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 > Other Discussions > Homebrew Equipment

Notices

Homebrew Equipment A place to show, design and discuss the weird and wonderful electronic creations from the hands of individual members.

Closed Thread
 
Thread Tools
Old 8th May 2020, 11:27 pm   #1
Steve_Bell
Hexode
 
Join Date: Sep 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 472
Default KEL80 EL34 Monoblock 80w amplifier

I'm thinking about building a couple of these 80w amplifiers. The design uses 4 x EL34 in parallel push-pull. It's a World Audio Design (WAD), KEL80, dating from 1999. The transformers are no longer available, so can anyone work out the A-A load of the output transformer from the circuit diagram excerpt?

HT is 510v, G2 supply voltage is 250v. Bias is a combination of cathode resistors giving cathode voltage of 9v and a fixed bias supply of -10v to the EL34 G1.

thanks,
Steve.
Attached Thumbnails
Click image for larger version

Name:	DSCF0826.jpg
Views:	223
Size:	63.1 KB
ID:	205305  
Steve_Bell is online now  
Old 9th May 2020, 6:53 am   #2
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
Default Re: KEL80 EL34 Monoblock 80w amplifier

Your first stop is the EL34 datasheet then you need to draw some load lines, first off your HT is 510v & 250v g2, which looks a bit low, wonder why they chose that value? (see below) These voltages will affect the result, but a quick cheat is to check out Hammond's OPT page here - http://hammondmfg.com/1608.htm . As you see 3k4, 4k3 & 5k are common Zpri. Your OPT choice dictates every other aspect of the amplifier circuit design, so unless you can find the exact OPT they used in the protoype, you will have to alter and tweak the amp design, even with an OPT of exactly the same Zpri.

I had a quick look at a EL34 datasheet, there is no example there that fits your HT and g2 voltages. Looking at the schematic I can see a few things that look a bit different to most EL34 PPP OP stages I've seen, firstly the afore mentioned 250v g2, second Rk = 180r, meaning the valves Vk will be high, so Ik will be high too, however you have -10v biased applied & having the screen grids at 250v balances this, so it's halfway between fixed and auto biased, which is interesting.

BTW, is this a guitar amp? Looks like it, checking out some similar amps might be an idea.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 9th May 2020, 9:38 am   #3
Steve_Bell
Hexode
 
Join Date: Sep 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 472
Default Re: KEL80 EL34 Monoblock 80w amplifier

It's not a guitar amplifier. WAD also had a 40W design (KEL34) that used just one pair of EL34s in an identical design, with the the low G2 volts and the same bias arrangement.
Steve_Bell is online now  
Old 10th May 2020, 4:23 pm   #4
Aub
Nonode
 
Aub's Avatar
 
Join Date: Mar 2005
Location: Nuneaton, Warwickshire, UK.
Posts: 2,034
Default Re: KEL80 EL34 Monoblock 80w amplifier

Hi Steve,

A lot of designs, for 4 x EL34 use 1.75k or 1.9k anode to anode. As Andy says, other factors will influence the value. The mixture of fixed and auto bias is unusual.

Cheers

Aub
__________________
Life's a long song, but the tune ends too soon for us all.
Aub is offline  
Old 11th May 2020, 7:00 am   #5
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
Default Re: KEL80 EL34 Monoblock 80w amplifier

Not a guitar amp... "The mixture of fixed and auto bias is unusual." It's something I've been meaning to have a look at for some time, it means matched valves aren't essential and you get more OP power. Apart from drawing load lines this source (100 amplifiers) might have something similar with more info - http://lilienthalengineering.com/ also here has some useful info on drawing loadlines - https://wtfamps.com/load-line-calculations/ as well as - http://www.valvewizard.co.uk/pp.html

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 11th May 2020, 8:00 am   #6
Aub
Nonode
 
Aub's Avatar
 
Join Date: Mar 2005
Location: Nuneaton, Warwickshire, UK.
Posts: 2,034
Default Re: KEL80 EL34 Monoblock 80w amplifier

I guess one benefit of having the two biasing systems is, that if the fixed bias supply fails, there is still the auto bias, so no blown output valves?

