Diode between emitter and collector?

Started by Guurf2, January 09, 2018, 11:25:01 AM

Previous topic - Next topic

Guurf2

Hi,
So lately i've been noticing a strange thing.
Some of the transistors i test on my transistor tester (diy atmega328 based) appear to have a diode(?) between the collector and emitter. I thought i was only something that appears when i test 4-leggers (RF germaniums) but it appeared today when i tested some black glass Tfk, happens sometimes with Mosfets too (between source and drain).
What is it? Does it mean anything? I remember it didn't always happen when i tested 4 leggers and now it's always there when i test them. Maybe something's wrong with the tester? Maybe it means nothing? Could it be a sign of huge leakege even though iceo seems low?
Does anyone know what it is?
Picture for refference:


Thanks,

Guy.

anotherjim

Probably is leakage. Have you seen it with any silicon BJT's? Is it a diode or a voltage drop across a faulty junction - that's a common fail with Ge.

Some power BJT might have a C-E diode as flyback protection from inductive loads. I've never seen one, but it could exist.
Diode should exist with MOSFET (source to drain), probably all of them.

vigilante397

As Jim said, MOSFETs (and some others) absolutely should have an internal diode. This is to block current in one direction when the FET is off, then when the gate threshold is reached and the FET turns on the body diode is shorted and current is allowed to flow.

https://en.wikipedia.org/wiki/Power_MOSFET#Body_diode
  • SUPPORTER
"Some people love music the way other people love chocolate. Some of us love music the way other people love oxygen."

www.sushiboxfx.com

Guurf2

Didn't see it in Si's but i a lot of Ge RF transistors, almost never in usual Ge's (Ac128, Ac125, Oc71 Etc) didn't seem to have them at all.
today was the first time i've seen this in other Ge's - tfk-Oc615V/Oc615M, Valvo-Oc47.
does it help anyone figure out?

Thanks!

Rob Strand

#4
What people of said above makes more sense to me.  (MOSFET = body diode, Ge = Leakage, Si = ???)

Your best bet is to look-up the datasheets, especially for the RF devices, you might see a diode on one of the pics.  Since EMC and static discharge have become more and more importand I wouldn't be surprised to see a diode on some modern devices.

Darlingtons with diodes have been around for some time,
http://pdf.datasheetcatalog.com/datasheets/savantic/2882.pdf
Send:     . .- .-. - .... / - --- / --. --- .-. -
According to the water analogy of electricity, transistor leakage is caused by holes.

iainpunk

Heyy. What is the direction of the diode? Some transistors have a high leakage which may be confused with a diode by the tester. If it even shows a diode when in reverse, its leakage. This is very common in GE nuggets.

Hope you find it out.

Iain
friendly reminder: all holes are positive and have negative weight, despite not being there.

cheers

Guurf2

Ill check some datasheets.
whats weird is that i had some super leaky transistors tested and it just showed super high Iceo (once it showed like 2.5Ma) so i think the tester can usually tell.
i started seeing this diode sign consistently in Ge RF models test results, and lately in some OC615/OC47 (i'm not sure they are RF).
so that's rather confusing  :-\

thanks people!

Guurf2

Hi everyone.
i performed some more checks and i'm proud to say results are in (and they are pretty happy too):

the diode symbol is probably actually a diode within the transistors, and not leakage/faulty tester.
and the reasons are:

i tested with a multimeter a few different transistors that show similar iceo on the transistor tester, but some had the diode sign and some did not.
their collector-emitter resistance was almost identical (and very high).
i compared with some super leaky SFT308 i have for refference (that doesn't show a diode sign, just 1.5ma Iceo) and it's collector-emitter resistance was much lower.

also turns out OC615 is actually an early telefunken RF transistor (stated in Radiomuseum) and it showed a diode sign, alongside af178 that showed a diode sign. compared them to similar PNP low frequency transistors that didnt show a diode sign and had a fairly low leakage as well.

so i call this one SOLVED

Guy.

anotherjim

I'm not sure I entirely trust your tester. Have you confirmed a functional transistor by the more traditional way with a multimeter and looking for only 2 diodes from the base?
Note that 4wire "RF" germaniums  are not hugely different from 3wire AF types. They have less miller capacitance at the expensive of gain (smaller junction area?) and the 4th terminal is just the connection to the can for screening use. The can should have no connection to the transistor - if there is any, the transistor is faulty, but might be usable if you simply don't connect the screen.
If you have any glass 4-wire ones, you found something new to me.

Guurf2

#9
Quote from: anotherjim on January 10, 2018, 12:15:28 PM
I'm not sure I entirely trust your tester. Have you confirmed a functional transistor by the more traditional way with a multimeter and looking for only 2 diodes from the base?
Note that 4wire "RF" germaniums  are not hugely different from 3wire AF types. They have less miller capacitance at the expensive of gain (smaller junction area?) and the 4th terminal is just the connection to the can for screening use. The can should have no connection to the transistor - if there is any, the transistor is faulty, but might be usable if you simply don't connect the screen.
If you have any glass 4-wire ones, you found something new to me.

Oh hi, i think i didn't explain myself well.
My tester actually did it's job, at first i suspected it shows a diode between C and E because it's misinterpreting high leakage. but i checked several different transistors (some did show a diode between C and E on the tester and some did not) with the same ICEO value on the tester for collector - emitter resistance (using an analog multimeter), and the resistance in all of them was the same.
also i put a few (of these that showed the diode symbol between C and E on the tester) in sockets in a fuzz face circuit and they sounded nice and quiet, not like typical leaky devices.
i used the multitester just to confirm that my transistor tester was doing it's job and it did.

now for the 4 leggers - i know these are pretty much the same as 3 leg pnp's but designed for higher frequencies (like radio for example) and that the 4th leg isn't part of the junction. what i did notice is that all of these 4 leggers i tried showed the diode symbol between C and E on my transistor tester and also showed very low ICEO (which i confirmed with the multimeter as well). also some 3 leg black glass OC615 i have constantly show this diode symbol on my transistor tester and low leakage as well (and i read that theyr'e designed for high frequencies, like most 4 leggers i have).

so i assumed that it is a common thing for high frequency Ge PNP transistors to have a protecting diode inside because that's the only explanation i could find, for the fact so many high frequency transistors i have show this diode symbol on my tester.

i hope someday someone will find this thread useful for the same kind of question  :)

Guy.