Anyone used 6T8 valves?

[Boater Sam]

6T8 triple diode triode similar to EABC80 but shorter.

Any experience with these?
Sam.

[usradcoll1]

That valve has been in use in the US for ages. The manufacturers only had to furnish one valve and socket for the ratio detector and first audio, instead of a separate dual diode such as a 6AL5, EAA91 and a high mu first audio triode in both TV sets and AM-FM radios.
The Zenith AM-FM table radios used a 19T8 for a radio detector, AM detector and first audio stage.

[Synchrodyne]

As far as I know, the 6T8 was the first noval-based triple diode-triode, released in 1947 by GE:



It was preceded by the octal-based 6S8GT, a post-WWII release during that brief interlude when it was thought that new valves requiring more than 7 pinouts would continue to use the octal base, that coming to an end when the noval base suddenly appeared on the scene, apparently unexpectedly.

From what little I have seen, those Zenith post-WWII valved chassis might be the basis of an interesting narrative. The 7-valve FM-only models also used the 19T8, with one diode unused, presumably as a convenient way of obtaining FM demodulation without requiring an extra valve. There were later double diode-triodes with separate cathodes, such as the 6FM8, that could have been used, but they seem not to have attracted Zenith’s attention. The 7-valve AM-FM models had a shared front end, 12BA6 + 12BE6 very early on, then 12BA6 + 12AT7, then 6BJ6 + 12AT7, and perhaps others. Interesting is that in most cases there seemed to be variants both with and without AM RF amplifiers. In the without case, the AM signal went direct to the 12AT7 (triode) mixer section. Late 1950s or early 1960s, Zenith seemed to have adopted single-valve FM front ends such as the 12DT8, and with it separate AM front ends. The with or without AM RF amplifier options mirrored its AM-only table receivers, which as far as I know into the early 1960s could be had in AA5 form or 6-valve with an RF amplifier. That contrasts with the European setmakers, many of whom seemed to have developed an aversion to AM RF amplifiers by the later 1950s.

Zenith worked with GE on some valve-related projects, notably the 6BN6 gated beam valve in the 1940s, so it was probably kept apprised of GE’s FM work that led to the 6T8/19T8 and 12AT7, allowing it to become an early user. One wonders if its (unfortunate?) predilection for AC-DC chassis had anything to do with the fact that the 12AT7 had a four-way heater that made it suitable inter alia for 150 mA series strings. Zenith also worked with GE on the 6BU8 TV valve. But despite the history of cooperation, the Zenith-GE pilot tone FM stereo system was not a joint effort. It was the FCC who combine the two very similar systems, after which there was evidently some cross-sniping between the two which each claiming to have been the real inventor.

Brimar offered the 6T8 in the UK, and I imagine it might have been used by some of the setmakers in the Brimar camp.


Cheers,

[G8HQP Dave]

I seem to recall reading somewhere that the 6T8 is worse for hum than the EABC80. Maybe the latter was a European improvement on the former, just as the ECC83 was an improvement on the 12AX7?

[Leon Crampin]

Other than a difference in length (6T8 is shorter) there does seem to be a difference in the diode sections of these valves.

With reference to the Brimar book No 6, the 19T8 (6T8 by reference) has a diode current rating of 5 mA per section - implying that the 3 diodes all have the same characteristics.

Conversely, the EABC80 in the same reference book has a diode rating of 1 mA for the AM detector, and 10 mA for the ratio detector diodes. The triodes are the same for each device, based on the 6AT6.

I suspect that when Philips redesigned the 6T8 for European use, coding it EABC80, they realised the importance of low impedance diodes for good ratio detector performance. The 6AL5 is somewhat lacking in this respect.

With the benefit of time as the ultimate testing authority, my findings are that most EABC80s reach the end of their useful life due to FM diode mismatch - probably caused by low emission. The detector becomes unbalanced and the set will not tune correctly. The AM detector diode and the triode seem to work adequately with very little emission.

I have not encountered enough 6T8 devices to know if this problem extends to this device - but it was a smart move on the part of Philips to give the EABC80 the same pinout. I wonder why it was bigger though...

Leon.

[turretslug]

Could making the envelope slightly bigger have been a sneaky way of side-stepping licensing agreements/controversy- as in small-print marketing of something that could be excused as not an obvious identical interloper in certain markets?

[usradcoll1]

Originally Posted by Synchrodyne

As far as I know, the 6T8 was the first noval-based triple diode-triode, released in 1947 by GE:

Attachment 166422

It was preceded by the octal-based 6S8GT, a post-WWII release during that brief interlude when it was thought that new valves requiring more than 7 pinouts would continue to use the octal base, that coming to an end when the noval base suddenly appeared on the scene, apparently unexpectedly.

