Fidelity RG32 - what were they thinking?

[Studio263]

Here's an interesting circuit from the 1965 season, the Fidelity RG32 stereo radiogram. At a glance it looks OK but look again...

One channel uses and EL84 and the triode part of the inevitable EABC80. The other uses an ECL86 for both functions, not really the same thing.

Then, they've used a common cathode resistor for both output valves, ensuring that the bias is always a bit out for each. The triodes share a common anode resistor too, which is only mildly decoupled.

Finally, one channel has a single resistor in its negative feedback loop where as the other has a resistor and capacitor combination, giving an LF boost on one channel only.

Fidelity's natural rivals for the bottom end of the market, Alba, went the other way that year with single ended EL34s driven by ECC83s and a proper tone control circuit. Models 9904 and 9905 had AFC on FM too, and the facility for an FM stereo decoder.

Madness both.

[Nuvistor]

First thought is they were using what they had in stock before making next years model use transistors.
It’s so odd they must have had a reason for making it that way.

[TonyDuell]

I wonder how much difference it really makes. There is negative feedback applied round this amplifier which should cover up the differences between the channels to some extent.

Having the 2 stereo channels not the same was not uncommon I think. After all, Philips (and others I guess) sold 4 track tape recorders where you could play a stereo tape using the internal amplifier and speaker for one channel and a preamplifier + your radio amplifier (signal going in the pickup sockets) for the other. And weren't there some demonstration stereo broadcasts where the 2 channels were sent, for example, on band 2 FM and 405-line TV sound (AM)?

[Nuvistor]

Probably didn’t make a great deal of difference to the sound and Fidelity were sold as budget equipment.
Still if they had to design and order in the parts it would have easier to have both channels matched, that’s why I think they were using what they had in stock, interesting to know if it was a short production run.

[TonyDuell]

I assume the diodes of the EABC80 were used for the radio detector stages. So you have a triode left in that valve which was used with a plain output pentode for one channel. In the other channel you also need a triode and output pentode, might as well get them in the same bottle I guess...

I was going to suggest this minimised the total heater current, and then realised an EL84 has a higher heater current rating than an ECL86.

[Synchrodyne]

For the same valve envelope count, an ECL86 pair could have been used instead of an EL84 and ECL86 combination. In that case the EABC80 triode would have been unused. (Although it could have been diode-strapped and used to provide delayed AGC on AM….)

Possibly maximum commonality with a corresponding mono chassis was required, hence one EL84 was a given. One operating mode for the EL84 listed by Mullard was emulating the 9-watt EL41, so perhaps that was used and considered to be an adequate match for the ECL86 9-watt pentode. But the ECL86 triode is 100 mu, whereas the EABC80 triode is 70 mu. Evidently that was not a cause for concern.

Perhaps a worse precedent for stereo output asymmetry, given that it was at a much higher price point, was found in the Ekco 9-octave record player. The main unit had an ultra-linear EL84 pair, whereas the add-on stereo unit had an ECL82 pair, not ultra-linear.


Cheers,

[Techman]

Originally Posted by Studio263

Fidelity's natural rivals for the bottom end of the market, Alba, went the other way that year with single ended EL34s driven by ECC83s and a proper tone control circuit. Models 9904 and 9905 had AFC on FM too, and the facility for an FM stereo decoder.

Madness both.

Yes, Alba!

You may have seen a thread a while back where I had an Alba record player for repair which used an ECL80 as the audio output valve - well actually it was a small radiogram as it had the addition of a small radio. I can tell you that it was one of the worse sounding record players I've come across for its size and type. I tried to improve things slightly by playing around with capacitor values, particularly the grid coupler, but had to leave as original in the end due to instability which this valve in this type of circuit is prone to. So the verdict? Total pile of its only saving grace being that it's a rather unusual unit with it using that particular valve. I suppose it wasn't a great surprise to me when its owner ended up giving it to me in return for another favour, he thought it sounded rubbish as well, and I agreed!

[Studio263]

Originally Posted by TonyDuell

There is negative feedback applied round this amplifier which should cover up the differences between the channels to some extent.

In this case it could make it worse, since the NFB networks are completely different.

The proper way to do it is to use an EBF89 was the IF amplifier (with its diodes connected for AM detector and AGC delay), two germanium crystals for the FM ratio detector and two ECL82s for the audio stages. Still not extravagant, but sensible.

[Paul Stenning]

I wonder if they used a mono radio or radiogram chassis along with the chassis from a record player to make up this odd arrangement from available parts.

Maybe they didn't care if the two channels sounded a bit different as they thought would emphasise the stereo effect (making people think that even mono radio was stereo)? Like the early "electronically reprocessed stereo" which was bass on one channel and treble on the other...

[dseymo1]

I suspect that many of us started our stereo listening using two different mono amplifiers (and loudspeakers) of some sort. It didn't seem to matter much at the time!

[1100 man]

I still do!
One channel through a 1970's Mitsubishi amp, the other though a Rotel '90's amp:- no idea why, I just never got round to fixing the Mitsubishi!! Sounds fine to me

[Ted Kendall]

Originally Posted by TonyDuell

And weren't there some demonstration stereo broadcasts where the 2 channels were sent, for example, on band 2 FM and 405-line TV sound (AM)?

Indeed there were - I have tapes of a couple, complete with advice about reversing speaker phase if the opening announcement doesn't appear between the speakers. They actually work OK, despite the different transmission systems. The Beeb had the devil's own job getting the two channels into step for these - on one occasion the TV sound went to Manchester and back to get the group delays about right...

[Studio263]

Originally Posted by Paul Stenning

I wonder if they used a mono radio or radiogram chassis along with the chassis from a record player to make up this odd arrangement from available parts.

Could be, sadly there are no chassis views in the book (R+TV S 1965 > 1966). I do remember having a Philips stereo radiogram years ago which used their standard small AM/FM chassis of the era (AC/DC, Permeability tuned FM front end with a big plastic wheel at the front with a spiral cut into it to move the coils, wavechange switch concentric with the tuning knob) with an extra chassis tacked on the back which had not only an extra output stage but also a rectifier valve (UY85) and a mains dropper as well. The functions to do with stereo operation were selected by an extra set of piano key switches mounted under the dial. I've not seen one since, or even the circuit diagram for it. Nice set though.

[Paul Stenning]

Originally Posted by Studio263

Could be, sadly there are no chassis views in the book (R+TV S 1965 > 1966).

No, just checked the manufacturer's data, forget that suggestion, it's all on one chassis!

[Nuvistor]

The AM/FM tuning capacitor is interesting, I have not seen it done this way before. The normal full capacity for AM, with fixed trimming capacitors for FM, using a standard AM gang instead of one made with separate sections for AM and FM.