Hacker RP25B hissing

[Andy 1964]

I am having problems with this radio, turning on from cold it hisses at a constant volume for about 5 min before any audio comes through. Cleaned all the contacts, cleaned the 5 pin connector at SK1, replaced all the electrolytic capacitors and replaced most of the lockfit transistors with BC548 and 2N3904. Eventually all the wave bands work, but not the next day. Any ideas please.

[paulsherwin]

This does sound like a bad transistor in the amp. Which ones haven't been changed?

[Andy 1964]

on the amp board A205, I changed T1 & T2, from BC148 to BC548, which sounded better.

[Radio Scotland]

Maybe a bad pre-amp transistor, I'm not familiar with this radio so just a guess

[agardiner]

Has it got those typical resistors which have a sort of end cap to them? Sorry, don't know the official name.

I have had a few of these Hackers which behave exactly as you describe. In every case it has been a fractured resistor on the amp board. (exact one varies) Have a good look at them under a magnifier, and gently try moving them ever so slightly with a pair of pliers. That is usually enough to find the culprit.

[agardiner]

to add to this - the hairline fracture seems to close up as the resistor slightly increases in temperature, hence why the radio eventually works normally. So another method would be to use a freezer spray once running OK.

[Philips210]

Hi.

One thing I've found when replacing Lockfit transistors in some Hacker radios is the mixer/oscillator transistor is more critical than say the IF and audio transistors. If replacing the BF194 with an alternative I use a BF494 which I believe is the same transistor in a modern package.
I remember having problems with the oscillator not always starting when a similar RF transistor was used as a substitute. By putting back the original BF194 the problem was solved.

Regards,
Symon.

[Nickthedentist]

I would second using freezer spray too.

[paulsherwin]

Originally Posted by Andy 1964

on the amp board A205, I changed T1 & T2, from BC148 to BC548, which sounded better.

I don't have the circuit handy, but I think there is another BC148 hiding somewhere.

[majoconz]

Turn the volume down low - still hissing? - then the problem is in the power supply rails or in the audio amp. Run it on an external power supply - no hiss? - suspect the power rail bypass capacitors - still hissing? - then it must be in the audio amp, maybe that TAA920 IC.

Just beware that if you run it on an external power supply then it needs 18volts plus a centre tap - it should draw somewhere around 20mA.

[Andy 1964]

I tried the amp board from a RP18, and it worked straight away. So its definitely the amp board. It does have resistors with capped ends, so will check those.

[Andy 1964]

I was poking around and grounded the emitter of T3, labelled T2, and I briefly got sound. I came across this page https://www.markhennessy.co.uk/artic...sistors.htm#t2
So I replaced it with a 2N3904 and it works perfectly! But its a NPN type and the mysterious T2 is a PNP type, surely it shouldnt work!

[paulsherwin]

I think you have found the bad transistor.

You should replace it with another Ge PNP type even if it seems to work with the 2N3904 - it can't be operating as the designer intended. Anything you have to hand should be OK.

One of the recently offered ACY20s should be fine if there are any left: https://www.vintage-radio.net/forum/...d.php?t=188015

[Radio Scotland]

Sure i have some ACY20s in my stash

[Andy 1964]

I will try a GT322b, which is PNP, and report back.

[paulsherwin]

That should work.

[Andy 1964]

It didn't like that GT322b at all, T5 and T6 got very hot and fried the russian. T3 which is a T2 transistor checked out on my tester OK, so put it back in.
From cold there is brief audio and then just hissing for 5 mins.
I measured the transistor voltages from cold, the collector to emitter voltages were low, about 6v except T5 which was high at 11v.
After 5mins when audio starts the voltages are almost correct.
The only things I have not replaced are T4 AC128, T5 AC176 and T6 AC128.
The amp board from a RP18 works fine anytime.

[paulsherwin]

That doesn't make sense. Are you sure you fitted the GT322B correctly? It's easy to get the leadouts wrong.

[Radio Wrangler]

So, during the hissyfit, the DC bias voltages are wrong?

Then the source of the trouble ought to be findable with a multimeter, a circuit diagram and some logic (mental, not TTL, sort)

Transistor audio amplifiers aren't very amenable to the just-keep-changing-parts approach because you have several transistors all interlinked so that each affects the DC conditions of the others and there is a tendency for one faulty device to blow others up. In this way, by blindly replacing only one part at a time, you can get trapped in an endless loop.

There are two ways out of the impasse. One is to remove and test all the devices. All! at once. Reassemble with new parts where needed and it ought to work. With the transistors/diodes out, it's a good time to check all the resistors and capacitors to bend the odds a bit more on your side. This method can fail if the semiconductor testing isn't realistic. Multimeters with transistor test functions and those little semiconductor testers don't have the power supply voltages to run even modest devices at realistic conditions. Hfe figures can be quite different in their regimes, and they don't get a chance to spod devices which would start to break down in the actual circuit.

The second method is to understand the circuit, to see how the DC conditions (voltages and currents) are established some logicking should then identify a group of suspects, though rarely gets it down to a single part.

There are techniques for extreme audio amp fixing.... (remember those reports of extreme ironing with likeable loonies taking ironing boards up pointy mountains?) This is needed when you have a newly designed amplifier first prototype before you and it isn't working. Current limited lab power supplies can be substituted for the normal ones to limit destruction. The amplifier can be broken into sections and jigs made to bias and stabilise each so they can be tested one at a time. THis is where you really get to understand what can go wrong.

Transistor audio amps are right b'stards to fix quite often, and have a stinker of a reputation. But they can be handled. The first step is to accept that their nature often militates against quick fixes or obvious faults.

But having to fix a bias problem is usually much easier than a noise problem without any clues in bias values.

David

[Andy 1964]

I checked that very carefully, I did not connect the shield to anything.