UK Vintage Radio Repair and Restoration Powered By Google Custom Search Vintage Radio and TV Service Data

Go Back   UK Vintage Radio Repair and Restoration Discussion Forum > General Vintage Technology > General Vintage Technology Discussions

Notices

General Vintage Technology Discussions For general discussions about vintage radio and other vintage electronics etc.

Closed Thread
 
Thread Tools
Old 29th Aug 2022, 10:42 pm   #41
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Triode Connected?

Quote:
Originally Posted by joebog1 View Post
Synchrodyne, The circuit from Hafler is more a cath amp design very similar to Quad. Requires another power supply voltage.
Although Hafler did equate it with the “conventional” approach where the screen is returned to a tap on the output transformer primary, fig. 2 in the patent.

He said: “Figures 1 and 2 represent then equivalent alternative constructions, Figure 2 being the simpler and more economical, but being subject to the limitation that the screen grid 5 and the plate 2 are at identical D. C. static potential.”

US2710312 19520520 Acro Hafler & Keroes Ultralinear.pdf

One could say that he had addressed right at the start the potential objection to the UL circuit that the anode voltage was limited to the maximum that the screen could handle.

I was pondering the customary practice of connecting the suppressor grid to the cathode when triode-strapping pentodes with separate suppressor pinouts. I think it is reasonable to assume that in normal pentode operation, the suppressor grid at cathode potential had negligible effect on the forward electron flow from the cathode to the anode, but does tend to stop the reverse flow from anode secondary emission. With triode strapping and the screen at very close to anode potential, I imagine that the suppressor grid at cathode potential would still have negligible effect on the forward electron flow, and so would not materially affect the triode nature obtained by that strapping. But it would still stop the reverse flow. Whether the latter was a material issue with triode-strapping I do not know. Presumably the distribution of current between the screen and the anode would need to ensure that the screen dissipation was not exceeded. So on balance, it might be safer to strap the suppressor to the cathode rather than to the anode, even though functionally, in terms of the overall transfer characteristic, there might not be a difference.


Cheers,
Synchrodyne is offline  
Old 29th Aug 2022, 11:10 pm   #42
joebog1
Banned
 
Join Date: Jun 2015
Location: Mareeba, North Queensland, Australia
Posts: 2,704
Default Re: Triode Connected?

I am actually wrong ( again ) about the circuit being similar to a Parry Cathamp as it was called. My apologies.

Joe
joebog1 is offline  
Old 30th Aug 2022, 6:50 am   #43
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

Quote:
Originally Posted by joebog1 View Post
He explains phase splitters that have uneven output impedances and it doesnt use a buffer to drive the output valves ?.
Lots of phase splitters have uneven output impedances (a good transformer is one of the few that doesn't!) however most of them preserve balance nevertheless.
kalee20 is offline  
Old 30th Aug 2022, 7:02 am   #44
joebog1
Banned
 
Join Date: Jun 2015
Location: Mareeba, North Queensland, Australia
Posts: 2,704
Default Re: Triode Connected?

Even driving EIGHT output valves from if I remember correctly a 12AX7, which once again from memory is something like 68K plate resistance. I guess the cathode will be lots lower.

I cant find the book right now, but I do remember the circuits ( to some degree ). I placed them in the "audiophool " category LONG before I joined this forum, perhaps before the forum existed.

I do appreciate your explanations and even criticism's though kalee. It keeps me checking up on what I have said. Its in fact, the best reason to be a member here!! One never forgets they are learning, CONSTANTLY.

With respect,

Joe
joebog1 is offline  
Old 30th Aug 2022, 8:12 am   #45
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,953
Default Re: Triode Connected?

Must admit, I have been following this thread with great interest - so many circuits and design considerations that I had never come across.

It's been wonderfully enlightening, thanks for all of you who have contributed (and sent me down a load of rabbit holes following the links you've posted!!)

That's the great thing about this forum, even us oldies can learn something new.
__________________
I'm the Operator of my Pocket Calculator. -Kraftwerk.
G6Tanuki is offline  
Old 30th Aug 2022, 1:47 pm   #46
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

Quote:
Originally Posted by Synchrodyne View Post
I was pondering the customary practice of connecting the suppressor grid to the cathode when triode-strapping pentodes with separate suppressor pinouts. I think it is reasonable to assume that in normal pentode operation, the suppressor grid at cathode potential had negligible effect on the forward electron flow from the cathode to the anode, but does tend to stop the reverse flow from anode secondary emission.
I'd agree! Though if g2 and anode were connected, there would be no reverse current flow anyway (tetrodes had the 'kink' when anode voltage fell below screen-grid voltage - and in triode operation with g2 linked to anode, that wouldn't happen...

