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 > Components and Circuits

Notices

Components and Circuits For discussions about component types, alternatives and availability, circuit configurations and modifications etc. Discussions here should be of a general nature and not about specific sets.

Closed Thread
 
Thread Tools
Old 27th Oct 2020, 10:54 am   #1
vinrads
Dekatron
 
Join Date: Feb 2009
Location: Chesterfield, Derbyshire, UK.
Posts: 3,737
Default EF86 as a phase splitter.

I am building a stereo amp ,from two Ekco units using EL 84's, I have just noticed it use an EF86 as a phase splitter , question! is this a good use for this valve ,does it have any benefit over the usual ECC valve. Mick.
Attached Thumbnails
Click image for larger version

Name:	IMGP0042.jpg
Views:	211
Size:	85.5 KB
ID:	218997  
vinrads is offline  
Old 27th Oct 2020, 11:12 am   #2
barrymagrec
Octode
 
Join Date: Dec 2017
Location: Morden, Surrey, UK.
Posts: 1,552
Default Re: EF86 as a phase splitter.

The EF86 is strapped as a triode. It`s a very basic circuit and not nearly as good as a balanced circuit using a double triode, however probably good enough with high gain otput valves such as the EL84.
barrymagrec is offline  
Old 27th Oct 2020, 11:28 am   #3
vinrads
Dekatron
 
Join Date: Feb 2009
Location: Chesterfield, Derbyshire, UK.
Posts: 3,737
Default Re: EF86 as a phase splitter.

Thanks Barry I was going to re build as original, but as I have more ECC'S than EF86'S I think a slight mod is req. Mick.
vinrads is offline  
Old 27th Oct 2020, 11:36 am   #4
barrymagrec
Octode
 
Join Date: Dec 2017
Location: Morden, Surrey, UK.
Posts: 1,552
Default Re: EF86 as a phase splitter.

You`l have a lot more gain with a balanced circuit, which you may not want. If I remember correctly a triode strapped EF86 has pretty similar characteristics to half an ECC82.
barrymagrec is offline  
Old 27th Oct 2020, 12:05 pm   #5
snowman_al
Octode
 
snowman_al's Avatar
 
Join Date: Dec 2018
Location: Llandeilo, West Wales, UK.
Posts: 1,092
Default Re: EF86 as a phase splitter.

It's a standard 'concertina' phase splitter. (Also called cathodyne or split load.)
It has advantages and disadvantages just like the cathode coupled phase splitter using 2 triodes. It has just less than unity gain, is well balanced, is very simple and effective. Why change it?

It was (is) used on classic amps, like the Williamson, Audio Innovations and many more.

Plenty of info if you search 'cathodyne phase splitter'. One example here: http://www.valvewizard.co.uk/cathodyne.html
__________________
Never Leave Well Enough Alone...
snowman_al is offline  
Old 27th Oct 2020, 12:21 pm   #6
barrymagrec
Octode
 
Join Date: Dec 2017
Location: Morden, Surrey, UK.
Posts: 1,552
Default Re: EF86 as a phase splitter.

It was used on the Williamson amplifier but in conjunction with a following amplifier stage to increase the drive.

I wouldn`t change the design here though although I might use a different valve if they were more available. Never liked the EF86 much anyway....
barrymagrec is offline  
Old 27th Oct 2020, 12:29 pm   #7
daviddeakin
Hexode
 
daviddeakin's Avatar
 
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
Default Re: EF86 as a phase splitter.

Quote:
Originally Posted by vinrads View Post
I am building a stereo amp ,from two Ekco units using EL 84's, I have just noticed it use an EF86 as a phase splitter , question! is this a good use for this valve ,does it have any benefit over the usual ECC valve. Mick.
No benefit to the EF86 (except maybe less hum in the old days), you can use a typical triode and you won't notice any difference. You can chuck anything at a concertina splitter and it will always perform beautifully, best splitter by far!
daviddeakin is offline  
Old 27th Oct 2020, 12:39 pm   #8
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,797
Default Re: EF86 as a phase splitter.