Aub
__________________
Life's a long song, but the tune ends too soon for us all.
Aub is offline  
Old 11th May 2020, 9:32 am   #7
AdrianH
Octode
 
AdrianH's Avatar
 
Join Date: Mar 2020
Location: Blackburn with Darwen, Lancashire, UK.
Posts: 1,567
Default Re: KEL80 EL34 Monoblock 80w amplifier

Just a simple question on the circuit. Should there not be decoupling caps to ground on g2 of the valves, or, because the resistors are relatively low at 100 Ohm it is not thought required as the 250 Volt would have adequate smoothing?

Adrian
AdrianH is online now  
Old 11th May 2020, 10:42 am   #8
Aub
Nonode
 
Aub's Avatar
 
Join Date: Mar 2005
Location: Nuneaton, Warwickshire, UK.
Posts: 2,034
Default Re: KEL80 EL34 Monoblock 80w amplifier

Hi again,
I've just remembered that VOX made an early version of their AC100 amp, which was stated as 80 watts. It used 4 x EL34 with cathode bias. The later version used fixed bias and was stated as 100 watts. There is a website dedicated to the Vox AC100. A search should come up with the URL. There's loads of stuff on there.

Cheers

Aub
__________________
Life's a long song, but the tune ends too soon for us all.
Aub is offline  
Old 11th May 2020, 10:59 am   #9
Steve_Bell
Hexode
 
Join Date: Sep 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 472
Default Re: KEL80 EL34 Monoblock 80w amplifier

Thanks all for the links and pointers. Plenty to look at.
Steve_Bell is online now  
Old 11th May 2020, 1:29 pm   #10
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: KEL80 EL34 Monoblock 80w amplifier

Quote:
Originally Posted by AdrianH View Post
... because the resistors are relatively low at 100 Ohm it is not thought required as the 250 Volt would have adequate smoothing?

That's the explanation. The 100 ohm resistors are so-called 'stoppers' for the screen grids. They're there, among other things, to prevent unwanted RF oscillation arising from grid capacitances resonating with reactances in the external wiring.

Cheers,

GJ
__________________
http://www.ampregen.com
GrimJosef is offline  
Old 11th May 2020, 2:12 pm   #11
AdrianH
Octode
 
AdrianH's Avatar
 
Join Date: Mar 2020
Location: Blackburn with Darwen, Lancashire, UK.
Posts: 1,567
Default Re: KEL80 EL34 Monoblock 80w amplifier

OK thank you, it was a query after reading a section in Scroggie's book.

Adrian
__________________
Asking questions and learning, or trying to!
Youtube EF91Valve
AdrianH is online now  
Old 12th May 2020, 5:38 pm   #12
bikerhifinut
Octode
 
Join Date: Nov 2013
Location: Penrith, Cumbria, UK
Posts: 1,993
Default Re: KEL80 EL34 Monoblock 80w amplifier

Try this forum.

Somebody there will definitely have experience. And info on transformer ratios.

http://www.world-designs.co.uk/forum/index.php

Andy.
bikerhifinut is offline  
Old 12th May 2020, 5:46 pm   #13
bikerhifinut
Octode
 
Join Date: Nov 2013
Location: Penrith, Cumbria, UK
Posts: 1,993
Default Re: KEL80 EL34 Monoblock 80w amplifier

Ah I see you already have been there...………...
bikerhifinut is offline  
Old 12th May 2020, 9:44 pm   #14
Steve_Bell
Hexode
 
Join Date: Sep 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 472
Default Re: KEL80 EL34 Monoblock 80w amplifier

Well, thanks to the World Designs forum and those who directed me there via PMs, etc., the answer is 4.6K. I actually have the HiFi World DIY supplement of April 2001 that gives the specs - just a case of knowing where to look.

Incidentally, this current thread "KEL80 EL34 Monoblock 80w amplifier" has already been indexed by Google.

Thanks again everyone for your inputs.

