RGD 1046 Radiogram.

I recently acquired an RGD 1046 radiogram. Kindly donated by forum member Ed Dinning.
This was a upmarket product made in 1946. Some consider that RGD made much better products than HMV, even Dynatron.

I've been fortunate enough to hear a Decca 'Kensington' (Decola in expensive frock), an RGD 1046 and a Dynatron 'Ether Princess' side by side.

I have to say the Dynatron was superior to the RGD, but both were definitely better than that overrated lump, the Decola. The build quality of the Dynatron is absolutely superb, whilst Decca stuff looks positively thrown together.

I've also heard the 1046's little brother, the 746 - and that is actually a very nice piece of kit. Given the choice between it and the 1046, I'd actually pick the 746.

One day, I hope to get my McMichael DL1 running as I think this will give many of the established 'quality' grams a run for their money, along with a potential 'dark horse' - KB's FG50. KB were quite advanced in acoustic research after the War, pioneering and patenting the "slot diffuser" and they were also great users of the Decca XMS pickup head, but with quite advanced equalisation. I'd also love to find (but I doubt I'll ever find one) a Ferguson 501RG.

Does your 1046 still have the original Garrard 'High Fidelity' pickup?

Unfortunately the RGD has no turntable. However, I have a Goldring Lenco GL78 which will fit in the gram space.
Your are right about the McMichael DL1, a little known dark horse. Information for this magnificent radiogram can be found on page 32 of volume II Radio Servicing book. Audio output is 16 watts @ less than 1% distortion. Two 6L6G in push-pull. Three EF37 low noise pentodes are employed in the audio amplifier, noise limiter and tone control stages.
McMichael also made about the same time as the DL1 a brilliantly over engineered TV set.

The reason why I wish to part with the HMV 801 is because it is just too big. Nevertheless, it must go to someone who will appreciate this instrument. K3 turntable.

For the amplifier I intend to build how about employing the more common and cheaper 6L6 rather than the much more expensive KT66?
I'd imagine a triode connected 6L6 will have very similar characteristics as the KT66. Triode connected KT66 = PX4 triode. heater different.
Read about the 6L6 here: http://en.wikipedia.org/wiki/6L6

Till Eulenspiegel.

Hmmm. So really, this'll be a 1046 in name only, won't it...?

It's notable that the Brimar handbooks warn that a 6L6 produces considerable distortion single-ended, and really should be used in push-pull.

However, I would mention that a triode-strapped KT66 is not equivalent to a PX4, but a PX25. There was an article many years ago in Radio Bygones (I think) where the author made a sub-chassis carrying two triode-strapped 6L6s that effectively plugged in place of the two triodes. The 6L6s had the advantage of a more modest heater current requirement than a KT66.

In a sense, yes. The HMV 801 radiogram was modified to employ a pair of KT66 to replace those PX25 triodes. The valves are operating close their absolute limits as the HT voltage is 450V.
Others have used triode connected EL34s which can be used a very high HT voltages.

Till Eulenspiegel.

You should be fine with KT66s at 450V. Proper ones are rated for 500V (design max) or 550V (absolute max) on both the anode and the screen. That said you do have to be careful with modern ones. Quite a few are just 6L6s with a KT66 label printed on them. If you want to use a modern beam tetrode which should be just about indestructible then the Electro-Harmonix KT90 was developed especially for very high voltage operation. They're rated for 600V when triode-connected and 750V anode/650V screen when run as a tetrode. I fitted 8 (yes, really) into an ultralinear stereo amp which was conservatively rated at 100W/ch and had a 575V HT with a separate bias supply, so the full HT was between anode/screen and cathode. It was eating 6550s for breakfast but the 90s have given no trouble.

VB

I recently acquired an RGD 1046 radiogram. Kindly donated by forum member Ed Dinning.
This was a upmarket product made in 1946. Some consider that RGD made much better products than HMV, even Dynatron.

I've been fortunate enough to hear a Decca 'Kensington' (Decola in expensive frock), an RGD 1046 and a Dynatron 'Ether Princess' side by side.

[...]
I've also heard the 1046's little brother, the 746 - and that is actually a very nice piece of kit. Given the choice between it and the 1046, I'd actually pick the 746.