It has to be triode strapped in that circuit, or else the screen current would cause problems with the balance.

I rate the 'concertina' phase splitter as OK-ish Despite the equal resistors in anode and cathode, the output impedances of this splitter are very badly matched. One path is 1/Gm in parallel with the resistor, the other path is Ra in parallel with the resistor.

It would be esy to say that any valve where Ra isn't bigger than 1/Gm is going to be lousy for gain, in this case it's the triode strapped Ra which is lower than pentode mode. Gainwise, it doesn't matter too much here, but impedance wise, it matters.

The long tail pair double triode is a step up in ability. It gives some actual gain which reduces the demands on the input stage. It's outputs are more symmetrical but not perfect. there is asymmetric feedback from one anode to the driving grid while the other anode has capacitance to a decoupled grid.

The best splitter is the Bailey/Radford pentode-triode long tail pair. The pentode by way of its screen and suppressor grids constitutes a cascode structure breaking up the capacitive feedback path on the input valve, while the triode section with its decoupled grid constitutes a folded cascode to the other output. This improves the output balance a lot. and pushes out a pole beyond the frequency range of the loop. If I was building something valved, I wouldn't use anything else.

Cascoding in transistor amplifiers is even more valuable as it not only staves off Ccb issues it also staves off varactor non-linearity and Early effect.

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 27th Oct 2020, 12:47 pm   #9
daviddeakin
Hexode
 
daviddeakin's Avatar
 
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
Default Re: EF86 as a phase splitter.

Quote:
Originally Posted by Radio Wrangler View Post
Despite the equal resistors in anode and cathode, the output impedances of this splitter are very badly matched.
That's a common misconception; the output impedances are identical, roughly 2/gm. Mismatch is dependent only on the matching of Ra and Rk, which can be as close tolerance as you care for.

Quote:
The long tail pair double triode is a step up in ability. It gives some actual gain which reduces the demands on the input stage.
But it needs two triodes, has worse balance (dependant on triode matching as well as external components), lower input impedance, larger output impedance... If you used the same extra triode for gain before the concertina, you get more total gain from the same parts! True the LTP might be a bit more tolerant of grid current in the power valves, but if you're not thrashing the output stage then the concertina has to be the splitter of choice.

The Bailey/Radfor is a rod for your own back, IMO. It buys some better bandwidth performance (compared to regular LTP, but still worse than the concertaina) for the price much worse sensitivity to component choice and ageing.

Last edited by daviddeakin; 27th Oct 2020 at 12:52 pm.
daviddeakin is offline  
Old 27th Oct 2020, 1:31 pm   #10
jjl
Octode
 
jjl's Avatar
 
Join Date: Jan 2003
Location: Ware, Herts. UK.
Posts: 1,082
Default Re: EF86 as a phase splitter.

Morgan Jones refers to unequal output impedances in concertina phase splitters in the first edition of his "Valve Amplifiers" book. He suggests use of a "build-out" resistor in series with the output of the cathode connection of the phase splitter to counter this.
By the third edition of the book I can find no mention of build-out resistors.

John

Last edited by jjl; 27th Oct 2020 at 1:46 pm.
jjl is offline  
Old 27th Oct 2020, 1:56 pm   #11
daviddeakin
Hexode
 
daviddeakin's Avatar
 
Join Date: Apr 2007
Location: W Yorks, UK.
Posts: 406
Default Re: EF86 as a phase splitter.

Quote:
Originally Posted by jjl View Post
Morgan Jones refers to unequal output impedances in concertina phase splitters in the first edition of his "Valve Amplifiers" book. He suggests use of a "build-out" resistor in series with the output of the cathode connection of the phase splitter to counter this.
By the third edition of the book I can find no mention of build-out resistors.
In the fourth edition on page 462 he does the derivations and ends with "Because Zk = Za, the frequency response at each output is forced to be the same, so the output resistances must also be equal, and rout(k) = rout(a)."
daviddeakin is offline  
Old 27th Oct 2020, 6:50 pm   #12
G6Tanuki
Dekatron
 
G6Tanuki's Avatar
 
Join Date: Apr 2012
Location: Wiltshire, UK.
Posts: 13,951
Default Re: EF86 as a phase splitter.

