Hey there,
I cant find a clear answer to this question…. Do transistors (if stored properly) degrade while sitting on the shelf? I often find new transistors to have a too high hfe for repair jobs on old gear. Is it a safe bet to replace them with nos transistors? Or am i better off to buy new production? Im talking silizium transistors - not germanium.
Cheers
I cant find a clear answer to this question…. Do transistors (if stored properly) degrade while sitting on the shelf? I often find new transistors to have a too high hfe for repair jobs on old gear. Is it a safe bet to replace them with nos transistors? Or am i better off to buy new production? Im talking silizium transistors - not germanium.
Cheers
Unless they have been improperly stored, I don't see much risk using old transistors. If in doubt, use one of those cheap transistor testers. Static damage while improperly handling them over the years is much more likely to happen. So testing still is a good idea.
There are only a few things that suffer from storage, mostly electrolytic caps, and modern SMD parts (moisture, leading to cracks when being soldered), and of course any mechanical parts made from/with plastic.
There are only a few things that suffer from storage, mostly electrolytic caps, and modern SMD parts (moisture, leading to cracks when being soldered), and of course any mechanical parts made from/with plastic.
Modern SMD parts should simply be baked out if they have been in storage a while. Just warm enough for the water to out-gas but not boil off destructively. SOME old through hole types too -if they have an epoxy seal instead of a molded case. I tend to avoid those types, as many were prone to failure in the first place.
Never had a bit of problem with old TO-92, TO-220, TO-126 unless they were the air cavity epoxy sealed type (the old style TI TO-220 being notorious). Old aluminum TO-3’s are usually fine - other than the fact that specs and uniformity were pretty loose in those days. Static can build up on old mosfets - you’ve got to check those for gate leakage.
Germanium transistors often need to be “cleared”. They can grow conductive tin dendrites inside, and of course test as a short. You simply have to apply enough current between two terminals to fuse it, but not hurt the junction. I’ve never seen silicon transistors do this because of the different metallurgy involved. But I have seen innards of old Russian single diffused 2N3055-oids that damn sure look like old germanium parts to me - they may very well grow tin whiskers.
Never had a bit of problem with old TO-92, TO-220, TO-126 unless they were the air cavity epoxy sealed type (the old style TI TO-220 being notorious). Old aluminum TO-3’s are usually fine - other than the fact that specs and uniformity were pretty loose in those days. Static can build up on old mosfets - you’ve got to check those for gate leakage.
Germanium transistors often need to be “cleared”. They can grow conductive tin dendrites inside, and of course test as a short. You simply have to apply enough current between two terminals to fuse it, but not hurt the junction. I’ve never seen silicon transistors do this because of the different metallurgy involved. But I have seen innards of old Russian single diffused 2N3055-oids that damn sure look like old germanium parts to me - they may very well grow tin whiskers.
If your reference for the hfe is a tran from the device in operation, the old tran may have experienced beta degradation. I've had a case where I had to replace a front end tran in an amp because of beta degradation. I measured the other channel front end devices and they too had degraded, but not to the point of causing an offset voltage failure like the one that caused me to open up the amp. I replaced both channel's front end tran's since it was just a matter of time until the other channel had the same problem.
Beta degradation in differential input stages is quite possible if there is no protection against reverse Veb current (ie when clipping). It’s the most common cause of field failures in matched devices (won’t match anymore, and if critical all hell breaks loose). It won’t happen from sitting.
I have experienced phenomena that maybe related to your question. Recently i have discovered a small stash of electronic parts in the basement left from my student years 30 years ago. I found metal can small signal transistors ( BC107/BC109 types) covered with looking like rust. But after examined closely found that, that is not rust instead looking more likely a tin whiskers. Entire outer shell of the transistors were like a sandpaper instead of usual smooth surface. I am guessing storing them in a humid environment for a long time is not a good idea. Although i have not check electrically yet so they may be still functional.
That's just a reaction with the case plating, threy should be fine internally unless the seal has failed.
Old tin plated steel cases will do that. It’s just a nasty oxide that doesn’t like to come off. In the case of old steel TO-3’s you usually need to sand a bit around the mounting screws to get a good non-rectifying contact. The internal tin whiskers are not formed by oxides, but tend to grow in the direction of an electric field. Silver will do it to some extent too (silver-filled conductive epoxy being the source) but tin is by far the worst that was in general use.
Semiconductors are manufactured at very high temperatures. At room temperature, any chemical reactions have essentially stopped.
Ed
Ed
Internal chemical reactions, with the exception of those involving water if there is a way for it to get in. External chemical reactions can degrade heat transfer, electrical contact, or solderability. Those usually involve water too. Damn those polar molecules.
Sorry if it appeared as if I was saying that beta degradation would occur by sitting. I was saying it occurs during use, albeit ever so slowly. In my case, some 35 years of use. It may be worse with higher beta devices like my amp used in the front end, I've not researched it that deeply. Here is one reference https://www.dfrsoft.com/DfRSoft - Transistor Aging.pdf
It’s not so slow under reverse avalanche. It’s almost all recombination, occurring at higher energies which are basically like hitting the crystal lattice with a hammer.
Higher beta devices have more to lose, and it occurs faster. It’s normal to run switching BJTs in emitter base avalanche during fall time. They don’t have squat for beta to begin with.
Higher beta devices have more to lose, and it occurs faster. It’s normal to run switching BJTs in emitter base avalanche during fall time. They don’t have squat for beta to begin with.