Avantic BM611 Stereo Decoder

[PJL]

I have just finished renovating this deluxe tuner with a cleanup and replacement of Hunts capacitors and out of tolerance resistors. It is working OK although FM is not as good as I would have expected and may have been twiddled with.

I now want to try to fit a stereo decoder. I have bought a cheap MC1310P decoder on the basis that I am not sure it's going to work out!

The circuit is attached and I also have the alignment instructions but I am assuming I will need to try to extend the bandwidth to about +/-100kHz to get stereo to work properly. My first problem is I do not recognise the demodulator design? There is also a high impedance output for a stereo decoder but this was probably more of a marketing ploy.

My next problem is going to be patching up some of my ancient test gear so that I can sweep the 10.7 IF. I have various options and hopefully one of them will come good.

Any advice from the experts as this is early days and I have yet to to get to grips with FM demodulation and stereo decoding theory.

[Ted Kendall]

First thing, obviously, is to get the tuner alignment correct.

The decoder output will (or should) be free of the 50uS de-emphasis network - stereo requires a flat bandwidth to about 60kHz feeding the decoder to accomodate the sidebands of the multiplex signal. You may need an emitter-follower to feed the decoder - I can't remember the input impednace of the 1310 off-hand.

[G8HQP Dave]

I think it is a Foster-Seeley discriminator. The multiplex output should have sufficient bandwidth, but you will need a buffer.

[PJL]

Agreed, it must be Foster Seeley. I can't see any obvious signs of de-emphasis but maybe that was in the pre-amp.

Options for sweeping 10.7MHz are a sort of working Polyskop but that would need a detector as it is low impedance but it does have markers, a similarly sort of working TF2370 for which I did pick up an amplified probe for but it is untested. The TF2370 probably wins as the Polyskop is in the garage and I am not sure I have the strength to bring it inside.

[Nuvistor]

There is a resistor/capacitor filter on the audio output of the Foster Seeley, are those de-emphasis or are they the wrong values, cannot read them on the diagram?
Frank

[PJL]

It's a 100pF so I over looked it but you are absolutely correct, the time constant with the 470K is 47uS. The modulator output is connected after this which will need changing which could be a bit of a challenge as we are working with very high impedances and even a few pF is going to attenuate the higher frequencies from the stereo signal.

The service data is for an older version and the cathode follower is now AC coupled and the bias is taken from a potential divider in the cathode. The more invasive solution would be to strip out the 470K/100pF de-emphasis and use the cathode follower as the buffer for the decoder then replace the volume control with a rotary switch for stereo/mono. Almost as bad as sticking a DAB tuner in the box!

[G8HQP Dave]

Strange to have the 'multiplex' output after the de-emphasis components! You may need to leave some filtering there to reduce the IF signal. You could simply add a resistor of a few k in parallel with the 470k and take output from the cathode follower. At lower volume settings the decoder will switch to mono.

[PJL]

I presume for stereo the emphasis is applied to each audio channel before multiplexing? When I remove the de-emphasis is the cathode follower going to have enough headroom? Maybe best to check this with a scope when I have sorted the alignment.

[Nuvistor]

Perhaps the Stereo decoder was not user installable and required some modification. When was the radio made, using an EF80 for RF amp would perhaps put it quite early and there would be few Stereo transmissions about. Perhaps it was sold as "convertible " like so many 405 TV's were in the early 60's.
Frank

[G8HQP Dave]

Pre-emphasis is applied to each channel before encoding. As higher frequencies are smaller amplitudes anyway the follower should be fine. The FM system bandwidth will fail before the follower does.

[Synchrodyne]

Agreed that it does appear unusual to have a dedicated multiplex (MPX) output jack that is fed via the de-emphasis circuit. Perhaps it was intended that the de-emphasis capacitor be disconnected for stereo operation? Usually for valved FM tuners, MPX outputs were taken direct from the discriminator, without any attenuation of other treatment, as shown by these examples:




This was the case even where the audio output went via a cathode follower. Thus such MPX jacks they could then be used to connect a separate MPX decoder, without any internal circuit changes being required. The MPX output was typically in parallel with the audio output, but the de-emphasis resistor, usually ahead of the de-emphasis capacitor in the audio output acted as a stand-off, isolating the MPX output from any de-emphasis effects.

In some cases, internal switching in the tuner allowed the audio output to double as an MPX output. In the MPX position, de-emphasis and any attenuation was bypassed:



There was an implicit assumption that any connected decoder would have a suitably high input impedance, usually obtained with an appropriate buffer stage. For example, Quad used a bootstrapped emitter follower for its MPX decoder unit:



Notwithstanding the “conventional” MPX output arrangement used by Leak in its Troughline II, it did differently in its Troughline Stereo, which had an internal transistor decoder. In this case, the MPX signal went through the cathode follower (ECF80), which thus served as the buffer between the tuner and the decoder. The mono output was also taken from the cathode follower, via a de-emphasis network. (In the Troughline II case, the de-emphasis network was ahead of the cathode follower.)



