Transistor AF117

[misterp]

Here is a question for all you AF117 experts: Among my old radios I have three which have AF117's in them. Two are not working at the moment but one is: it is an Ace Challenge de-luxe that I purchased new in 1963 (I've read somewhere these were made by Perdio). I've been refurbishing it case wise and a few other minor things. It seems to work perfectly and I've tried the tap test on the two AF117 transistors, which are in the RF section, and nothing happens. What would be the symptoms if they had the tin whisker problem? Also would it be beneficial to change them for new transistors, maybe the BF450 or AF125?

[paulsherwin]

If it's working then you don't have a tin whisker problem (yet).

There are zillions of threads discussing tin whiskers and AF117 replacement if you search the forum.

[Paul_RK]

Originally Posted by misterp

What would be the symptoms if they had the tin whisker problem?

Most often just loss of signal, so all that's heard is a quiet background hiss much the same as what's heard between stations on a healthy set. Tapping affected transistors will sometimes, but by no means always, produce noises and occasionally restores more or less normal operation for a while. It won't usually have any effect on a set that's already working.

Paul

[misterp]

Thank you Paulsherwin. It's good to know I don't have the problem yet, I will look through the posts. It makes you wonder how it starts as this one is 58 years old and has been stored in lofts and garages over the years so has been subjected to a wide range of temperatures.

[Boulevardier]

Originally Posted by misterp

It makes you wonder how it starts

It's an interesting topic. There is (or used to be - haven't checked recently) information on the "whiskers" issue on the NASA website if you're interested in background and details.

Mike

[Graham G3ZVT]

Originally Posted by misterp

Thank you Paulsherwin. It's good to know I don't have the problem yet, I will look through the posts. It makes you wonder how it starts as this one is 58 years old and has been stored in lofts and garages over the years so has been subjected to a wide range of temperatures.

These transistors are a little unusual in having a 4th wire connected to the case and normally grounded.

The whiskers are like microscopic wire-wool they grow out of the tin plating inside the case, they are growing inside your AF117's as we speak! The problems start when one or more whiskers reach the germanium or one of the leads. Often initially, only one electrode is involved and there is an easy fix, simply cut the screen wire!
In the case of my Bush TR82, it took 40 years for the oscillator stage to reach that stage and I cut the screen and restored operation. But it was only a matter of time until the whiskers engulfed more than one electrode, and this took a further 10 years to reach that point.

I removed the transistor and shorted EBC together and zapped it with 30v wrt the case. Normal service once again restored.

[Electronpusher0]

The one thing we learn from history is that we never learn from history.
Lead free solders have re-introduced the tin whisker problem.
The link below is an official NASA paper on lead free solder, it includes some pictures of tin whiskers on SMD boards soldered with lead free solder

The tin whiskers part starts at page 41
https://www.google.co.uk/url?sa=t&rc...CEb3tpbALljy6v

Peter

[Ted Kendall]

And now we have self-driving cars chock full of tin whiskers?
No, thanks - I'll take my chances.

[Graham G3ZVT]

Originally Posted by Ted Kendall

And now we have self-driving cars chock full of tin whiskers?
No, thanks - I'll take my chances.

One of the exempt industries, along with Aerospace and Medical, is
Automot{#`%${%&`+'${`%&NO CARRIER

>>><<<

[crackle]

I have repaired may radios with these transistors, but found that out of them 3 of the radios which had AF117 etc. transistors and which were marked with Holland on the transistor were working perfectly. It made me think that maybe the Holland factory had a different manufacturing process to the majority of AF11x transistors.
Mike

[paulsherwin]

I think it's just the luck of the draw.

[Nanozeugma]

Originally Posted by Electronpusher0

The one thing we learn from history is that we never learn from history.
Lead free solders have re-introduced the tin whisker problem.
The link below is an official NASA paper on lead free solder, it includes some pictures of tin whiskers on SMD boards soldered with lead free solder

The tin whiskers part starts at page 41
https://www.google.co.uk/url?sa=t&rc...CEb3tpbALljy6v

Peter

An extremely interesting read / resource Peter, thank you.

[Graham G3ZVT]

Originally Posted by Nanozeugma

An extremely interesting read / resource Peter, thank you.

This paper is closer to home though!

https://nepp.nasa.gov/whisker/anecdo...ransistors.pdf

[Refugee]

It is not just tin.
Zinc can be pretty bad too. When I had a look on the web a few years ago I found a story about a data centre that had some wiring upgraded. After this all the computers started to suffer many random failures.
They eventually discovered that the zinc plated ducting had zinc whiskers on it and they had been stirred up by the cabling work and floated into cabinets through the cooling fans and ended up shorting components around parts than needed cooling.
My best whisker photos are of zinc whiskers on the chassis of a valve signal generator. Fortunately the component spacing is wide enough that they are unlikely to cause a problem.
I reckon the most reliable solution is to remove circuit boards and blow them over with compressed air at regular intervals well away form other equipment.
Zinc is fine if the spacing between components is wide enough and there is enough current available to blow the shorts open circuit without causing disruption.
They have been using zinc on high voltage pylons for years without issue.
The weather just clears the whiskers off before they become visible to the naked eye let alone long enough to cause a short.

[ms660]

They grow underground as well, used to get them down the Tin mine.

Tin whiskers have been known about for lots of years going back to the 1960's at least.

Lawrence.

[misterp]

There is (or used to be - haven't checked recently) information on the "whiskers" issue on the NASA website if you're interested in background and details.

Mike

I've had a look at the NASA website and there's some interesting reading about it there.

Thank you all for your replies, I didn't realise the scale of the problem in other fields.

[Maarten]

Originally Posted by crackle

I have repaired may radios with these transistors, but found that out of them 3 of the radios which had AF117 etc. transistors and which were marked with Holland on the transistor were working perfectly. It made me think that maybe the Holland factory had a different manufacturing process to the majority of AF11x transistors.
Mike

It might also depend on production date. I've read a quote from someone on this forum who worked at Mullard, who said that by the end of the production run the problem was known and may have been solved. This could have been at a different point in time depending on the factory or even production line.

[CambridgeWorks]

I've seen them on 1990s comms equipment. On tinned screening shields of of rf modules. Sometimes up to 6 or 7mm long. Always growing at right angles to the original surface. Breathing closely to them made them move about.
Rob

[G6Tanuki]

I've seen tin-whiskers extending _inside_ the cases of some gain-control-type potentiometers - thankfully the whiskers would have to grow near a centimetre to intrude on the working-bits.

But equally I've seen whiskers extending into a 4-stage VHF helical antenna-filter in the front-end of a 160MHz mobile-radio; though the whiskers didn't short-out the filter their naughty capacitance between the 'hot' end of the helical-filter coils and ground did result in something like 6dB of lost-gain detuning.

The entire fleet of radios concerned were promptly removed-from-service and crushed to prevent anyone else suffering the problem.

[paulsherwin]

Just a reminder to stay on topic please.