Hacker AF117 best alternatives?

[mhennessy]

Originally Posted by nebogipfel

Yes, might it be possible to make up a simple zapper so that a transistors could simply be plugged in then a switch thrown?

Yes, I've often thought about making a magic box to automate the process. One day...

Anyway, here's what I do:

1. Remove from circuit, gently twist EBC together, and solder to one end of 100uF capacitor. Bend leads of capacitor so that other end of capacitor is just a few mm away from screen lead.

2. Measure DC resistance between EBC bundle and screen - it might be just a few ohms, but even if O/C continue anyway...

3. Charge capacitor to the highest DC voltage you have to hand. I have a dual PL320, which gives just over 60V.

4. Touch screen to other end of capacitor. Take care not to touch both ends of capacitor as 60V might tingle, depending on how conductive your skin is.

5. You might hear a small arc, and perhaps the inside of the transistor will light up (the base is glass).

6. Turn down PSU, discharging capacitor in the process.

7. Measure resistance again, confirm it is O/C. Leave DMM in place (use croc-clips) and tap the case in all directions, while watching for signs of intermittent connections.

<following is optional if no sign of incipient shorts, but recommended anyway>

8. Solder screen to capacitor, charge up capacitor again.

9. Hit case of transistor from all angles, in the hope of "catching" stray whiskers. Watch PSU metering - you might spot the output drop briefly as shorts make contact. And try not to s/c the capacitor, as this makes bit sparks!

10. Turn down PSU, desolder transistor and double-check resistance of each electrode to screen. If O/C, re-install in radio, then check for signs of intermittency by tapping once again. If in any doubt, remove and repeat. If OK, you can be confident in a medium-term fix.

Pretty easy. What I do differently to most is the high voltage, the tapping (8 onwards), and the expectation of doing it again. It works though.

All you need is a capacitor and a power supply. Even a couple of PP9s will work, but the higher the voltage the better, IMHO...

Good luck,

Mark

[Keith]

Thanks Mark,

100uF is smaller than I would have expected - but then the whiskers are presumably almost microscopic. I use a similar technique on shorted NiCads but with a 20,000uF capacitor. It ususally works - but the fix doesn't last long!

[nebogipfel]

Very clear description and explanation. Thanks Mark

[mhennessy]

Pleasure!

The most time-consuming part is removal, and to lesser extent refitting. That's why I go through the procedure, even if it seems OK at step 2. The high voltage might jump across a whisker that is about to reach the junction - also the tapping might cause a whisker to move to within range. For these reasons, the high the voltage the better in my opinion. When I do eventually build my box (boost converter and big "fire" button!), I'll probably go beyond 60V.

Charge is 0.5 times C V squared

You get a surprisingly big spark from 100uF at 60V. Take care - especially when tapping them.

And yes, the whiskers are tiny. You might have already seen it, but the NASA website is a fascinating read: http://nepp.nasa.gov/whisker/background/index.htm

When I'm back home, I'll try to post some images taken from a chassis that have tin whiskers forming. It's in a Sanyo cassette deck. It's all furry

Mark

[crackle]

Hi Mark
Why use a capacitor, do you have a specific reason for this.
I have had 100% (sample of 4) success with just connecting the EBC leads and connecting my bench power supply on max. (30v) to the case/shield and EBC, simple.

Mike

[mhennessy]

Fair question - especially as your bench PSU will have a capacitor of a similar value mounted on the output terminals. Using another capacitor, actually at the transistor itself ensures that a much higher peak current can flow - but how necessary or worthwhile this is, I honestly can't say. I do like to solder the transistor EBC leads together, and solder them to the capacitor at the same time - it's all quite neat.

But yes, there are no absolutes with this - everyone has their own preferred method, and no-doubt has a similar success rate to me. What I do works and has become a habit, and that's really all I can say.

All the best,

Mark

[PJC58-Hythe]

I recently bough 10 AF124's from here http://www.donberg.co.uk/search?query=af124

Took a while to arrive but were NOS and all worked fine.

Peter

[nebogipfel]

Originally Posted by PJC58-Hythe

I recently bough 10 AF124's from here http://www.donberg.co.uk/search?query=af124

Took a while to arrive but were NOS and all worked fine.

Peter

Thanks for the link Peter. Shame about the P&P charges

[davidgem1406]

I have suffered the problem of replacing multiple AF11x's and finding that AF12x's are quite highly priced now.

I went looking for possible alternatives but only germanium devices. I checked the specification for each device I found against the AF11x's and AF12x's, then picked out those that were at least equal, preferably better.

What I ended up with is the following:

AF106/AF109R/2N3783 - £0.10 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.
Not good for use as FM oscillator.

SFT358 - £0.57 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.

2SA518 - £1.46 each - This is a direct equivalent to the AF124 so should work OK for the full range of AF11x's. I have not used any so far.

P416/2N384/2N990/AF137 - £0.28 each - This is a Russian transistor & should work for all the AF11x range. I have not as yet used any.

