Measuring Beta - what am I missing?

Mayank

Mayank

Well-Known Member
Hi:

I have an older Conar transistor tester model 214 which I use to check transistors using the following steps (a) clip the leads on to the E, B & C (b) select PNP/NPN (c) switch to Beta Calibrate (d) carefully adjust the calibrate knob to set the needle to 1 (for x10) on the meter (e) switch to Beta and take the reading.

I recently ordered 6 Onsemi MJ21193G PNP transistors from Mouser. While 3 show the beta to be ~500, the other 3 register NO MOVEMENT on the meter (on the Beta Calibrate). I've verified that these 3 transistors are OK (diode setting with my multimeter). But the beta measurements are still dead.

What am I missing :scratch2: ?

Rgds
Mayank
 
Register to hide this ad
Just checked another batch of different transistors. They all have beta in the range of 180-220. So the tester is working OK.

Questionable batch of Onsemi devices perhaps?

Mayank
 
This datasheet, from on-semi web site, specs the max hFE of those transistors is 75

http://www.onsemi.com/pub/Collateral/MJ21193-D.PDF

Fake transistors can measure very high (I've read in some service manual this advice: if XX transistor measures more than XXX value, don't use it), or something is wrong with the measurenent device or the measurement procedure. Try to check with another tester to isolate the problem.

What's the scale of the device? Perhaps you are exeeding some value, or you are reading low values for a large scale.... just thinking...

Last time I've matched some 16A TO-3 transistors (On semi MJ15024) all measured around 40, nothing close to 200 or 500... Not even 100.
 
Last edited:
Make sure you have good connections to the leads of the transistors - possibly the suspect ones have a film of oxide on the leads?
 
You assume that your meter can measure everything, but each test device has its limits. Power transistors like the MJ21193G may inflict a too large current draw from the Tx tester, upon which the test circuit saturates (or potentially even burns out when exposed for extended period, all depending on how the tester's internal circuit is built and protected)

Therefore, try to reduce the current draw and add a 2~5 watt resistor at the emitter of the transistor under test. Start with 47 ohm, then 22 ohms, then 10 ohm, then 4.7 ohm and see up to where the tester still works.

Then again, beta-matching of output devices should rather be done at real output currents, and real life Vce across the transistor, as Vce and Ic directly impact various characteristics. Not much sense to do that with small current transistor testers. But better something than nothing :)
 
Mayank said:
Hi:

I have an older Conar transistor tester model 214 which I use to check transistors using the following steps (a) clip the leads on to the E, B & C (b) select PNP/NPN (c) switch to Beta Calibrate (d) carefully adjust the calibrate knob to set the needle to 1 (for x10) on the meter (e) switch to Beta and take the reading.

I recently ordered 6 Onsemi MJ21193G PNP transistors from Mouser. While 3 show the beta to be ~500, the other 3 register NO MOVEMENT on the meter (on the Beta Calibrate). I've verified that these 3 transistors are OK (diode setting with my multimeter). But the beta measurements are still dead.

What am I missing :scratch2: ?

Rgds
Mayank
Click to expand...

The Vbe voltage of that particular transistor is unusually high. My guess is that it is right at the limit of what an old tester can deal with, and some percentage of the transistors have Vbe's that are slightly above the threshold. (Quite normal.)

If the 214 is a battery model, you could try putting new batteries in. If all you need is beta, there are any number of $20 DVM's now that will test beta. And, there are $30 component testers that will read beta, Vbe and various other parameters, too.

-k
 
elnaldo said:
This datasheet, from on-semi web site, specs the max hFE of those transistors is 75

http://www.onsemi.com/pub/Collateral/MJ21193-D.PDF

Fake transistors can measure very high (I've read in some service manual this advice: if XX transistor measures more than XXX value, don't use it), or something is wrong with the measurenent device or the measurement procedure. Try to check with another tester to isolate the problem.

What's the scale of the device? Perhaps you are exeeding some value, or you are reading low values for a large scale.... just thinking...

Last time I've matched some 16A TO-3 transistors (On semi MJ15024) all measured around 40, nothing close to 200 or 500... Not even 100.
Click to expand...

I have some MJ15001 MJ15002, ON-semiconductor also, they came from Farnell or RS afaik. I was allowed to take them home when equipment using it went obsolete at work. We did use them at replacements for some high power CRT deflection amps so they were doing quite some voltage swings and amps 24 hrs a day and did not give any problems.
Also these have very high HFE (above specs), some about 300.
High HFE stays at some more current than a microamps hfe tester gives, too.
90% is between 200 and 300, but some 10% are below 100, also...and yes they are from a few datecodes, only...

btw I also measured using power supplies dvms and resistors setup. Still, it puzzles me a bit.
 
Last edited:
I've measured a lot of 20 MJ15024 (the spec says max hFE 75) with my cheap Multimeter, and measured all almost the same, around 40-50, but it's true, they were measured at that low current. I can't tell what's the hFE at high current.
 
Thanks for all the responses. I wanted to circle back on the suggestions.

The MJ21193G transistors in question were ordered from Mouser. They are still new/unused with no oxide film on the leads (I cleaned them anyway).

The measurement devise (Conar transistor tester) appears to be working fine - yes, I bought new batteries to make sure. I had a couple complimentary MJ21194G transistors and other power transistors on hand - they all test OK with beta between 75-85.

To try and reach a conclusion, I ordered 10 MJ21193G directly from Onsemi (they shipped them all they way from Singapore!). These show beta (on the same Conar transistor tester) at the "expected" levels - 80,75,75,80,80,80,75,70,75 & 80.

The "original" MJ21193G transistors still show wierd results - 3 show ~500 and the other 3 register NO MOVEMENT on the meter.

My conclusion: either these were very good "looking" fakes or there came from a bad manufacturing batch. I've never had a problem ordering from Mouser before.

I will henceforth test each new transistor before using.

Rgds
Mayank
 
Interesting, maybe contact Mouser if they are interested testing the ones you have to see if they got fakes?
 
You must log in or register to reply here.
Back
Top Bottom