Dormer & Wadsworth (D&W)

[John Caswell]

A very minor correction to David's comments above. VITS = Vertical Interval Test Signal.
The manufacturer was Gorler in W. Germany (at the time) and their units, both valve and semiconductor were a staple diet in many West German radios as well as B&O, Heathkit etc
Larsen and Hoedholt (L&H) the Danish manufacturer produced both "front ends" based on dual gate MOSFETS and a complete FM tuner module using that front end, plus the ubiquitous CA3198E for the IF and a KA2265 stereo decoder, you added varicap voltage controls, plus meters etc. and a frequency counter.
Using this module, the L&H 7255, is was possible to build, which I did, a state of the art FM tuner which knocked spots off off most things available at the time.
I also built, a bit earlier, a mono valve FM tuner from a circuit in the American "Radio Electronics " magazine that had 3 RF stages, plus bandpass coupled IF stages. It caused me a few headaches in the beginning, but goodness me how sensitive and selective it was, if signal was passing the aerial it was sucked in. Useless for stereo of course.
On another of David's points about small worlds, I would never have an affair as I know I could walk out of a hotel on the dark side of the Moon, and the next door would open and comment like "****** me John what are you doing here" would come hurtling out
Sorry for the digressions.

John

[Synchrodyne]

I think that Görler FM front ends were also used by some of the UK manufacturers, such as in the Wharfedale WFM-1 tuner. So D&W would have been competing with Görler as well as other UK manufacturers.

Given that D&W was a relatively obscure organization, with no apparent electronics history prior to its entering the FM front end field, I wonder if it was say an engineering company that for one reason or another elected to enter a specialized electronics area. FM (and TV) front ends were as much about mechanical engineering as as electronic engineering. That was the case with Cyldon, originally an engineering company that stepped into the TV front end market c.1953 with a turret tuner based upon the Amercian Standard Coil prototype, ahead of any similar move on the part of the established setmakers.


Cheers,

[Synchrodyne]

Looking at the chronology of the D&W solid-state FM front ends, it is known that the silicon bipolar version of the D&W 341/AFC FM front end was used in the Radford FMT1 FM tuner. This became available early in 1966, as shown in this excerpt from the 1966 Audio Fair catalogue.



This timing would have made it a reasonably early application of silicon planar transistors.

The two dual-gate mosfet version of the same front end, the 341/AFC/FET/2 was used in the Rogers Ravensbourne 2 FET FM tuner, which was shown for the first time at the 1968 Audio Fair.



The same 341/AFC/FET/2 front end was used in the Radford FMT3, which was listed (at that time alongside the FMT2) in HFYB 1970.

A FET-based front end, presumed to be the 341/AFC/FET/2, had been offered as an option on the Radford FMT2 from early 1967, as shown in this excerpt from the 1967 Audio Fair Catalogue.



If the presumption is correct, then this would have been a very early application of dual-gate mosfets in an FM front end. What appears to have been RCA’s initial commercial release of a dual-gate mosfet was the TA2644, announced in Electronics World 1966 October. TA2644 was in fact the development type number for what became the 3N140, although when the latter name was applied I do not know. RCA had used the TA-series development type numbers quite often when discussing mosfets in its publications, and they were shown in early edition datasheets under the definitive type numbers. The RCA 4060n series of dual-gate mosfets (3 for TV applications, 2 for FM) were in the 3N140 group, although the three TV types had their own TA-numbers as well. Anyway, it does appear that the basic type of dual-gate mosfet used in the 341/AFC/FET/2 was commercially available by early 1967. Of course, it is possible that D&W had used jfets in 1967, such having been used in American FM tuner front ends since late 1965.