Quote:
Originally Posted by Radio Wrangler View Post
In triode-strapped service, the screen can't be above the anode voltage, so it never appears more attractive than the anode.
... as Radio Wrangler has observed.

Quote:
Originally Posted by Synchrodyne View Post
Presumably the distribution of current between the screen and the anode would need to ensure that the screen dissipation was not exceeded. So on balance, it might be safer to strap the suppressor to the cathode rather than to the anode, even though functionally, in terms of the overall transfer characteristic, there might not be a difference.
Hmmm... I'm thinking the reverse, that if suppressor was strapped to anode/g2, that it would itself intercept any secondary emission electrons - either way the screen dissipation would be kept manageable!

Of course, there's the matter of suppressor-grid dissipation, if g3 current was at all significant - the suppressor is not designed to dissipate power (no radiators at the top of the support rods, for instance). Current measurement would be a good guide, methinks, and apply some 'sensibleness.'

It's worth considering that connecting g3 to cathode introduces another set of capacitances from the 'anode' to cathode: Cg2-g3 and Cg3-anode would be added. It may not be significant at audio frequencies, but the thread title isn't frequency-restricted. So at RF, it would be better to connect g2, g3, anode together than g3 to cathode.

Lack of a g3 dissipation rating didn't stop Ever Ready doing exactly this for the triodised DF97 in their Sky Monarch FM front-end...
kalee20 is offline  
Old 30th Aug 2022, 3:06 pm   #47
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Triode Connected?

I'd have assumed that the mindset of a valve manufacturer would be along the lines of:
"Why strap a pentode/tetrode as a triode, let's just flog 'em a triode in the first place!"

But maybe they didn't see the point of triode operation and didn't look beyond the greater gain and power of beam tetrodes? After all, power valves did shift to beam tetrodes and the power triode was left out in the cold. (RF jobs were still made, though)

I still keep suggesting the 3-500Z to audiophiles. With looks like that they ought to sound wonderful, but maybe I'm just wanting to thin their numbers a little with 2.5kV on a top cap...

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 30th Aug 2022, 4:36 pm   #48
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

Quote:
Originally Posted by Radio Wrangler View Post
I'd have assumed that the mindset of a valve manufacturer would be along the lines of:
"Why strap a pentode/tetrode as a triode, let's just flog 'em a triode in the first place!"
There is the flexibility offered by a triodisable BT or pentode though, which reduces the type numbers that have to be built - such as the aforementioned DF97 (triode-connected for FM front-end, pentode for IF amplification). Rather than a separate triode type in one place (yes I know the DC90 existed but AFAIK no receiver in Britain used it).

Quote:
Originally Posted by Radio Wrangler View Post
I still keep suggesting the 3-500Z to audiophiles. With looks like that they ought to sound wonderful, but maybe I'm just wanting to thin their numbers a little with 2.5kV on a top cap..
You're not trying David, for a medium-power (30W), affordable triode, how about an amplifier built around the PD500? We see 'L used for power pentode of BT, rarely 'D' used, as for a power triode, which itself creates instant appeal. Should be lots around (oh, and 25kV on the top cap anode...)
kalee20 is offline  
Old 30th Aug 2022, 4:43 pm   #49
ms660
Dekatron
 
ms660's Avatar
 
Join Date: Apr 2011
Location: Cornwall, UK.
Posts: 13,454
Default Re: Triode Connected?

You won't have to look far to see -ve bias on g3 in common use...Last IF (plus) ....

Lawrence.
ms660 is offline  
Old 30th Aug 2022, 5:08 pm   #50
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Triode Connected?

Quote:
Originally Posted by kalee20 View Post
for a medium-power (30W), affordable triode, how about an amplifier built around the PD500?
Did they ever make a rotary anode version of the PD500?

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 31st Aug 2022, 12:12 am   #51
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Triode Connected?

Regarding the suppressor grid (g3) disposition in triode-strapped pentodes, it would appear that in practice it did vary somewhat, as shown by some examples.

Firstly, the Dynatron T139 tuner/control unit (part of the Ether Marshal suite) had three triode-strapped pentodes, at least in one of its variations.

