It was a germanium PNP transistor. I might've pressed the button too. I saw transistors used as capacitors that were connected like this, but they weren't germaniums. What current (if any) has flown through the transistor? According to the datasheet, the reverse collector current at VCB=20V is 5μA, reverse emitter current at VEB=2V is 50μA, VEBO=3V, for what it's worth.


simulate this circuit – Schematic created using CircuitLab

  • \$\begingroup\$ What makes you think you might have fried it? Unexpected behavior or the smell of smoke, heat, etc.? \$\endgroup\$ – Billy Kalfus Sep 24 '18 at 20:52
  • \$\begingroup\$ You do not seem to have a clear goal with this circuit. In this circuit the current is limted to 9V / 470 ohms = 19 mA which will not harm any component as this current is quite small. But what is the purpose of all this? \$\endgroup\$ – Bimpelrekkie Sep 24 '18 at 21:02
  • \$\begingroup\$ @BillyKalfus It was one out of 50 and I don't remember which one, so I don't know how it's going to behave. I measured the whole batch again (VBE and hFE) and all of them seem fine, but those measurements say nothing about noise (some of these germaniums are hissy as hell, I haven't checked them yet). Can some parameter deteriorate after hooking it up like that? Like I heard about microwave transistors that had their fT decrease after being subjected to static electricity, other than that they still worked. \$\endgroup\$ – HaveIFriedIt Sep 24 '18 at 21:12
  • \$\begingroup\$ @Bimpelrekkie It's a test circuit for measuring VBE, the transistor was put into the socket the wrong way, it should've been connected as a diode with its base and collector tied together. \$\endgroup\$ – HaveIFriedIt Sep 24 '18 at 21:15
  • \$\begingroup\$ Just to be sure, the 9V supply is upside down here, right? \$\endgroup\$ – Passerby Sep 24 '18 at 21:53

The theoretical max power dissipation inside the transistor can have been 43 milliwatts. That's the case if the BE reverse breakdown voltage happens to be 4,5 V. If you have the datasheet of the transistor, find the thermal resistance from junction to the free ambient. Multiply it by 43 mW and see how much the internals were hotter than your room. Datasheet can also tell the max. allowed internal dissipation in some usual ambient temperature like 25 degrees centigrade.

If 2N3904 gets 43 mW, it would be about 9 degrees centigrade hotter inside than the room. I do not believe a discrete germanium transistor is destroyed due 43 mW dissipation.

ADD: Not any more so sure. I just found that I have had germanium rf transistors which can stand about max. 80 mW internal dissipation and that's in collector, which is several times larger than the emitter.

  • \$\begingroup\$ From the datasheet: thermal resistance 200°С/W, max temperature of the p-n junction 90°С, max collector power dissipation 200mW, max ambient temperature 70°С. This transistor seems relatively sturdy and should be fine, but I want to understand what was going on. Which junction we should worry about in my circuit, BE? What about BC? This link says CE is as fragile as BE when reversed, but there's nothing about BC. rohm.com/products/faq-search/faqId/232 When the allowed reverse voltage is exceeded, what exactly happens? Does it fail immediately, or heats up and melts? \$\endgroup\$ – HaveIFriedIt Sep 25 '18 at 0:56
  • \$\begingroup\$ @HaveIFriedIt The breakdown is like in zener diode, nothing is melted if the power dissipation is low enough not to rise the temperature too high for the material. Emitter is small when compared to collector, it can get hotter with the same dissipation than collector. Remake the wrong connection, but instead of 470 Ohm, have a 4,7 kOhm resistor. Then you can safely measure the reverse breakdown voltages, disconnet in turns C and E. Think the junctions as zener diodes. \$\endgroup\$ – user287001 Sep 25 '18 at 10:48
  • \$\begingroup\$ The results are in: i.imgur.com/4k5hB8M.png BC in reverse doesn't act like a zener, maybe because 9 V is not enough (in the datasheet they measure the reverse collector current at 20 V and it still leaks only 5μA). Found this, but it's vague, who knows if it applies to my 30-year-old unobtainiums. rohm.com/products/faq-search/faqId/231 \$\endgroup\$ – HaveIFriedIt Sep 25 '18 at 20:53
  • \$\begingroup\$ @HaveIFriedIt You put reverse 9V through 4,7kOhm resistor to BE junction and got about 1mA current - The breakdown has happened. The voltmeter shows it acts like about 4,2V zener. CB junction has higher breakdown voltage. \$\endgroup\$ – user287001 Sep 25 '18 at 21:01
  • 1
    \$\begingroup\$ Overheating sounds fun, gonna try that. Things are starting to make sense. How do I fry it though? Exceed the CE forward voltage? Found on wiki: "If the emitter-base junction is reverse biased into avalanche or Zener mode and charge flows for a short period of time, the current gain of the BJT will be permanently degraded." I don't have the measuring gear, so I'll have to listen to each transistor. Picked pairs (hfe, Vbe) for the left and right channels of the amp, and gonna make a 3rd channel, a test bench. Will switch between the two transistors on the fly and see if they sound similar. \$\endgroup\$ – HaveIFriedIt Sep 25 '18 at 22:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.