On the basis of circumstantial evidence, yet to be corroborated, the varicap derivative of two-mosfet 341 front end was used in the Sudgen R21/R51 tuner. This was first shown at Sonex 1971 (as reported in WW 1971 April), with the first HFYB listing being 1972. This was not a very early use of varicap tuning in a domestic product. For example, AREF (Denmark) had offered varicap-tuned FM front ends from late 1967 (as reported in WW 1967 December). However, as early varicap applications were reputed to adversely affect signal-handling and spurious response performance, D&W (and its customers) may have chosen to wait for some progression along the learning curve. On the other hand, from the start, dual-gate mosfets offered benefits as compared with small-signal bipolar devices.

Re D&W’s use of bandpass input tuning with a single-tuned interstage in its 341 series, support for this approach may be found in a 1966 RCA paper “A Comparison of Solid-State and Electron-Tube Devices for TV-Receiver RF and IF Stages”. Therein it was said, in respect of the poorer cross-modulation characteristics of bipolar transistors as compared with valves:

“Further improvement in cross-modulation can be achieved, at the cost of additional sacrifice in noise figure, by use of additional pre-selection prior to the transistor input. For this reason, it is good design practice to use a double-tuned input circuit in transistor rf stages. As shown in Table IV, a noise figure of 3 dB is typical for bipolar transistors. Therefore, a degradation of as much as 1.5 dB will still provide a device comparable to tubes.”

Although RCA was discussing VHF TV front ends, I think that as Band II is between Bands I and III, interpolation to the FM case is reasonable. Normal practice for valved TV front ends was a single-tuned input with a double-tuned bandpass interstage, and that was retained when dual-gate mosfet RF amplifiers were introduced (initially by RCA in 1968).


Cheers,

[Radio Wrangler]

This is where we get into geographic differences. In some parts of the world with a congested FM band full of commercially competitive stations in a power race, tuners with all the front-end selectivity you can get are needed. In other places with sparsely-populated and well-ordered bands, noise figure is to the fore to give people in fringe areas better S/N ratio.

I find the 'Tuner Information' website amusing with all their shoot-outs trying to find 'The Best' tuner. It's a hopeless pursuit without splitting it up into categories, however they have lots of interesting information on there.

David

[Ted Kendall]

Originally Posted by Radio Wrangler

I find the 'Tuner Information' website amusing with all their shoot-outs trying to find 'The Best' tuner. It's a hopeless pursuit without splitting it up into categories, however they have lots of interesting information on there.

Curious site, that. Lots of useful stuff, agreed, but some of the assertions about subjective audio quality are risible, given the lack of control over programme material. Mind you, they aren't alone - I saw a site somewhere which largely consisted of one man's pugnacious assertions about the sound quality of practically every receiver brought to market between the 1960s and the 1980s...

[Radio Wrangler]

I find it very difficult to keep a straight face when confronted with assertions about subjective audio quality, so I stopped even trying.

When someone next praises a sound for 'air', cock your head on one side, listen intently for a while and then question whether it's normal air... is the oxygen/nitrogen ratio correct? it does sound a little stale. Do this totally deadpan (if you can!) and see how they react. If you act like a pundit, their training kicks in and they have to accept what you just said.

I have some quite decent speakers and an amplifier which can handle them. No brand marks anywhere but very definitely well made. It's quite amusing when you come across people who don't know what to say about what they've been hearing because they would have to see the make and model names before they would know what they are supposed to say they've heard. It's either very very funny or very very sad. And I can't decide which.

David

[Synchrodyne]

The simpler, three-gang D&W FM front end used in the Rogers Ravensbrook FM tuner had a single-tuned input, a dual-gate mosfet RF amplifier and a bipolar mixer. It seems likely that it had an all-bipolar predecessor of otherwise similar configuration. If so, this might have had the lowest noise, but the poorest strong signal performance of the group of 3- and 4-gang models.

Conceivably D&W might also have offered a two-gang all-bipolar FM front end with an aperiodic input, although that type would have been more for setmaker than for hi-fi applications.

D&W was not alone in choosing a bandpass input and a single-tuned interstage for its four-gang, mosfet FM front ends. Armstrong did the same for its 600 series, I think using an in-house design and not a third-party unit.



