Kenwod TK-40SW , Intermittent problem in Ratio detector

Hi,

I have an intermittent failure in the FM ratio detector stage of a TK-40SW that's been driving me nuts. :tears: The output on FM sometimes disappears although the tuning meter indicates that the signal is there. Usually happens when turning the set on after it has been off for a long time.

Since the meter works even when the ratio detector fails, I sniffed around the detector driver, Qb5. When things were working, the DC bias looked OK, but when the stage died, there seems to be no current flow in the transistor.
DC Bias Qb5:
Working: B:12.45V, C: 4.27V, E: 12.51V
Dead: B:12.12V, C: 0V, E: 14.44V (Equal to supply rail)

And here's the kicker - I found that repeatedly touching the base lead of this guy with a wire would often bring it back to life. I just can't figure out why this happens. :dammit:

I can't detect any microphonics which would indicate a cold solder joint. I banged the transistor and the board with a plastic tool and nothing happens - it either continues to work or continues to be dead.

So suspecting a parasitic oscillation I built a diode based RF probe and checked again. Absolutely no RF activity when the stage dies. I also tried probing with a frequency counter and even a wideband ICOM radio, but there is no sign of any oscillation when the stage malfunctions.
RF Probing Qb5
Working : B : 0.78V, C : 10V, E: 0.06V
Dead : B : 0.77V, C : 0.06V, E: 0.02V

Before going on a wild goose chase for prehistoric transistors on Fleabay I was wondering if the Gurus on this forum could give me an insight into what's going on. BTW the TK-40SW is a FM Mono receiver.

Thnx and Cheers,
Sant

Ok, looks like it's time to bite the bullet and try replacing the transistor.

Does anyone know of where I can get a 2SA234 transistor ( Hitachi germanium PNP). I see it on Fleabay but the prices are a rip-off.

- Sant

Not a clue, but I'd install three small machined sockets from a sacrificial IC socket so you can change parts. I'd try a silicon device just to see if it works at all and the problem goes away. Then you can try various germanium devices. I've had very poor luck replacing germanium devices in some circuits because even the same type numbers don't always work the same.

Thanks, that's a good suggestion. I try the Si 2n3609 first as it seems to fit the bill on most parameters ( Vceo, hfe, Ft, Co )

Cheers,
Sant

Update - I wired up a transistor socket in parallel with the existing 2sa234. I didn't want to pluck out the old transistor until I was sure that it was the cause of the problem.

I left the socket empty, but the next time the FM died out I put a SI 2sa933 in the socket and lo and behold - the FM came back on loud and clear. And it went back off when I pulled out the 933. So now I know that the problem is with the 2sa234. And as usual, tickling the base of the 234 with a long wire got it working again.

Cheers,
Sant