On the higher HF bands (including the bandspreads), there was a grounded grid, aperiodic RF preamplifier ahead of the main RF amplifier (a W77 pentode). In this case g3 was strapped to the anode.

The gramophone input stage was a triode strapped Z729 (EF86). Here g3 was grounded, rather than strapped to the anode or the cathode. (The stage had normal cathode bias.)

The tone control stage was a triode strapped Z77. As with the RF preamplifier, g3 was strapped to the anode.

In all three cases, g2 was strapped directly to the anode, with no intervening resistor. The latter may have been used more with AF power amplifiers. One might infer from the above that the disposition of g3 varied by valve type.


Secondly, the Eddystone 680X HF receiver used a triode strapped Z77 as its local oscillator. In this case, g3 was connected to the cathode, which was grounded; g2 was strapped directly to the anode.

The BFO was a triode-strapped 6BA6, g3 to ground, g2 direct to anode, with its cathode tapped into the oscillator coil.

This tends to nullify the notion that in the Z77, g3-to-anode was normal, and perhaps specified by the valve maker. It looks more as if it were designer choice.


Thirdly, the Marconi Atalanta marine receiver; this used a triode-strapped EF85 as 1st local oscillator with g3 to cathode, both grounded, and g2 direct to anode.


Fourthly, the RCA 6X8 triode-pentode was an interesting case. Its prominent application was as a VHF TV tuner frequency changer, amongst the first of its kind. But it was also intended for use as a frequency changer for both bands in AM-FM radio receivers. In the FM case, the mixer side could be used either as a pentode or a triode, according to designer preference. Either way, g3 was grounded. That was consistent with RCA’s position at the time that RF pentodes should have a separate pinout for g3, and that it should be grounded directly and not connected to the cathode. So in the triode-mode case, g3 was grounded and g2 was strapped directly to the anode.

One may guess at RCA’s reasons for what it did here. It needed a VHF TV triode pentode following the move to the “40 MHz” standard IF, where the closeness of the IF to TV channel 2 made the established double triode frequency changers problematical due to feedback. It also needed to match GE’s 12AT7 double triode which had been established inter alia as a single frequency changer in AM-FM receivers. RCA had originally addressed that requirement with the 6BE6 pentagrid, and when that did not work too well, with the 6SB7Y and 6BA7 improved pentagrids, which might not have been hugely better. Thus the 6J6 was pressed into that service, notwithstanding that its microphony was a less desirable feature for FM. With the 6X8 it could match the 12AT7 in AM-FM receivers where the designers wanted a double triode frequency changer, but it could also go one better in that for receivers with enough low-noise RF gain on FM, the mixer could be used in pentode mode. In “dual-mode”, the mixer section was switched between pentode mode on AM and triode mode on FM, with g3 grounded in both cases; g2 was accordingly switched between HT and strapped to the anode.


Looking through the valve datasheets:

The Z77 was specified for both pentode and triode operation, but for the latter case, the disposition of g3 was not mentioned.

The GEC Z729/EF86 was specified for both pentode and triode operation. In the triode case, g3 was shown with the same voltage as anode and g2, so strapped to the anode. On the other hand, Mullard had triode operation of the EF86 with g3 strapped to the cathode.

The EF85 was specified only for pentode operation, as was the 6BA6.


Cheers,
Synchrodyne is offline  
Old 31st Aug 2022, 7:42 pm   #52
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

That's a good variety thanks once again to Steve's encyclopaedic knowledge! Several disciplines - audio and HF, power and signal... great stuff!

Quote:
Originally Posted by joebog1 View Post
I do appreciate your explanations and even criticism's though kalee. It keeps me checking up on what I have said. Its in fact, the best reason to be a member here!! One never forgets they are learning, CONSTANTLY.
Absolutely - and it works both ways too, Joe! It's great to read the many thought-provoking posts here.
kalee20 is offline  
Old 31st Aug 2022, 7:57 pm   #53
turretslug
Dekatron
 
turretslug's Avatar
 
Join Date: Nov 2011
Location: Surrey, UK.
Posts: 4,385
Default Re: Triode Connected?

The Marconi CR100 used a triode-strapped KTW62 variable-mu pentode as local oscillator, with both g2 and g3 connected directly to anode.

The Eddystone 750 (conceived, I believe, as a compact, reduced-price sibling of the 680X) also featured the same triode-connected Z77 arrangement, g2 to anode and g3 to chassis.
turretslug is offline  
Old 1st Sep 2022, 6:03 am   #54
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Triode Connected?