Cheers,

[Synchrodyne]

It is known that D&W were active in the valve era, and upthread it was speculated that the Radford FMT1 valved FM tuner might have used a D&W front end. That is, Radford’s FM tuner essay started with a D&W front end and it stayed with that supplier into the solid-state era with its FMT2 and FMT3 tuners.

By similar reasoning, it is possible that Rogers had used valve D&W FM front ends before its use of same for its early solid-state models. I think that with its early FM tuners, Rogers might have used its own front end designs integrated with the chassis, but later on made the change to third-party units.

A comparison of the Radford FMT1 and Rogers Mk II front ends shows that they were very similar, and probably from the same source, which could well have been D&W. The Rogers version had RF AGC, whereas the Radford version did not. Presumably that feature was optional.



Both of these tuners appear to have been presented at the 1962 Audio Fair. The single-valve (ECC85) front end was more of an economy than a performance design, so seemingly out of place in higher performance hi-fi products. However, where the use of a third-party unit was necessary, that type was probably unavoidable simply because more elaborate designs were not offered by the unit manufacturers. It could be that D&W had designed its ECC85-based FM front end to extract maximum performance from the configuration, thus aiming at more of a niche market. It may not have wished to compete head-on with the likes of Plessey in the mass market. Certainly, the D&W solid-state tuners, particularly the two-FET models, were comparable to other hi-fi units of their time.

Perhaps to offset their relatively simple front ends, both the Radford FMT1 and Rogers Mk II had four-stage IF strips, unusual in British practice (but not in American practice). The following Rogers Mk III had a similar valve line-up to the Mk II, so quite probably used the same or a similar front end. It had AFC, which was probably another option on the third-party front end.

Another receiver with very similar front-end circuitry was the Dynatron TRV14 Pathfinder of mid-1961.



That makes the TRV14 another candidate for having used a D&W FM front end. Previously Dynatron appears to have designed and built its own FM front ends. In the case of the FM1 and FM2, these were 3-valve units integral with the chassis. The T10 AM-FM tuner had a single-valve unit that looks to have been of the bolt-on type but probably made by Dynatron. The T10A AM-FM tuner and T11 AM-FM Tuner/Control Unit had a two-valve “bolt-on” unit known as the FE2, also probably of Dynatron manufacture. With the TRV14, Dynatron was heading downmarket (particularly on the AM side), and economics might have dictated the use of a third-party FM front end.


Cheers,

[Radio Wrangler]

The Radford circuit diagram above looks a little dodgy, there is no DC path for the grid of the first half of the ECC85. No AGC feed, no grid leak. All paths have series capacitors.

The Rogers diagram has C5 on the AGC line, but it looks to be at the wrong end of the AGC input resistor. As it is C5 winds upin parallel with C3. This would not seem to be good for the input matching tap and the neutralisation bridge. It'll also introduce a resonance.

The solution may be that the Radford diagram lacks an AGC injection point at the vertex near to the capital 'T' of 'Trimmer'. The 100k resistor then makes sense.

David

[Synchrodyne]

I wonder if there were errors in the schematics.

Presumably the basic circuit was designed to accommodate agc if required. If agc facility was not used, then one might expect the agc feed resistor to be returned to ground to provide a DC path from the grid.

In the Rogers case, there was another 1000 pF capacitor between the agc line and ground on the other side of the 100 k agc feed resistor, but this was lost when I cropped out the front end from the full schematic. Even so, the 1000 pF on the grid side of the resistor would appear to ground the grid in respect of signal, whereas the basic circuit looks as if it is of the in-between kind, neither fully grounded grid nor fully grounded cathode, but with both electrodes driven.