I have several Roberts and at least one Hacker that need the AF11x's replaced so no doubt I will be trying the units I have not as yet used.

Dave

[paulsherwin]

A very wide range of transistors will work as AF11x replacements - most of the time. There really isn't a problem where transistors are easily accessible and trial and error can be used. The difficulty comes when transistors can only be changed after lots of dismantling, leaving the radio in a non-working state. Only the very brave would use anything other than an AF12x in these cases.

I've had success with obscure Japanese RF transistors scavenged from 1960s Hong Kong radios and ex computer switching transistors, though the latter tend to be a bit low gain.

[robjkmannering]

Thanks Mark for the tip regarding tapping the case during the zapping procedure, it had not occured to me that this physical action may help.

Regards,
Rob

[nebogipfel]

Originally Posted by davidgem1406

I have suffered the problem of replacing multiple AF11x's and finding that AF12x's are quite highly priced now.

I went looking for possible alternatives but only germanium devices. I checked the specification for each device I found against the AF11x's and AF12x's, then picked out those that were at least equal, preferably better.

What I ended up with is the following:

AF106/AF109R/2N3783 - £0.10 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.
Not good for use as FM oscillator.

SFT358 - £0.57 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.

2SA518 - £1.46 each - This is a direct equivalent to the AF124 so should work OK for the full range of AF11x's. I have not used any so far.

P416/2N384/2N990/AF137 - £0.28 each - This is a Russian transistor & should work for all the AF11x range. I have not as yet used any.

I have several Roberts and at least one Hacker that need the AF11x's replaced so no doubt I will be trying the units I have not as yet used.

Dave

Unfortunately a quick search online has turned up pretty high prices on all the numbers you mention. I presume as time goes on the situation with obsolete transistor pricing will get worse.
There are some Russian transistors listed on Ebay which claim to be equivalents but I think this discussion has led me to believe the best solution is either zapping or for total reliability replacement with Af12*'s

Thanks for the input chaps

On a different but related subject my recently purchased and extensively got at RP18 appears to be fitted with an RP25 amplifier board. I wonder if this will present problems aside from originality? I notice the correct board was only fitted to RP18 and 30/31

[paulsherwin]

Zapping is fine as a short to medium term solution, but the whiskers will grow back eventually. You may get ten years or more before it happens of course.

[davidgem1406]

Posts: 434

What I ended up with is the following:

AF106/AF109R/2N3783 - £0.10 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.
Not good for use as FM oscillator.

SFT358 - £0.57 each - This worked as a replacement for the AF114 & AF117 so should replace the whole range of AF11x's.

2SA518 - £1.46 each - This is a direct equivalent to the AF124 so should work OK for the full range of AF11x's. I have not used any so far.

P416/2N384/2N990/AF137 - £0.28 each - This is a Russian transistor & should work for all the AF11x range. I have not as yet used any.

The prices I quoted are the prices I paid per transistor some 12 to 18 months ago, all from ebay.
No doubt in that time they could well have increased. Likewise has the cost of AF12x's.

Hello Paul,

I have also used these transistors in the Mullard modules, I string the module in on wires long enough to allow me to work on it if need be.
No problems encountered to date.

Dave.

[JoshWard]

I have used BC557s in several radios and haven't had a problem yet. They are cheap and readily available.

No idea how well they'd cope in an FM section though. I have only used them in AM only sets so far.

[nebogipfel]

Originally Posted by JoshWard

I have used BC557s in several radios and haven't had a problem yet. They are cheap and readily available.

I like the price Josh

Have you used them in Am Hackers?

[crackle]

Originally Posted by robjkmannering

Thanks Mark for the tip regarding tapping the case during the zapping procedure, it had not occured to me that this physical action may help.

Regards,
Rob

Hi Rob
That is another good reason to connect direct to a bench PSU and dont bother with an additional capacitor, I can turn the transistor round to all angles and tap it whilst connected to the supply.
Mike

[boiss]

I Siliconed my Sovereign11 on AM, (FM is already silicon), It had already been got at and siliconed but the connections were wrong and there were a couple of O/C tracks at the transistors. Only problem was the LO didn't on LW, Changing the the bias resistors to the later silicon values sorted it.
I guess that for seudo originality (does that exist?)the bottoms of the AFs
could be opened BC/BFs fitted and hot glue sealed, would look perfect. After all for valve jobbies it's fine to fit new caps in old cans!!!!!
Cheers
Mike

[howard]

Hiyas,

I always replace AF11* with AF12*, never had any problems with them and they won't fail again as will repaired AF11*s. They are fiddly to fit sometimes, but nowhere near as fiddly as replacement silicon transistors like BF450s. The AF121 is similar to the AF124 so would probably work OK.

Howard

[mhennessy]

Originally Posted by crackle

That is another good reason to connect direct to a bench PSU and dont bother with an additional capacitor, I can turn the transistor round to all angles and tap it whilst connected to the supply.
Mike

So how does a capacitor affect this?

Mark