Anyway, the Leak Troughline Stereo case provides a precedent for adapting the Avantic BM611 to route the MPX signal through its cathode follower.

De-emphasis for stereo tuners was mostly done after decoding, along with filtering, the latter often involving a low pass above 15 kHz with a null at 38 kHz and another notch at 19 kHz. But with frequency division multiplex (FDM) decoding, as could be done with say the Philips TDA1005A IC, it was possible to place the de-emphasis ahead of the decoder, with the usual roll-off circuit in the base (sum) channel input to the IC, and an appropriately shaped bandpass filter in the difference channel input. But that approach may have been more difficult when it came to easily adjusting between 50 and 75 microsecond time constants.

By way of an aside, an odd feature of the Avantic BM611 was that it used both an ECF82 (FM oscillator-mixer) and an ECF80 (FM reactance and AF cathode follower). Both of these valves started life as TV and FM oscillator-mixer combinations, and both found multiple other uses. It is difficult to rationalize that the Avantic BM611 circuit requirements were such that two of one variety could not have been used.


Cheers,

[Radio Wrangler]

I've just had a look at the circuit. The tuner case, name and price may be rather deluxe, but the circuitry isn't at all special. It's closer to what you'd find in a radiogram than in a Leak Troughline.

This must have put people in a quandary. For someone who likes the Avantic amplifier, but wants best quality VHF/FM reception, then it would have been better paired with a Leak tuner. But that would have looked wrong.

David

The detector is a Foster-Seeley. The ratio detector can be spotted by the diodes facing opposite directions, the Foster-Seeley with them pointing the same way.

[Ted Kendall]

In the sixties, one Harry Leeming, a service engineer, had a column in Hi Fi News, which read in the same league as Mac Hellyer. One of his pieces was advice to the effect that you shouldn't blindly buy the same make of tuner as your amplifier without checking its performance. Granted, everybody was feeling his way into the stereo era at the time, but there was evidently a broad spread of competence, even among established brands.

[Radio Wrangler]

G3LLL, yes, I used to know Harry.

A nice guy, and knew what he was talking about. He became the number-one guru on the Yaesu-Musen FT101 series.

David

[KeithsTV]

Originally Posted by Ted Kendall

I can't remember the input impednace of the 1310 off-hand.

According to the MC1310 datasheet the input impedance is typically 50k so to connect it to the multiplex output you'd need a buffer, nothing too complex an emitter follower would do.

Keith

[PJL]

Pentode RF amp, Triode oscillator, Pentode Mixer, Pentode IF amp, Pentode IF amp, Cathode Follower output + AFC + tuning indicator. Better than a typical radio and similar to the Troughline but has AM too.

Somebody had preciously played with the alignment possibly to extend the lower end of the frequency range. Not having a wobbulator, I used my dodgy Marconi 2022 on widest FM modulation and Tek 454A on XY plot off the modulation output to sort it out. Not ideal but in the end it showed a pretty flat response over the 100KHz deviation and gain was up. Next job is to find some vero! To avoid hacking the tuner about, I plan to wire out the de-emphasis and take the cathode follower output to an external box with the MC1310 powered by a wall wart. I am hoping by focusing the re-alignment on providing a reasonably wide bandwidth I have a better chance of getting decent stereo.

[PJL]

Assembled the decoder. Not yet decided how to wire it to the tuner.
1. Hack it into the tuner case using a voltage doubler off the heater and put one channel through the existing AM/FM switching and the other to the multiplex phono socket. Not very satisfactory as one channel would be affected by the volume control.
2. Take out the de-emphasis, keep the volume/cathode follower and put the decoder in an external box. A switch would be required for AM/FM to bypass the decoder and another for Mono and the volume would need to be set to the correct amplitude for the decoder.
3. Use a voltage doubler to power a J-FET buffer with a new connection prior to the de-emphasis and wire to the existing multiplex output. Put the decoder in an external box with a Stereo/Mono switch which bypasses the decoder. This seems to be the best option and in mono leaves the set exactly as designed.

[Ted Kendall]

The separate box solution is nearest to period practice - Quad and Heathkit come to mind. Separate amplifier inputs were sometimes used for mono and stereo feeds, so in practice it was quite elegant.

[PJL]

I am listening to stereo but it does seem a bit muted at the top end. To prove it's going to work I just lifted the capacitor that formed the de-emphasis and am using the cathode follower output. It fairly easily locks on to stereo so it may be my recalculation of the de-emphases for the 1310 and I also need to check the pre-amp is internally DC isolated.

PS: Ooops it's not decoupled so back upstairs to add a few output caps!

[Radio Wrangler]

How about putting the decoder in a diecast box painted to match the Avantic chassis?

David