madbeanpedals::forum

General => Open Discussion => Topic started by: sgmezei on February 13, 2012, 08:01:47 PM

Title: how to build a transistor tester
Post by: sgmezei on February 13, 2012, 08:01:47 PM
Okay, so I ordered a batch of 100 Russian transistors (GT108V)  and 20 GT402E and I want to build a rig on perf board to test them. I am following RG's article seen here:
http://www.geofex.com/article_folders/fuzzface/fffram.htm

An excellent article!!!

What I would like to do, is use my 1spot adapter which puts out about 9.4 volts but it is constant.
Do I have to adjust the resistors to make my DMM read the "correct" values at the end of the procedure?

Any advice or help with this project is much appreciated.
Title: Re: how to build a transistor tester
Post by: jrod on February 13, 2012, 10:42:36 PM
That should work just fine with your adapter. Not sure what you mean by "adjust the transistor" but considering the way ge's are sensitive to temperature that .4V over 9V should not make much of a difference if any at all. Just pop the transistors into their socket, let em settle, measure, do the math, and you should be good! Those Russian transistors are great. I have some GT109 that I like a lot and GT402's that are great too. 
Title: Re: how to build a transistor tester
Post by: sgmezei on February 13, 2012, 11:09:20 PM
whoops!!! I meant resistors, I fixed it  ;)
Title: Re: how to build a transistor tester
Post by: sgmezei on February 16, 2012, 05:32:18 AM
How many good sets did you get out of your batch?
Title: Re: how to build a transistor tester
Post by: cjkbug on February 17, 2012, 03:10:27 PM
you could use a 9volt regulator after the 1spot to get a stable voltage. and make sure you let them stabilize for a few minutes before writing the measurements as handling the transistors will add heat.
Title: Re: how to build a transistor tester
Post by: sgmezei on February 24, 2012, 07:53:32 AM
Okay, I got a big batch of Russian Germaniums and started to test some of them out. It is late and I need to ask some dumb questions.
I tested a bunch of them and am not quite sure I am doing it right.
RG method from his fuzz face article

I measure across the 2k7 resistor and got my first reading.
I then connected the 2.2M resistor and measure the DC across the 2k7 resistor again.

Is this correct? Am I measuring the 2k7 resistor both times?

Well I got these readings on a few:
The first number is the leakage and the second is the gain (if I am doing this right)
0.55 , 1.45
0.4 , 1.12
0.14 , 0.83
2.07 , 3.37
0.22 , 0.91

So would the first transistor have a gain of 90 hFe?

I figure this out right because I have 140 to test.

Help me Team!
Title: Re: how to build a transistor tester
Post by: sgmezei on February 24, 2012, 07:54:34 AM
This is the meter setting I am using. I want to make sure.
Title: Re: how to build a transistor tester
Post by: plesur on February 24, 2012, 08:51:49 AM
If you are using 2.2M and 2.49K resistors from what I understand, you multiply your readings by 100 then subtract one number from the other.

So (1.45 - 0.55) * 100 = 90

So you are correct.

RG thinks that any leakage over 500uA is bad. What's the leakage on this one?

0.55/2472 = 0.000222 = 222uA

So anything up to 1.23V on the leakage side is 'usable'.

So the second transistor is also somewhat leaky and the fourth one is faulty.

But you'll need to adjust for your resistor values, as this is approximate.


( Edited to clarify leaky transistors )
Title: Re: how to build a transistor tester
Post by: claytushaywood on February 24, 2012, 09:26:52 PM
Isnt the first tranny getting close to "too leaky" as well?

So I have a multimeter that tests transistors... this does not test for leakage though, correct?

Is this method that RG describes as accurate as a bought transistor testing device?

For those of you that have bought large quantities of transistors off ebay or something... how are they? 

Any of you sell them?
Title: Re: how to build a transistor tester
Post by: sgmezei on February 25, 2012, 02:48:47 AM
Thanks Plesur, that is a huge help.  Good to know I am somewhat on the right track.

Claytushaywood, if you haven't read the RG keen article above, Do It, I found it really helpful. New/cheaper multimeters don't really test for leakage or account for it AFAIK.

