Function of this transistor in 60s radio?

[Circlotron]

My dad bought one of these radios, an Astor P12L, in the mid 1960s. I still have it. Goes okay but the speaker voice coil is open cct. Anyway, just today i got a circuit for it and there is one transistor function I can't work out. It is the one at the bottom labeled "ATI". It's a PNP with the emitter to the negative supply rail and the collector feeds power to the self oscillating mixer. At first I thought it was some kind of voltage regulator circuit and it's output would depend on the current drawn by the following first IF stage. But now it looks like it is drawn connected backwards with C and E swapped. What's the verdict? What does it do?

[Quarlo Klobrigney]

That's a funny looking symbol for a transistor.
I see the oscillator 72 and the 1st IF 73 at the top.
I can only guess it's role. 
Maybe bias control for the oscillator over the varying battery voltage?
We are after all talking primitive transistors....

[SilverSolder]

Never seen that symbol for a transistor before - awesome!   And positive ground?  Cool!

[Quarlo Klobrigney]

There were lot's of strange goings on back in the day. Mostly PNP Germaniums and not many NPN with rigged circuits to get the fT high enough to work the "high" frequencies.

Quote from: SilverSolder on Today at 21:59:12

Never seen that symbol for a transistor before - awesome!   And positive ground?  Cool!

[Chris56000]

Hi!

It's an anti–ovsrload device/a.g.c. amplifier transistor!

Under quiescent or low–to medium signal, TR73 s passing maximum collector current and the voltage drop across R36 is at maximum, so TR74 is fully conductive (almost) and the maximum supply voltage is supplied to the mixer/oscillator collector.

If a burst of high–amplitude interference occurs or an exceptionally strong signal is tuned In, a large positive d.c. is developed across the detector load (volume control) by the detector diode, and this is fed back to the 1st i.f. stage as reverse a.g.c., by R41, reducing the collector current of the stage and therefore the drop across R36, which will reduce the conduction of TR74 to reduce the mixer/oscillator's supply voltage to compensate!

Chris Williams

[Circlotron]

Hi!

It's an anti–ovsrload device/a.g.c. amplifier transistor!

Under quiescent or low–to medium signal, TR73 s passing maximum collector current and the voltage drop across R36 is at maximum, so TR74 is fully conductive (almost) and the maximum supply voltage is supplied to the mixer/oscillator collector.

If a burst of high–amplitude interference occurs or an exceptionally strong signal is tuned In, a large positive d.c. is developed across the detector load (volume control) by the detector diode, and this is fed back to the 1st i.f. stage as reverse a.g.c., by R41, reducing the collector current of the stage and therefore the drop across R36, which will reduce the conduction of TR74 to reduce the mixer/oscillator's supply voltage to compensate!

Chris Williams

Yep, I can follow what you are saying, but for this function to happen TR74 would have to be an NPN. I'm pretty sure that is not the case. The symbol is the same as the others but I could look inside the radio and have a look at the transistor part number.

[Quarlo Klobrigney]

It's AGC. See page 5 of the pdf below. It spells out the functions of the transistors.
https://www.kevinchant.com/uploads/7/1/0/8/7108231/p12l.pdf

ATI crosses to an NTE126, PNP Germanium

[orbanp]

Doing calculations there is not much C-E voltage on that transistor. Voltage calculations from the voltage on the collector of TR78, or from the collector of TR77 and on the resistor of R48.
If ATI is indeed a PNP transistor, it could be operated in the "reverse mode", collector and emitter swapped. After all a transistor is a "symmetrical" device, except  the manufacturing process is "optimized" that the device would work well with dedicate C and E terminals. ATI could be a transistor optimized for reversed operation?
You could just take out that transistor and measure it, would be interesting to see what it is.
Regards, Peter

[Circlotron]

The transistor is marked Anodeon ATI. Must be something special to be working in reverse polarity like orbanp says. A quick google search found nothing.

[Quarlo Klobrigney]

Those transistors look almost like a Ruski design. Better change those electrolytics soon. See Shango's videos

[Gyro]

Never seen that symbol for a transistor before - awesome!   And positive ground?  Cool!

It's actually pretty representative of (maybe even based on) the transistor construction of the day - The die supported on a ring shaped base electrode contact, with the collector connection grown on one side with a bond wire and the emitter on the other.

The modern symbol is a bit more representative of a modern planar transistor, although still not right because it's the collector that is the substrate, not the base - try drawing a symbol that accurately represents that! 

[Bruce Abbott]

Under quiescent or low–to medium signal, TR73 s passing maximum collector current and the voltage drop across R36 is at maximum, so TR74 is fully conductive (almost) and the maximum supply voltage is supplied to the mixer/oscillator collector.