That's a perfectly-satisfactory split-load phase-splitter which I've used in the past [even using a 6AT6 double-diode-triode because it was all-I-had-available]. it works OK, just as the classic 'see-saw' double-triode-with-unbypassed-common-cathode-resistor does.

Both are, IMHO, better than the double-triode phase-splitter that uses a resistive potential-divider across the output of one triode which feeds the grid of the other triode as a 'phase inverter'. This makes horrible assumptions about the long-term gain-stability of the phase-inverter triode and the resistive-stability of the potential-divider resistors. We all know how 50-year-old carbon-stick resistors can gain/lose tens of percent of their initial resistance over the years.

My favourite phase-splitter is the Jeffrey Bootstrap circuit, which gives good gain and balance at the same time [see below]. - the 'trick' here is that the triode bootstraps the anode-resistor of the pentode, so there's essentially the same audio-voltage appearing at either end of this resistor - meaning the pentode is 'looking' into a much higher (and more-consistent) impedance and so can generate a greater voltage-swing.

Of course once you move away from simplistic Class-A modes and your output-pair start drawing significant grid-current [like all honest hard-working output-valves should] you need something more than just phase-splitters.
Attached Thumbnails
Click image for larger version

Name:	jeffrey.jpg
Views:	166
Size:	59.0 KB
ID:	219047  
G6Tanuki is offline  
Old 27th Oct 2020, 7:07 pm   #13
Trevor
Octode
 
Join Date: Aug 2006
Location: Stoke-on-Trent, Staffordshire, UK.
Posts: 1,082
Default Re: EF86 as a phase splitter.

I remember this circuit presented by gilbert briggs the why and how of good audio I think
The same book featured the Garner circuits
Very interesting circuit where medium dive voltages to the output valves is considered to be ok
Very nice revival
Trev
Trevor is offline  
Old 28th Oct 2020, 1:18 am   #14
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,797
Default Re: EF86 as a phase splitter.

There is an analysis which says the same current must flow in both the cathode and the anode of a concertina splitter, and that therefore as long as the load resistors are equal then the two outputs should match.

However loading on one output affects the grid/cathode voltage and hence affects the valve current, while loading on the other output (anode) does not. At least doesn't affect it in exactly the same ratio. The valve downstream on the cathode because of cathode follower action thinks it sees a different source impedance to what the valve driven from the anode does.

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 28th Oct 2020, 1:58 am   #15
martin.m
Hexode
 
martin.m's Avatar
 
Join Date: Mar 2018
Location: Bishop Auckland, County Durham, UK.
Posts: 373
Default Re: EF86 as a phase splitter.

Quote:
Originally Posted by jjl View Post
Morgan Jones refers to unequal output impedances in concertina phase splitters in the first edition of his "Valve Amplifiers" book. He suggests use of a "build-out" resistor in series with the output of the cathode connection of the phase splitter to counter this.

John
Is this why the Pye Black Box amp has a 100K resistor in series with the cathode output of it's concertina phase splitter? R15 on the version that uses EL42 output valves.
__________________
Regards
Martin
martin.m is offline  
Old 28th Oct 2020, 8:35 am   #16
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,631
Default Re: EF86 as a phase splitter.

If you use one ECC82,one triode for the IP/voltage gain, one triode as cathodyne PS you should have just about enough gain to drive two EL84's which need about 10v RMS + a bit more for NFB,it depends what you want, if it's a simple circuit with no frills this does the job Mick . Better if you can direct couple the IP to PS, no cap to muck up frequency response.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 2nd Nov 2020, 3:02 pm   #17
AC/HL
Dekatron
 
AC/HL's Avatar
 
Join Date: Nov 2003
Location: Heckmondwike, West Yorkshire, UK.
Posts: 9,637
Default Re: EF86 as a phase splitter.

Continued here: https://www.vintage-radio.net/forum/...d.php?t=172700
AC/HL is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 9:00 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.