[...]

Does your 1046 still have the original Garrard 'High Fidelity' pickup?

Hi Cath.
I had an RGD 746 for many years, it was truly an excellent radiogram. It now lives in Hungary as far as I'm aware. I wish I still had it.

The one I heard had the Garrard High Fidelity pickup and was pretty original - and it really did sound great. The pickup in my DL1 by contrast is a Connoisseur high-fidelity unit. The nice thing about the DL1 is that it has a proper crossover feeding 12" and 4" loudspeakers, and a twin-t notch filter for needle scratch. Plus a socket for a table lamp!

Pictures of the radio unit. Note the high standard of construction and also the tuning capacitor which has separate vanes for the short wavebands.
Valve line up is: VP41 RF amplifier. TH41 frequency changer. VP41 IF amplifier. HL41DD detector and delayed AVC rectifier, AF amplifier gram only. The tuning indicator is a Mazda ME41.
The amplifier is quite simple. A later model, the 850G has a similar amplifier circuit except for the added refinement of negative feedback.

Till Eulenspiegel.

Hmmm. So really, this'll be a 1046 in name only, won't it...?

I now have the original amplifier chassis, albeit it has been stripped of all it's components. So it follows I could rebuild the chassis almost exactly like the original, that is with four volt valves and a pair of PX4 triodes. Well, there is no way I'm employing the over-priced and scarce PX4 so it'll have to be the triode connected KT66 or 6L6. In the original circuit the audio amplifier valve was an SP41. In a 6.3 volt heater environment I can use the 6.3 volt SP61, or, better still the low noise EF37A/CV358 pentode. http://www.radiomuseum.org/tubes/tube_ef37a.html The original phase-splitter was an HL41 triode, an HL61 6.3 version was never made so it will have to be the 6J5 or L63. Or, perhaps a para-phase splitter circuit employing the 6SN7GT double-triode?
According to the RGD museum a later version of the 1046G was made, the 1046G3. The radiogram employed a three speed Garrard record player and the amplifier had separate bass and treble tone controls. http://www.rgd.org.uk/rgd_exh_pg_1046G3.htm

Till Eulenspiegel.

... The nice thing about the DL1 is that it has ... a twin-t notch filter for needle scratch ...

I'm puzzled by this. A twin-t notch filter is used to produce a very sharp and, therefore, very narrow, notch yet 'needle scratch' will essentially cover a range of frequencies, all of which need to be removed if the filter is going to be effective, surely?

On this subject, I got a surprise listening to my MP3 player the other night. A lot of the content has been transcribed from numerous sources including my late brother's CD collection which contains, amongst other things, several magazine cover compilations including several classical collections. The player contains around 1100 tracks at the last count and I usually have it on random play but it is quite obvious I haven't heard everything on it yet!

The track I heard was Menuhin and the LSO playing Elgar's Violin Concerto in B Minor - which I noticed was in mono. What I didn't expect, a little over halfway through was, during a quiet patch, to hear the once familiar ssshhh - ssshhh - ssshhh - ssshhh - of a 78rpm record nearing its end!

The piece is 12 minutes long and I've listened to it again, very carefully, but apart from a few almost indiscernible clicks, the rest of it is fine!

One thing I can't detect is where the joins are between what I would imagine was four 12" record sides, so great care was obviously taken in the transcription but it is a bit strange that there wasn't a blameless copy of the master available to transcribe it from!

Hmmm. So really, this'll be a 1046 in name only, won't it...?

I now have the original amplifier chassis, albeit it has been stripped of all it's components. So it follows I could rebuild the chassis almost exactly like the original, that is with four volt valves and a pair of PX4 triodes. Well, there is no way I'm employing the over-priced and scarce PX4 so it'll have to be the triode connected KT66 or 6L6. In the original circuit the audio amplifier valve was an SP41. In a 6.3 volt heater environment I can use the 6.3 volt SP61, or, better still the low noise EF37A/CV358 pentode. http://www.radiomuseum.org/tubes/tube_ef37a.html The original phase-splitter was an HL41 triode, an HL61 6.3 version was never made so it will have to be the 6J5 or L63. Or, perhaps a para-phase splitter circuit employing the 6SN7GT double-triode?