If a burst of high–amplitude interference occurs or an exceptionally strong signal is tuned In, a large positive d.c. is developed across the detector load (volume control) by the detector diode, and this is fed back to the 1st i.f. stage as reverse a.g.c., by R41, reducing the collector current of the stage and therefore the drop across R36, which will reduce the conduction of TR74 to reduce the mixer/oscillator's supply voltage to compensate!

I think you have this backwards. TR74's Collector is connected to the primary side of the 1st IF transformer (driven by TR72), so if turned on it will short out the signal.

So under normal conditions TR74 should be turned off, which it will be because the Base is reverse biased (-4.8V on the Base, -5.7V on the 'Emitter').

When a burst of high level interference occurs the AGC goes hard on causing TR73 (1st IF amp) to turn off. This causes voltage across R36 to decrease until TR74 turns on, shorting out the IF transformer and reducing the signal level.
 
The reason for wiring TR74 in 'reverse' (Collector and Emitter swapped) might be to reduce its current gain for softer limiting.

[bsfeechannel]

The modern symbol is a bit more representative of a modern planar transistor, although still not right because it's the collector that is the substrate, not the base - try drawing a symbol that accurately represents that! 

The modern symbol is reminiscent of the point-contact transistor. The very first.

[Circlotron]

I think you have this backwards. TR74's Collector is connected to the primary side of the 1st IF transformer (driven by TR72), so if turned on it will short out the signal.

The reason for wiring TR74 in 'reverse' (Collector and Emitter swapped) might be to reduce its current gain for softer limiting.

Yeah Bruce, I think you've got it. The mistake I made was thinking this transistor fed power to the first IFT, but power is fed to the tapping point on the primary. The collector looks like it will load the primary. This business of having the collector and emitter swapped is a new on on me. I wonder what is special about that transistor to work this way, reversed? I haven't been able to find data sheet on it.

[SilverSolder]

The modern symbol is a bit more representative of a modern planar transistor, although still not right because it's the collector that is the substrate, not the base - try drawing a symbol that accurately represents that! 

The modern symbol is reminiscent of the point-contact transistor. The very first.

I like the junction transistor symbol.  I'm going to try to use it in a diagram!

[David Hess]

I wonder what is special about that transistor to work this way, reversed? I haven't been able to find data sheet on it.

When the collector and emitter are swapped, the current gain is lower, possibly even less than 1, and the collector-to-emitter saturation voltage is lower, which matters in chopper and muting applications.  I do not know that it matters here.

Before JFETs became available, they made special "chopper" bipolar transistors which had a symmetrical emitter and collector so their gain was equal, or at least more equal, either way.  These can be identified by their high base-emitter breakdown voltage which may equal their base-collector breakdown voltage if they are truly symmetrical.

Could transistor 74 be an RF noise blanker which works by quenching IF transformer 63?  It is in the right spot but I do not see how it could work either unless it is rectifying the tapped signal from IF transformer 64, and I do not see how that could happen.

[Circlotron]

Probably not a noise blanker, at least not an effective one. I remember riding around in dad’s 1959 Hillman Minx and before we did something about it, ignition static was wall to wall.

[David Hess]

Probably not a noise blanker, at least not an effective one. I remember riding around in dad’s 1959 Hillman Minx and before we did something about it, ignition static was wall to wall.

Maybe it did not work because they put the wrong polarity of transistor in.  Or maybe the noise would have been much worse without it.

[vk6zgo]

Never seen that symbol for a transistor before - awesome!   And positive ground?  Cool!

It's actually pretty representative of (maybe even based on) the transistor construction of the day - The die supported on a ring shaped base electrode contact, with the collector connection grown on one side with a bond wire and the emitter on the other.

The modern symbol is a bit more representative of a modern planar transistor, although still not right because it's the collector that is the substrate, not the base - try drawing a symbol that accurately represents that! 

The "modern" symbol actually predates the symbol shown in the radio circuit.
Around the early Sixties, there was a push to adopt the latter, as it supposedly was more related to the diode symbol.

It was adopted by some companies, including the radio magazine "Electronics Australia".
"EA" hung in there with the symbol for ages, till they finally had to admit they were "backing the wrong horse".

[SilverSolder]

[...]
The modern symbol is a bit more representative of a modern planar transistor, although still not right because it's the collector that is the substrate, not the base - try drawing a symbol that accurately represents that! 

This is an attempt:

[Circlotron]

^^ Reminds me of an SCR symbol.
I wonder how they came up with that?