Philips-Mullard TV tuners 1955-1960

[Pieter H]

Hi,

impossible to find any circuit diagrams of the Philips/Mullard TV's for the 1955-60 periode. I'm looking for the codes of the tuners used in these sets, both the product code (usually in the AT75xx or AT76xx range) and the factory code (A3 xxx yy if made in Eindhoven, probably MW xxx yy if produced in Micham).
As far as I know it concerns the following TV families:

• 1955/56: 1756/1757/2157U
• 1955/56: 1458/1758U
• 1956/57: 1468/1768/2168U
• 1956/57: 1772/1778U
• 1958/59: 1796/2196U
• 1959/60: 17TG100/17TG200/17TG300

Any help much appreciated!

Cheers, Pieter

[FERNSEH]

From the components list in the service manual for the Philips 2155U. Related models are: 1458U, 1468U and 1772U.
The tuner unit employed in these receivers is given as: complete tuner MK.892.82. Tuner unit less valves HY.142.26.
Spare parts for the tuner have MK prefixes. E.G. the tuning spindle assembly is MK.958.26. Only a few parts in these receivers have A3 prefixes.
Service manual is dated February 1957.

DFWB.

[Jac]

Originally Posted by Pieter H

impossible to find any circuit diagrams of the Philips/Mullard TV's for the 1955-60 periode. I'm looking for the codes of the tuners used in these sets, both the product code (usually in the AT75xx or AT76xx range) and the factory code (A3 xxx yy if made in Eindhoven, probably MW xxx yy if produced in Micham).

Pieter,

I should have the service manual for probably all the above types.
As well as for the AT7520 & A3.791.64 tuner modules.

Jac

[Heatercathodeshort]

Just a few from manuals that are at hand. The rest are at the back of the garage..

1792/2192U Feb 1959 Including FM radio Less valves. HY.142.38

1756U August 1955. MK.892.71

1768/2168 October 1967. HY.142.30

Note. Later 1768U models were fitted with a tuner that employed printed circuit clip in coils similar to the tuner fitted in the 17TG100U series. John.

[Pieter H]

That's a good start, thanks to all!
I assume all mentioned tuners are based on the PCC88-PCF80 valve set.

Interesting to see that Philips/Mullard used a two-step approach for the tuner manufacturing. Apparently the mechanical and electrical assembly was done in a first factory, coded HY, while valve insertion and subsequent tuner alignment were done in the Mitcham factory, coded MK. Does anybody have an idea about which factory this "HY" might be? Alternatively, but less likely, this was also Mitcham, with HY the factory code for unfinished sub-assemblies and MK for complete modules.

Jac, I'll contact you off-line.

[FERNSEH]

Doesn't help much, but here's a picture of the miniature tuner fitted in the Philips 19TG108U. Valves are PCC89 and PCF80. Made in 1961 the 19TG108U was a development of the 1959/60 model 17TG100U.

DFWB.

[Heatercathodeshort]

The standard valve line up in the UK from 1954-59 was PCC84 and PCF80.

The frame grid PCC89 did not appear until 1959 but strangely the earlier PCC88 was overlooked by all manufacturers other than in a range of very high gain Pye models introduced in 1958. John.

[Pieter H]

Hi John,

something interesting happened here.
Contrary to your statement, the "continental" Philips tuners changed to the frame grid PCC88 (in combination with the PCF80) in 1956 and it was used massively in two generations of tuners until 1963. Apparently Mullard stuck to the PCC84, for reasons not clear to me.

In contrast, Mullard was the first to switch to the PCC89 in 1959 (still in combination with the PCF80), whereas the "continentals" waited for the PCC189. This one was introduced in 1960 in the French tuners from the Dreux factory, again still in combination with the ubiquitous PCF80.

In 1961 the Dutch and German factories then finally changed over to the new combination PCC189-PCF86, both frame grid valves. I still have to find out where the PCC89 and PCC189 differ, but it is clear that the latter quickly replaced the 89.

To be continued!
Cheers, Pieter

[Heatercathodeshort]

Sorry Pieter, I should have said the PCC88 was overlooked by British manufacturers. I can't understand why due to the much superior gain of the PCC88 over the PCC84.
Philips manufactured truly excellent tuners. John.

[AC/HL]

Could it have been that our service areas were smaller, or perhaps the smaller bandwidth requirements of the 405 system?
They would have been forced to change when the PCC84 inevitably went from current to maintenance status, possibly the 88 wasn't far behind it?

[FERNSEH]

I believe the only difference between the PCC89 and PCC189 is the pin connections to the electrodes.
PCC89: https://www.radiomuseum.org/tubes/tube_pcc89.html Pin connections same as PCC84.
PCC189: https://www.radiomuseum.org/tubes/tube_pcc189.html Pin connections same as PCC88.

DFWB.

[Pieter H]

Thanks a lot for your inputs, the picture is getting clearer.

So the Philips tuner organization at large made the switch from PCC84 to PCC88 in 1956, which required a re-arrangement of the valve socket connection. However, with the socket rotated 180 degrees the connections were roughly identical, so in that respect the effort was limited. But the benefit was considerable, as John pointed out, due to the almost doubled effective gain of the frame grid construction. When the PCC88 was replaced by the pin-compatible PCC189 that was an almost drop-in replacement. This switch-over was made in 1960.

I agree that it's difficult to judge why the Mitcham tuners stayed with the PCC84 for another 3 years. Every TV standard, irrespective of the channel bandwidth, should have benefited from the improved valve. So the most plausible explanation is an economic one. The UK Philips sets had stayed relatively long, compared to the continent, with single channel non-tunable front ends, the first tuner-equipped family being the 1446/1746/1747 in 1954, three years behind the Netherlands and Germany. Given the modest initial volumes it is not unreasonable to assume that in 1956 the first generation investments were not yet written off and a conversion to the PCC88 not justifiable. A "thin" explanation, I admit, but for lack of a better one.