How about staying with 4 volt Mazda octals - but substitute triode-connected PEN44s or 45s? At least all the valve-holder holes will be the same size

... The nice thing about the DL1 is that it has ... a twin-t notch filter for needle scratch ...

I'm puzzled by this. A twin-t notch filter is used to produce a very sharp and, therefore, very narrow, notch yet 'needle scratch' will essentially cover a range of frequencies, all of which need to be removed if the filter is going to be effective, surely?

I wondered that myself, but maybe the idea is that it's to tune out a predominant band of frequencies. I'm wondering just how sharp the notch is to be honest, but I'm wondering aloud here - would the grain structure of the filler have a predominant frequency band? It'd be an interesting, if largely academic, exercise to run a spectral analysis of the noise on the surface of a 78.

Of course, if they're nice laminated Columbia discs, needle scratch won't be an issue.

How about staying with 4 volt Mazda octals - but substitute triode-connected PEN44s or 45s? At least all the valve-holder holes will be the same size

There is certainly nothing wrong with those 4 volt Mazda octal valves. Not sure about a triode connected PEN45, sure the PEN44 is beefy enough, but scarce. I believe the AC4/PEN is similar
There is another 4 volt output valve that might be well worth considering, the Mullard AL60 or CV9. Langrex have some.

Add the Baxandall NFB tone control circuit?

Add the Baxandall NFB tone control circuit?

If it can be done unobtrusively, it'd be a good idea. Have you the centre-tapped pot?

Add the Baxandall NFB tone control circuit?

If it can be done unobtrusively, it'd be a good idea. Have you the centre-tapped pot?

There are two cut-outs on the left side of the cabinet, one is for the toggle mains on-off switch and the other for the volume control. The bass and treble tone controls can replace these controls and the volume control would then replace the original top cut tone control which was on the right side of the cabinet. Then I'll have to reposition the on-off switch.
To make up the tone control controls I've ordered two eleven position switches at the very reasonable price of £4.22p each from an Ebay seller in Hong Kong. Similar specification switches from RS were too expensive. The switches will be wired for five positions of tone boost and cut. The centre position is the "flat" setting. The RGD console radio model 750C has separate switched bass and treble controls. The model 850 radiogram has similar tone controls. So because it is the fact that these RGD sets have switched tone controls I have decided that it would be a good idea to incorporate a similar style of tone switching.

Hi Jeffery,
It would appear that the centre tap treble control was in common use in early Baxandall tone control networks, mainly the circuits with negative feedback. The circuit shown in the attachment is from a 1952 Wireless World magazine.
The centre tap is easy with the switched resistors I intend to use. In 1955 Pye produced a high fidelity system which comprised of a pre-amplifier and power amplifier. Models PF91 and PF91A. The pre-amplifier has the Baxandall tone control circuit with the centre tap treble tone control. The circuit is very similar to the one that appeared in the 1952 Wireless World.

Till Eulenspiegel.

Pictures of the amplifier chassis as it is now and how it was before it was dismantled.

Till Eulenspiegel.

Well it seems to the designer of the Pye PF91A pre-amplifier decided to copy the Baxandall tone control circuit component by component exactly as the original as it appeared in the 1952 Wireless World article. Only significant difference is the choice of valves, the rare ECC40 double triode instead of the SP61 pentode.
The ECC40 is similar but not directly equivalent to the 6SN7GT and ECC82/12AT7.

Till Eulenspiegel.

A very good job on the amplifier. Looks really fine.

Gary

Hi Gary,
I'm sorry to say the pictures of the complete amplifier are not of mine. I should have wrote: this is how an RGD 1046 amplifier looked like before it was dismantled. It is unlikely my reconstructed amplifier will look exactly like the original.
I still have to decide which type of output valves to use. Triode connected KT66s or a common type like the 6L6. The amplifier pentode and phase triode splitter will be the original types, SP41 and HL41.

Till Eulenspiegel.

Very interesting, Andy. Thank you.

The DL1 circuit is in Radio and Television Servicing - worth looking up.