Vintage transistors and tin whiskers

Cursitor Doom wrote in news:r9rgu3$h74$ snipped-for-privacy@dont-email.me:

Put it in your ten below freezer for an hour and let the cold disintegrate the non alloyed tin.

Then give it a nice hard rap to break it all loosed like we used to do to the old tube type TVs to get them going again.

Kidding of course... except for the TV thing. That worked. So did taking the back off it and re-seating all the socketed tubes and components. Worked every time. It is almost always a mechanical connection.

I don't know about the whiskers but Ge transistors have high leakage and no sharp knee at low currents. I wonder if 10 ohms is too low depending on the voltage applied by your meter. I have a few of the later AC188 types. I'll dig them up tomorrow and measure their resistances.

For quick go-no go checks of transistors, I prefer analog MMs. The static forward resistance depends highly on the current applied through the probes.

I used to have some OC71s and OC72s a long time ago. They made neat phototransistors with the paint scraped off from their glass enclosures.

to freeze tin plague converts to dust))

OK, I got the AC188s. With a DMM they measure 25-50 ohms forward and a few Ks in the reverse direction.

I beg your pardon?

I think you've misunderstood the problem here, Pimpom. The leakage in question is between the case ("screen") and the die and significantly pre- date the black-painted phototransistors we all fondly recall from 50 years ago. :-)

White tin (a metal) is what is useful; grey tin, a semiconductor, is the allotrope that is stable at low temperatures. Grey tin is brittle, turns to dust.

More detail here:

it is necessary to measure the resistance of a cold transistor. it is so easy to distinguish metal from a semiconductor.

Tin whiskers in modern equipment are so fine the ohm meter will zap them away. Do you think you are seeing something else?

I have had some success with an Eddystone EB35 by replacing them with silicon transistors and adjusting the base bias to get approximately the same collector current.

(I was a tester on the production line at Eddystones when EB35s were going through, so I am the one to blame if they don't meet spec.)

CD, I'm not familiar with old transistors. If you have physical access to where you suspect the whiskers, then you might be able to see 'em, and even if not, knock them away with something.. maybe compressed air or with tooth brush bristles?

George H.

The whiskers will be on the inside of a hermetically sealed metal can.

and they are damned hard on razor blades!

access

Grin. Yeah, then the only option is to try and blow it out. I had a little bit of success blowing out whiskers for a short time with what I assume was a current that vaporized some of the whisker. This was in a rotary switch which I think was still actively growing whiskers. So x amount of time later said switch would fail again. What's the appeal of old Ge transistors? I might have some in an old parts box... want 'em?

George H.

On Monday, 18 May 2020 10:20:46 UTC-7, George Herold wrote: ...

...

The original poster was refferring to OC171 transistors. The can was filled with a gel in which the whiskers grow. It won't be possible to blow them out.

kw

For me personally, it's partly nostalgia and partly other things. For certain applications, their shortcomings compared to Si types were partly offset by their lower Vf, lower saturation and more linear gain.

You can make your own 'Ge' transistors by heating Silicon parts. Vbe drops and they go all leaky.

Problem is you need to keep them hot. Maybe a hybrid using a Silicon transistor with a valve (tube) heater! Steampunk transistors!

And if something starts oscillating because you have too much gain with the silicon replacements then adding a low value R in series with their emitters will usually fix.

There are already emitter resistors, but they are bypassed, so adding resistance in series with the bypass capacitors (or removing them altogether) would be the way to go.

I used fairly mundane silicon transistors with a lowish cutoff frequency and didn't have any trouble.