From what little I have seen, those Zenith post-WWII valved chassis might be the basis of an interesting narrative. The 7-valve FM-only models also used the 19T8, with one diode unused, presumably as a convenient way of obtaining FM demodulation without requiring an extra valve. There were later double diode-triodes with separate cathodes, such as the 6FM8, that could have been used, but they seem not to have attracted Zenith’s attention. The 7-valve AM-FM models had a shared front end, 12BA6 + 12BE6 very early on, then 12BA6 + 12AT7, then 6BJ6 + 12AT7, and perhaps others. Interesting is that in most cases there seemed to be variants both with and without AM RF amplifiers. In the without case, the AM signal went direct to the 12AT7 (triode) mixer section. Late 1950s or early 1960s, Zenith seemed to have adopted single-valve FM front ends such as the 12DT8, and with it separate AM front ends. The with or without AM RF amplifier options mirrored its AM-only table receivers, which as far as I know into the early 1960s could be had in AA5 form or 6-valve with an RF amplifier. That contrasts with the European setmakers, many of whom seemed to have developed an aversion to AM RF amplifiers by the later 1950s.

Zenith worked with GE on some valve-related projects, notably the 6BN6 gated beam valve in the 1940s, so it was probably kept apprised of GE’s FM work that led to the 6T8/19T8 and 12AT7, allowing it to become an early user. One wonders if its (unfortunate?) predilection for AC-DC chassis had anything to do with the fact that the 12AT7 had a four-way heater that made it suitable inter alia for 150 mA series strings. Zenith also worked with GE on the 6BU8 TV valve. But despite the history of cooperation, the Zenith-GE pilot tone FM stereo system was not a joint effort. It was the FCC who combine the two very similar systems, after which there was evidently some cross-sniping between the two which each claiming to have been the real inventor.

Brimar offered the 6T8 in the UK, and I imagine it might have been used by some of the setmakers in the Brimar camp.


Cheers,

Your research on the evolution of the post-war Zenith AM-FM receivers is spot-on!
I have many Zenith radios from that era. The last of the valve-based models did use the 12DT8 FM front end, but they changed the 19T8 to a 14GT8 valve. I don't have my later schematics here, so I can't see if there's a difference.

[G8HQP Dave]

Would lower impedance diodes require a slightly bigger cathode so need a bit more room inside the glass? Perhaps a bit more spacing also reduces hum?

[Leon Crampin]

Possibly, but the heater wattage allocated to the FM diodes in both the 6T8 and the EABC80 is the same - and also the same as the 6AL5/EB91, viz. 0.95W.

This may give a clue as to the poor long term performance of the EABC80 FM diodes, which have the highest anode current rating of the three devices for the same heater power.

I don't think the 6AL5 suffers from premature low emission as I have always found old tatty examples from overcooked TVs to be OK, even after a hard life.

Thinking further, I suspect that the EABC80 may well have better internal screening than the 6T8, which could account for its larger envelope. This would further explain the possible lower hum levels achieved with the European device.

Anode dissipation would not be a limiting factor in any section of these valves.

Leon.

[Synchrodyne]

The 6T8 and the ratio detector appeared at about the same time, the latter having been described in RCA Review 1947 June. So it is possible, perhaps probable that the 6T8 diodes were not designed with the ratio detector in mind.

Following American triple diode triodes might have been more accommodating of the ratio detector. The 6V8 of 1950 was described as having two high-perveance and one medium-perveance diode (as did the later EABC80) with the comment that the high perveance pair were suitable for use in balanced low-impedance detector circuits. Current for those two diodes at 5 volts applied was 40 mA, as compared with 20 mA for the 6T8, and 24 mA for the 6AL5.

The medium-perveance diode in the 6V8 had a current of 2 mA for 10 volts applied, whereas the third diode in the 6T8 was the same as the other two. It could be that to keep the heater current at 450 mA, and in turn the voltage for the 150 mA series-string version at 18.9 volts, a bigger proportion of that heater current had to be allocated to the two “FM” diodes in the 6V8 case, coming out of the allocation for the "AM" diode.


Cheers,

[Synchrodyne]

Originally Posted by usradcoll1

The last of the valve-based models did use the 12DT8 FM front end, but they changed the 19T8 to a 14GT8 valve. I don't have my later schematics here, so I can't see if there's a difference.

The 14GT8 was a double-diode triode with separate cathodes for each section, intended for use as an FM detector and 1st AF stage (72 mu triode). The diode current was 18 mA at 5 volts. Zenith appears to have stayed with the Foster-Seeley discriminator for FM until the stereo era, then switching to ratio detector. In some cases (e.g. MJ1035W) it then used the 6BC7 triple diode for FM and AM detection, each diode 35 mA at 5 volts. (Unlike GE, Zenith never seems to have used the FM limiter grid for AM detection.)


Cheers,