I'm helping a friend resurrect an Ampex 440 studio tape deck. Things ran fine for a couple of hours, then one playback channel suddenly got noisy.
Question: Why? What is an NPN transistor's internal mechanism that causes this? Back in my repair days I ran into this; it was relatively rare, but happened enough to notice.
Thanks.
ak
Question: Why? What is an NPN transistor's internal mechanism that causes this? Back in my repair days I ran into this; it was relatively rare, but happened enough to notice.
Thanks.
ak
The most usual cause of increased transistor noise is reverse current through the emitter-base junction. When you drive a transistor into emitter-base breakdown, the charge carriers can get high enough energy to cause damage. This results in a much increased 1/f noise component of the equivalent input noise current. There are device physicists on this forum who can tell you precisely what happens, I'm only a stupid circuit designer.
As to why this should happen to your friend's tape recorder when it is just playing, I have no idea. Are you sure it is the transistor and not a rotten contact somewhere?
As to why this should happen to your friend's tape recorder when it is just playing, I have no idea. Are you sure it is the transistor and not a rotten contact somewhere?
Last edited:
It’s likely a rotten contact - INSIDE the transistor. E-B breakdown just isn’t going to happen in a tape recorder - overdriving the input stages is needed and tape saturates so it’s just not likely. But surface contamination, diffusion, and other sources of high-energy electron-hole recombination can happen with older transistor processing. There are types we avoid like the plague because of this.
With the leads usually high current, of course moisture and impurities likely from old processing play in.
Ion exchange from the metal, screws up the chemistry.
Dont think BJT has oxide layers for function but believe it can be used for protection layers.
No idea if it plays in like it does with fet
Ion exchange from the metal, screws up the chemistry.
Dont think BJT has oxide layers for function but believe it can be used for protection layers.
No idea if it plays in like it does with fet
Oxides or nitrides are used for passivation - but there are always holes in it where the bond wires go. Moisture gets in from a poor epoxy encapsulation job,and it’s game over. At least as far a stable, noise free device. Older processing can also have poor surface or edge treatments, and you can get hot electron damage even if the damn thing is hermetic and there’s no tin inside to grow dendrites due to high field (early Si and almost all Ge did this). They just degrade over time, and it might not even take many mA to do it. Things they had to do to make ICs a reality did away with most of the problems.
Use a newer type transistor and it will shut it right up.
Use a newer type transistor and it will shut it right up.
Here's a thread on known bad transistors, (and replacements) for which you can review your schematic.