This is my first batch of tranny's off of ebay so I will post the gains I get. Don't know how many good ones I will get so I won't offer to sell any hahaha.
Title: Re: how to build a transistor tester
Post by: claytushaywood on February 26, 2012, 05:18:42 PM
cool!  let me know! yeh i had just been reading it.  though i was dozing off and my comprehension was awful.

but yeh, let us know about how it turned out.
Title: Re: how to build a transistor tester
Post by: sgmezei on February 27, 2012, 06:07:21 PM
Okay so here is what I got for GT402E, I ordered 20 of them and I believe 19 are useable.
The third number is the Hfe

        1.   0.16, 1.01,  85
   2.   0.28, 1.27,  99
   3.   0.30, 1.82,  152
   4.   0.20, 1.23,  103
   5.   0.10, 0.89,  79
   6.   0.30, 1.52,  122
   7.   0.23, 1.04,   81
   8.   0.32, 1.81,  149
   9.   0.20, 1.15,  95
   10.   0.68, 1.71,  103
   11.   0.25, 1.15,  90
   12.   0.49, 1.86,  137
   13.   0.21, 0.98,  77
   14.   0.28, 1.61,  133
   15.   0.24, 1.05,  81
   16.   0.40, 1.60,  120
   17.   0.26, 1.43,  117
   18.   0.48, 1.53,  105
   19.   0.29, 1.27,  98
   20.   2.14, 3.78,  bad bad bad
Title: Re: how to build a transistor tester
Post by: claytushaywood on February 27, 2012, 07:12:16 PM
very nice... got some fuzzing out planned do ya?
Title: Re: how to build a transistor tester
Post by: plesur on February 27, 2012, 07:32:17 PM
These look like a nice set of trannies, low leakage, good gains. There's a few fuzz face and tone bender sets in there!

I just might get me some of those too.  ;)
Title: Re: how to build a transistor tester
Post by: culturejam on February 28, 2012, 03:42:40 PM
Some of the Russian germaniums have an internal "shunt" resistor that makes measuring the gain accurately very difficult. The big-ass GT404 trannies don't see to have them, but the smaller flying-saucer trannies usually do.

On the other hand, the Russian trannies are generally far more modern in manufacture than the "coveted" US/UK germanium units. As a result, they tend to have lower leakage and lower failure rates. At least in my experience, that is.
Title: Re: how to build a transistor tester
Post by: sgmezei on February 29, 2012, 06:02:46 PM
IT308

   1.   6, 96,      90
   2.   18, 141  123
   3.   10, 97,    87
   4.   10, 121   111
   5.   30, 156   126
   6.   27, 186   159
   7.   8, 88,       80
   8.   15, 121   106
   9.   10, 90,     80
   10.   26, 156   130
   11.   12, 107    95
   12.   9, 98,        89
   13.   20, 120    100
   14.   10,112     102
   15.   7, 92,        85
   16.   11, 108,    97
   17.   20, 133     113
   18.   18, 121     103
   19.   12, 95,      83
   20.   17,117      100
   21.   7, 97          90
   22.   13, 103     90
   23.   7, 114       107
   24.   5, 83         78
   25.   10, 112     102
   26.   5, 88         83
   27.   8, 111       103
   28.   7, 103       96
   29.   6, 77         71
   30.   16, 111     95
   31.   9, 84         75
   32.   19, 120     101
   33.   10, 103      93
   34.   9, 104        95
   35.   7, 82          75
   36.   13, 111      98
   37.   9, 90          81
   38.   6, 86          80
   39.   12, 107      95
   40.   6, 101        95
Title: Re: how to build a transistor tester
Post by: Haberdasher on February 29, 2012, 06:06:18 PM
So 40 out of 100?
Title: Re: how to build a transistor tester
Post by: plesur on February 29, 2012, 06:19:37 PM
It's like you're starting your own numbers station (http://en.wikipedia.org/wiki/Numbers_station) ...  :D
Title: Re: how to build a transistor tester
Post by: sgmezei on February 29, 2012, 06:29:56 PM
Yes 40 out of 100, I got tired of testing so I am taking a break for a few days. They sound great.

And no, no number station....hahaha
Title: Re: how to build a transistor tester
Post by: Haberdasher on March 01, 2012, 02:39:24 AM
What about the gt402e's?  do those sound good too?
Title: Re: how to build a transistor tester
Post by: jimmybjj on March 01, 2012, 02:56:11 AM
wow, you are getting lucky with those gains. My last batch of tranny's (100) didnt have 1 fuzz faceable transistor.
Title: Re: how to build a transistor tester
Post by: sgmezei on March 21, 2012, 03:33:51 AM
Alright here is what I have come up with for testing all of them so far.


https://docs.google.com/spreadsheet/ccc?key=0AiuJFraWqwPxdDdUNE5fdmRuSE95S28tc2p0ZlJ0VWc


Also, I have to correct myself. I got the 108's not the 308's
Title: Re: how to build a transistor tester
Post by: sgmezei on March 21, 2012, 03:38:06 AM
Quote from: culturejam on February 28, 2012, 03:42:40 PM
Some of the Russian germaniums have an internal "shunt" resistor that makes measuring the gain accurately very difficult. The big-ass GT404 trannies don't see to have them, but the smaller flying-saucer trannies usually do.

On the other hand, the Russian trannies are generally far more modern in manufacture than the "coveted" US/UK germanium units. As a result, they tend to have lower leakage and lower failure rates. At least in my experience, that is.


Is there a better way to test these type of transistors then? A more accurate way for our purposes?