But even then it is difficult to understand why the Mitcham tuners even stayed with the PCC89, because that valve, like the PCC84, had the screen connected to the second gate. In the PCC88 and PCC189 the screen had a separate pin for direct grounding, which was in principle more effective. So a follow-up question: did the UK tuners stay with the PCC89 or did they also switch over to the PCC189, and if so when?

Cheers, Pieter

[Synchrodyne]

From its TV AGC papers of 1957-58, one may infer that for positive/AM TV systems, Mullard preferred to use remote cutoff valves in AGC’d positions, although the sharp cutoff type was seen as being acceptable for negative/FM systems.

The PCC84 comprised a pair of sharp cutoff triodes that when operated in series cascode, had a net remote cutoff characteristic.

The PC88 had sharp cutoff triodes that in combination provided a sharp cutoff result.

The PCC89 had remote cutoff triodes that in combination produced a curve similar to that of the PCC84

So, given its preference, it seems reasonable that for UK domestic production, Mullard would have stayed with the PCC84 until the PCC89 became available, rather than switching to the PCC88.

Mullard described the PCC189 as the PCC89 with a different pinning arrangement that facilitated alternative positioning of components in tuner units.


Cheers,

[Pieter H]

Hi,
I've seen this discussion before and initially assumed the statements were correct. However, diving deeper into the matter I think I have to disagree with your statement that the remote vs. sharp cut-off was valve-dependent. What I found in the Philips valve documentation is that it is circuit dependent!

See the first attached picture, illustrating the PCC84 Vg-Ia characteristic for circuit configuration A, where the cascode grid is biased from a series resistor in the cathode chain, vs. configuration B where the grid is independently biased. The curves show a remote AGC cut-off behaviour for A and a sharp AGC cut-off for B.
ALL Philips tuners used configuration B, albeit with over time higher resistor values of 1M. At least the continental (non-UK) tuners. I don't posses circuit diagrams of the Mitcham tuners yet, so can't verify whether they used configuration A to obtain remote cut-off as you state. You might well be right, but then it's not a valve but circuit issue.

This behaviour did not change with the new valves like the PCC88 and PCC189, which had almost identical DC behaviour as the PCC84 and PCC89. The second picture shows the (delayed) AGC curves for a PCC88-equipped tuner in combination with EF80 IF valves, clearly showing the 5V tuner AGC cut-off. When the EF184 was introduced that valve required a much higher (20V) cut-off whereas the equivalent PCC189 AGC control voltages remained identical.

Cheers, Pieter

[Synchrodyne]

The PCC89 data show a range of grid base numbers according to the exact circuit details, but the range is from 9 to 15 volts for divider bias in the upper triode grid, and around 16 volts for sliding bias. Given that the triodes themselves are of the remote cutoff type, this is not so surprising. The Mullard PCC89 was announced in Wireless World 1957 October, p.474, including the comment that it had a variable-mu frame grid construction to assist in the reduction of cross-modulation effects. Presumably that claim applied for both the divider bias and sliding bias circuit configurations.

In its original work on the series cascode circuit, RCA emphasized the sliding bias option, observing that this resulted in a stretched grid base and what was close to a square-law curve, which minimized cross-modulation. I imagine though that the divider bias option was attractive to some setmakers in that with its short grid base, around 5 volts for the early cascode double triodes, as compared with 10 volts or so for sliding bias, it required less gain from the AGC system.


Cheers,

[dragonser]

Hi,
while looking through some paperwork I found the following service sheet which I thought might help...
But it is for a later version of the tuners and is dated 1963.
I thought I would post it anyway !

[Pieter H]

Hi Peter,

thanks a lot, great addition since I didn't have this one yet. It remains interesting to see that Mullard apparantly offered its tuners without valves (coded HY xxx yy, the code became MK 892 zz once the valves were inserted), this in sharp contrast to at least some factories on the continent, where the valves were inserted in one of the first steps and left throughout the electrical assembly and alignment process. See my web site for detailed pictures of that.

Having the valves in place throughout production seems odd, given the increased risk of damage. However, apparently that was not such a big issue, and the advantage was that especially the pins of the valve sockets, which as we know were often not hard fixed in especially the ceramic sockets to allow valve pin tolerances, were much more stable for soldering with a valve inserted.

In the end the Mullard approach seems more problematic. It means that factory alignment took place with valves that were removed before delivery, requiring at least some level of verification and re-alignment once the final valves were inserted in the television factory. Doesn't look very efficient to me.

Cheers, Pieter

[Heatercathodeshort]

The tuners may have employed different heater rating valves, PCC84/PCF80 or UCC84/UCF80 or maybe even the E series for export.

I must admit I have never had to realign a Philips or thinking about it, any make of tuner after replacing just the valves. Philips tuners were incredibly stable and their range of valves consistent to the point that it was rarely necessary to adjust the user fine tuner on Band 3 after replacing the oscillator/mixer, typically a PCF80.

Dummy stainless steel 'valves' were supplied to setmakers but I doubt if they used them due to the valve holders being far more robust than at first thought. J.

[Maarten]

Just wondering... Were those tuners produced in Micham or did Mullard use other production sites as well?

Coincidentally, does anyone know what was made in Dunfermline? Tuners by any chance?

[Maarten]

I think that should have been "incidentally", right?

But anyway, in https://www.vintage-radio.net/forum/...00&postcount=9 I found that tuners would have indeed been made in Dunfermline. Does anyone have an idea from what year on?