Cheers,

[John Caswell]

More info update
Just had a look at a Dynatron T10 unit I happened to have handy in the workshop (as you do)
The FM front end is a permeability tuned Gorler unit the UT340 using the standard West German ECC85. I understand for the T10A and certainly according to the schematic, Dynatron made the two valve FM front end, using an EF80 as the RF amp and an ECC85 as the oscillator and mixer. This also was inductively tuned and was called the FE2.
I have to admit finding this thread very interesting, as I my youth I was an avid FM dxer, mainly due to Austen Uden of Hi Fi News. I was unable to have a large aerial in the block of apartments where I first lived but once married and moved house, I erected a Fuba UKA8 on a rotator which fed my TSL mono FM tuner. Still have the same aerial system but tuner is now a Magnum Dynalab MD100
John

[GW4FRX]

Originally Posted by Synchrodyne

I wonder if there were errors in the schematics.

Whatever the virtues of Radford products, generating accurate schematics didn't seem to be a specialite de la maison in Bristol.

[Radio Wrangler]

Inaccurate schematics seem to be a theme running across all hifi. A lot of manufacturers guard accurate information like a secret rocket fuel... they act as if their circuit was some great secret that couldn't be designed by anybody else, or bettered for that matter, only copied once Boris and Natasha had got their hands on the secrets.

Part of the game of course is bigging-up their products with the idea that if they are seen to be guarding things, there must be something clever being protected. Nil combustibus profumo!

Of course, there are honest-to-goodness whoopsies which the Radford ones are likely to be. but there was a recent thread on here about an amplifier with so many circuit variations, no-one knew which one was right, and the earlier examples would have been truly horrible (transistor amps with no bias and monstrous crossover distortion) these had to have been deliberate misinformation, didn't they? No-one would have bought anything so bad?

David

[GW4FRX]

Originally Posted by Radio Wrangler

Of course, there are honest-to-goodness whoopsies which the Radford ones are likely to be....

That's almost certainly true. By the look of some of the boards I've seen, the issue seems to be largely one of poor version control and the documentation not keeping up with design and component changes. But in the case of the SC.24 and SPA.50 (and possibly also the FMT2) the schematics and the hardware are alarmingly divergent.

I don't know whether similar strictures apply to the earlier thermionic amplifiers; perhaps their design was more settled early in their lives and component changes were less of an issue.

[Synchrodyne]

Originally Posted by John Caswell

Just had a look at a Dynatron T10 unit I happened to have handy in the workshop (as you do)
The FM front end is a permeability tuned Gorler unit the UT340 using the standard West German ECC85. I understand for the T10A and certainly according to the schematic, Dynatron made the two valve FM front end, using an EF80 as the RF amp and an ECC85 as the oscillator and mixer. This also was inductively tuned and was called the FE2.

The Dynatron FM front end progression looks rather odd in hindsight. For the FM1 and FM2 switched tuners, it designed its own three-valve form (2 x Z719 + 1 x B309) that was integral to the chassis. Then it went to a Görler single-valve unit for the T10. Perhaps this was seen as not good enough, hence the change to its own FE2 in the T10A (and I think also in the T11). But then it went back to the single-valve type, provisionally assumed to be from D&W, for the TVR14, for which corner-cutting was probably the order of the day.

Anyway, there is circumstantial evidence at least that D&W was active in the FM front end business as far back as 1961-62.

At the other end of the timescale, as said there is strong evidence that a D&W front end was used in the Sugden R21/R51 FM tuners. The R51 production life was 1971 through 1981, although of course it might not have retained the same front end throughout. The T48, introduced in 1976, had a somewhat different front end that looked rather like the Toko EF5603.



There could have been several reasons why Sugden made a change from D&W – price, performance, supplier diversity and so on. And possibly the same change was made in the R51 at that time, for the same reasons. But it could also have been that D&W was exiting the business, so that change was unavoidable. So we have c.1976 as maybe date for when D&W made its exit. If so, it would also have affected the Cambridge T55, at least if the working assumption that initially used a D&W front end is correct.


Cheers,