCB has "RV1" which I'd assume is a trimpot (variable resistor). In fuzz faces this is usually to set the bias of one of the transistors (usually Q2) in order to get the sound right, and if not set properly can result in problems similar to what you have described. Have you adjusted it? Do you have instructions for what you are supposed to with it?

[Mark Hammer]

I recommend using your hfe tester on your meter to confirm pinouts.  I try to have datasheets for all semiconductors I use, but have to admit that the diagrams are often ambiguous for me.  So I confirm pinout by sticking the transistor in the socket and looking for plausible hfe readings.  Assuming the transistor is good, flipping leads around will eventually provide a plausible reading.

Alternatively, if your meter has no transistor socket, but does allow you to measure diode forward voltage, you can identify whether something is NPN or PNP, and at least identify which pin is the base, by measuring Vf between the  various pins.

[Electric Warrior]

Yes, on old Ge parts you can bet the tab/dot is Emitter and be pretty sure the Base is in the middle.

The dot usually marks the collector.

[SJON]

Hello.

I can't check the readings now. My friend did so and everything seemed correct. I neither have a working multimeter nor know how to do that.

Is it permissible on this forum to give a link to the supplier's circuit? It is open on their website and can be downloaded free by anybody. A German company.

Best regards,

Simon

[SJON]

Also, we did vary Q2's trim pot but the problem doesn't seem to be there.

[GGBB]

Is it permissible on this forum to give a link to the supplier's circuit? It is open on their website and can be downloaded free by anybody. A German company.

Yes - that's fine.

[PRR]

> the supplier's circuit?

I suspect here:
https://www.musikding.de/The-Face-69er-Fuzz-kit

[SJON]

You may well be entirely right.

[R.G.]

The obvious answer is by function. C, B, E - collector, base, emitter. The key is in knowing which is which. For the transistor, you should always refer to the manufacturer's data sheet -

Yes - as I violated myself recently. 

However, as a rough and ready indication of pinout, one can set one'd multimeter to the 100's of ohms range and measure resistances between pins. The trick is that bipolar transistors look like two diodes with the anodes (for NPN) or cathodes (for PNP) tied together at the base pin. So your multimeter lets you measure resistance in both directions between all pins, and from that tell which pin is the base, and what polarity the transistor is. The collector and emitter both conduct to the base on one polarity, and show open in the other polarity. If the base pin is positive for this to happen, it's an NPN. IF the base pin is negative for conduction, it's a PNP. Collector and emitter don't show conduction in either polarity on most meters.

That leaves only the problem of which is collector and which is emitter. On modern transistors, the collector-base and base-emitter junctions are not identical; the collector base "diode" is optimized for holding off voltage. The base-emitter is optimized for best current flow in through the base and highest gain; this results in a base-emitter junction with a quite-low reverse voltage rating - generally 5 to 7V in modern devices. The difference in the junctions also makes the reverse-mode current gain much smaller than the normal mode current gain.

It was not always this way. Early germaniums had identical collector-base and base-emitter junctions. However I suspect later germaniums had more asymmetrical junctions too.

[PRR]

> Early germaniums had identical collector-base and base-emitter junctions.

Yes.

And IF they are essentially identical, you only need to identify the Base (your multi-meter test). Get that right, let C and E fall where they may.

Ah, the variability of early Ge processing is such that, unless you got chopper-rated parts, the gain will be slightly different one way or the other. And many old-Ge guitar effects are over-fussy about that. So I guess after you get it working one way, you should try swopping C and E and try the other way.