Re the Dynatron case, also interesting is the LF613 power amplifier that was the companion to the T139 tuner/control unit.

This had a triode-strapped Z77 input stage, another triode-strapped Z77 phase splitter, followed by a push-pull triode-strapped KT66 pair. Another triode-strapped KT66 was used as the voltage stabilizer for the tuner/control unit HT power supply.

With each Z77, g2 and g3 were strapped to their respective anodes. All three KT66s had their screens connected to their respective anodes by 100R resistors.

Dynatron was clearly still very triode-oriented on the AF side at the time. The T139/LF613 were announced around 1954 September, but were essentially updates of the T99/LF612 series from around 1951 September. (It switched to pentodes with the TC10 and LF10 of c.1955 September.)

The use of triode-strapped KT66s was hardly unusual – if you wanted an indirectly heated triode output valve at the time, then it was more-or-less a routine choice.

But given that all of the Z77 complement in the T139/LF613 combination were triode-strapped, the question comes to mind – why not use a triode, such as the L77 (6C4), which I think was GEC’s most prominent small-signal triode of the time. It might have been a matter of characteristics. The L77 had a µ of 17 and a slope of 2.2 mA/V. The numbers for the triode-strapped Z77 were 75 and 7.5 mA/V. A single triode with similar characteristics might have been hard to find. Also, pricing on the relatively fast-moving and very widely used Z77 might have been better.

At the RF end, some validation for the selective use of the triode-strapped pentode came from Wallman’s original cascode circuit, which for best performance had a triode strapped 6AK5 pentode input stage, and a 6J4 triode second stage. Dynatron’s use of a triode-strapped Z77 as a grounded grid RF preamplifier was probably predicated upon the fact that the “standard” remote cutoff RF pentodes of the time, W77, EF41, EF89, etc., were a bit noisier than desired at upper HF. The usual solution in comparable receivers was to use instead a high slope pentode, Z77, EF80, 6F1, etc., as was done by some other makers (e.g. Ambassador, Armstrong, Murphy, although Murphy also required such for its own image-rejection circuit.) But that complicated things from the AGC viewpoint. Given that with the T139, Dynatron also seemed to be aiming for an extraordinarily high valve count anyway, adding the grounded-grid RF preamplifier was aligned with that. Possibly too there was an RF selectivity angle. With the T139, Dynatron was stepping down to a three-gang front end from the four-gangs used in the T99 series, and so losing some front end selectivity. (This may have been an enforced change basis what was available in the way of third-party bandspread front ends at the time.) Having the grounded grid gain stage between the aerial input and the first RF tuned circuits (on bandspreads and upper HF) might have allowed those circuits to be tightened up somewhat as compared with the relative looseness required when they were looking back into the variable impedance of a random aerial. But that is speculation on my part.

(As an aside, getting sufficiently low noise at upper HF was solved more neatly by the use of cascode RF amplifiers, often configured to be semi-remote cutoff, which crept into HF receiver practice in the late 1950s, using second-generation cascode valves. One could say that these added more “quiet” than actually needed, and in some cases the extra benefit was taken up by using double bandpass input tuning – e.g. Eddystone 830 and later Racal RA17.)

Anyway, at least from post facto back-of-the-envelope rationalizations, it does appear that there were sound reasons for the triode-strapping of pentodes and tetrodes by the equipment makers, and thus for the valve makers to sometimes include the “strapped” characteristics in their datasheets. And there were some cases where “strappability” appears to have been an initial design objective.


Cheers,
Synchrodyne is offline  
Old 1st Sep 2022, 8:04 am   #55
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Triode Connected?

Quote:
Originally Posted by Synchrodyne View Post
But given that all of the Z77 complement in the T139/LF613 combination were triode-strapped, the question comes to mind – why not use a triode, such as the L77 (6C4), which I think was GEC’s most prominent small-signal triode of the time. It might have been a matter of characteristics. The L77 had a µ of 17 and a slope of 2.2 mA/V. The numbers for the triode-strapped Z77 were 75 and 7.5 mA/V. A single triode with similar characteristics might have been hard to find. Also, pricing on the relatively fast-moving and very widely used Z77 might have been better,
The Eddystone EA12 uses a couple of 6C4s as its oscillators.

These feed the triode sections of ECH81s as buffers and thence to the mixer sections.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 1st Sep 2022, 8:10 am   #56
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

That's fascinating as always.

