Testing gain of PNP germanium.

[Jamdog]

I've read up on geofex how to measure germanium PNP gain,  but trying to do it seems to not be working.

I did the schematics with a 2.205M resistor (measured)   and a potentiometer set to 2.475K, and all four germanium I tested had these results: about 9.5mV without pressing the switch,  and about 2.5mV when activating the switch,  plus heat.
Do I understand negative gain?


Checked and double checked the circuit,  all is OK..  it's like this:





In doubt,  I also tried an NPN transistor (doesn't work)  and a PNP "backward"  (inverting emitter and collector)  (identical results)

I don't believe I can have negative gain transistors? 
The battery reads about 8.7V without a load,  and is connected the proper way around.

What can I have missed?

[Jamdog]

I may add,  putting any of the tested transistors in my known working fuzz face leads to a non-working fuzz.  At least two of those had been previously tried as working in there.

I think I just trashed half my stash of vintage germaniums,  trying to rate them.

[Ben Lyman]

I have only ever used Steve Daniels' technique which appears to be about the same, except easier for me to wrap my own brain around. Check it out here, it's #24 on his FAQ page:
http://diy.smallbearelec.com/HowTos/FuzzFaceFAQ/FFFAQ.htm
it also took me a few minutes to understand how to set my DMM, I think mine says something like 2000u (or backward u for "micro")

[Jamdog]

Thanks Ben!

Setting it for current makes more sense than for voltage, I agree with you fully there.

But with four of my eight transistors now toast,  I am not sure I want to do further testing...   I only have enough left for one other fuzz face

I am pretty sure the issue was heat.  They still read the same on the diode scale,  maybe they'll still work after a long cooling period?
Why did they heat up almost instantly when applying current at the base is beyond me.  Maybe 9V is too much for those transistors or something. 
It's sad though.

[robthequiet]

I think I see where you have the ground side of the trim pot and the 2.2M resistor but not clear on where the ground is in relation to the trim pot. The 2.2M resistor connects to a black lead at the other end that goes off the bottom of the pic. I'm just trying to relate the breadboard to the schematic at Geo.

[hellwood]

it looks like you have your negative wire going directly to the switch, bypassing the 2.2M. When the switch is ON, you are smoking your transistors from having too much current

[Jamdog]

Yeah,  I you're right.  I placed batt neg on the wrong side of the 2.2M.
Strange is,  I rechecked the schematics a dozen times,  particularly when the first transistor got hot.


Are the transistors toast or just temporarily incapacitated?

[Finom1]

You could take a look at this; http://www.amazon.com/Atlas-ADVANCED-SEMICONDUCTOR-ANALYZER-Tracing/dp/B00O0BMNX2

[Jamdog]

Tried on my fuzz,  all are dead.  They only work on super high amplification.

Bummer.

[duck_arse]

those resistors will protect yr trannies from excess current when correctly wired. DON'T use your meter on the current range, because that will shunt the high-value protectors specified with a very, very low value resistance internal to the meter. this will then allow lots of current, as much as your battery has, and it will all flow through your precious geraniums.

kaa-FFffft! you are reading a voltage formed across a resistor, by a current flowing. use the volts range.

[Ben Lyman]

Is this the right setting? It's the only setting that gives a reasonable answer when I use the Small Bear method

EDIT: The part that I CIRCLED above ^
NOT where the selector is pointing

[LightSoundGeometry]

[duck_arse]

Mr Lyman - that is the current measuring range, yes, that (or the 200u) is the correct range for the smallbear method.

the RG method needs to be measuring voltages, though.

[Ben Lyman]

Ahh, I see now, the two methods must be very different then. Thanks Duck!

LSG: I watched the vid, pretty cool but also a bit complex for me.

SBE method is so quick and just one quick math equation at the end. I believe my readings to be accurate because I have made several FF's, two TB MKII's and two RM's, all resistors were stock values per schematics, gains/leaks were measured and tranny's plugged into their positions accordingly.
I highly recommend people check out Steve's method and give it a try at least once, it is certainly accurate enough for guitar FX w/Ge tranny's.
For Si, I just plug 'em into my DMM HFE scale and call it a day 

[LightSoundGeometry]

Ahh, I see now, the two methods must be very different then. Thanks Duck!

LSG: I watched the vid, pretty cool but also a bit complex for me.

SBE method is so quick and just one quick math equation at the end. I believe my readings to be accurate because I have made several FF's, two TB MKII's and two RM's, all resistors were stock values per schematics, gains/leaks were measured and tranny's plugged into their positions accordingly.
I highly recommend people check out Steve's method and give it a try at least once, it is certainly accurate enough for guitar FX w/Ge tranny's.
For Si, I just plug 'em into my DMM HFE scale and call it a day 

ben, that is Steves method, just in video form lol

[hellwood]

I'm lazy, so I made an Excel spreadsheet to do the math using RG's method.  I have 500 or so germaniums all categorized and numbered in nice columns with gain and leakage, and parts drawers that are sorted accordingly.

[Ben Lyman]

ben, that is Steves method, just in video form lol

Haha! so it is! I looked a little closer this time. There is a lot more math involved in the video though. I think the main difference is that when I put my DMM on the above ^ setting (2000u) I actually get the number that I want without any further math. Then all I do is subtract, divide, and done. That last step (subtract and divide) is the only math involved, everything leading up to that is just hookup and read the numbers

[tubelectron]

I had the same problem : valid Ge transistors measurement... So I have built a small (manual) bench test. The 1st picture lets you know it all :





It needs an external DC regulated PSU (variable if possible) and 2 DMMs : one for Ib, one for Ic.

This little bench allow me the measurement of Ice0, Ib, Ic for Hfe, Icb0, IceS, so I can calculate the real Hfe = (Ic - Ice0) / Ib

A+!