Several manufacturers used Z77 as a grounded-cathode pentode for FM RF amplifier, but I'd not heard of the same valve being used as grounded-grid, triode-strapped, doing the same job.

It's right on topic for this thread anyway!

I'm wondering if transit time between the strapped grids had any effect at VHF?

What we still don't know is the reasons behind the decisions to sometimes take suppressor to cathode and sometimes to the 'anode' conglomerate. I'm beginning to suspect there wasn't one - I find in inconceivable that nobody on this Forum would know the definitive answer if there was one.

Quote:
Originally Posted by Synchrodyne View Post
And there were some cases where “strappability” appears to have been an initial design objective.
Such as, maybe, g2, g3, anode pins right next to each other?
kalee20 is offline  
Old 1st Sep 2022, 8:53 am   #57
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: Triode Connected?

Pin assignment choices were probably influenced by the need to get bent wires away from each other as they went from electrode to pin. They definitely were influenced by the desire to keep sensitive inputs away from large signals and pins with large gain from them (instability risks)

Trying to collect things which might get strapped in a few applications might not find much choice left.

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Old 1st Sep 2022, 9:28 am   #58
kalee20
Dekatron
 
Join Date: Feb 2007
Location: Lynton, N. Devon, UK.
Posts: 7,061
Default Re: Triode Connected?

True. Though the Z77 / EF91 managed it!
kalee20 is offline  
Old 2nd Sep 2022, 6:36 am   #59
Synchrodyne
Nonode
 
Join Date: Jan 2009
Location: Papamoa Beach, Bay of Plenty, New Zealand
Posts: 2,943
Default Re: Triode Connected?

Quote:
Originally Posted by kalee20 View Post
What we still don't know is the reasons behind the decisions to sometimes take suppressor to cathode and sometimes to the 'anode' conglomerate. I'm beginning to suspect there wasn't one - I find in inconceivable that nobody on this Forum would know the definitive answer if there was one.
Likewise I am now doubting that there is a rigorous explanation. It could well have been left to the whim of the circuit designer. With the Z77, the adjacency of the anode, g3 and g2 pins may have encouraged the strapping of g3 to the anode. As you say, there is a good chance that this could have been the intent of the valve designer.

Re the Dynatron T139, I should have mentioned that it did include an L77/6C4 triode, used as the oscillator, with the X79 then used simply as a mixer, with its own triode inactive (anode strapped to cathode). I imagine that was done to reduce oscillator pulling and to ensure that the oscillator was robust right up to 30 MHz over a reasonable lifetime. One may infer that Dynatron saw the L77 as definitely better than a triode-strapped Z77 for the oscillator job, but that the obverse applied to the grounded grid RF preamplifier stage.


Cheers,
Synchrodyne is offline  
Old 2nd Sep 2022, 1:17 pm   #60
Robert Gribnau
Heptode
 
Robert Gribnau's Avatar
 
Join Date: May 2017
Location: Konongo, Ghana
Posts: 510
Default Re: Triode Connected?

Quote:
Originally Posted by joebog1 View Post
Even driving EIGHT output valves from if I remember correctly a 12AX7, which once again from memory is something like 68K plate resistance. I guess the cathode will be lots lower.

I cant find the book right now, but I do remember the circuits ( to some degree ). I placed them in the "audiophool " category LONG before I joined this forum, perhaps before the forum existed.
I think it would be fair to Menno van der Veen if you would find that particular book/design to prove your statements. It would really surprise me a lot if he did use a 12AX7 to drive an output stage of an amplifier intended for audio. I only know examples where he uses ECC81/12AT7, ECC82/12AU7, or ECC88/6DJ8 for that purpose. In some of his explanations he even describes that he chose one of these valve types because of their low output impedance.

Until now I only saw him use the ECC83/12AX7 in guitar amplifiers. And I never saw a design by him with 8 power tubes per channel.

To illustrate this a bit a link to one of his first designs (from 1987). In it he uses an ECC82 to drive 4 x EL34 push-pull.

https://mennovanderveen.nl/images/on...h_new_OPTs.pdf
__________________
Robert
Robert Gribnau is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:22 am.


All information and advice on this forum is subject to the WARNING AND DISCLAIMER located at https://www.vintage-radio.net/rules.html.
Failure to heed this warning may result in death or serious injury to yourself and/or others.


Powered by vBulletin®
Copyright ©2000 - 2024, vBulletin Solutions, Inc.
Copyright ©2002 - 2023, Paul Stenning.