Steve.
Steve_Bell is online now  
Old 13th May 2020, 5:41 am   #15
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
Default Re: KEL80 EL34 Monoblock 80w amplifier

4K6, that might be tricky to source, a 4K3 or 5K would probably do the job though, it's only 300r ish difference, if you get one with 4,8,16 secs it gives you the chance to use a 4r with an 8r load say, not optimal but custom OPT's arn't cheap. These lads in Poland do bespoke toroid OPT's and their OPT's are competively priced - https://sklep.toroidy.pl/en_US/index BTW, why did you want to go for this particular design?

Not come across that forum, looks interesting. Glad you found your answer, Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 13th May 2020, 8:32 am   #16
GrimJosef
Dekatron
 
Join Date: Sep 2007
Location: Oxfordshire, UK.
Posts: 4,310
Default Re: KEL80 EL34 Monoblock 80w amplifier

It's always worth remembering that the primary impedance is a reflection of the load on the secondary. We make fine judgements between 4k3 and 4k6 and 5k, +/-5-10% or so, assuming an 8ohm load on the secondary when in fact the impedance of an '8ohm' speaker varies much more than that with frequency. So in real life the valves will see a widely varying load.

Cheers,

GJ
__________________
http://www.ampregen.com

Last edited by GrimJosef; 13th May 2020 at 8:42 am.
GrimJosef is offline  
Old 13th May 2020, 10:47 am   #17
Aub
Nonode
 
Aub's Avatar
 
Join Date: Mar 2005
Location: Nuneaton, Warwickshire, UK.
Posts: 2,034
Default Re: KEL80 EL34 Monoblock 80w amplifier

Gave up trying to register with the World-Designs forum. Filled in all the fields several times. When pressing Complete Registration button, it empties some of the fields and doesn't say what's wrong. Repeat process - give up. 🤨
__________________
Life's a long song, but the tune ends too soon for us all.
Aub is offline  
Old 13th May 2020, 7:35 pm   #18
Steve_Bell
Hexode
 
Join Date: Sep 2004
Location: Newcastle upon Tyne, Tyne & Wear, UK.
Posts: 472
Default Re: KEL80 EL34 Monoblock 80w amplifier

In reply to Andy: Serendipity.

I was sorting through my large collection of hifi mags, with a view to thinning out the collection (not going to happen!). I came across a large collection of the HFW DIY supplements, one of which contained the KEL80 article. I had built a KEL84 in the past and liked that.

Also, I have the power transformers and a couple of sturdy chassis and the valves.
Steve_Bell is online now  
Old 14th May 2020, 5:38 am   #19
Diabolical Artificer
Dekatron
 
Diabolical Artificer's Avatar
 
Join Date: Jul 2009
Location: Sleaford, Lincs. UK.
Posts: 7,637
Default Re: KEL80 EL34 Monoblock 80w amplifier

"It's always worth remembering that the primary impedance is a reflection of the load on the secondary" It's easy to forget that sometimes when looking for OPT's, the 8r of a speaker being it's nominal impedance, a bit like pinning the tail on a donkey.

"Serendipity." Oh, her. With me it's usually the other way round , have OPT X and then find a design that fits. AFAIK a triode OP stage is the most tolerant of load differences, UL next, pentode being the most choosey . Class has something to do with it also I think.

Andy.
__________________
Curiosity hasn't killed this cat...so far.
Diabolical Artificer is offline  
Old 14th May 2020, 7:48 am   #20
Radio Wrangler
Moderator
 
Radio Wrangler's Avatar
 
Join Date: Mar 2012
Location: Fife, Scotland, UK.
Posts: 22,800
Default Re: KEL80 EL34 Monoblock 80w amplifier

Quote:
Originally Posted by Diabolical Artificer View Post
AFAIK a triode OP stage is the most tolerant of load differences, UL next, pentode being the most choosey . Class has something to do with it also I think.
Draw up sets of load curves for a few power triodes and a beam-tetrodes and draw on some load lines, then you'll start to see the effect the anode impedance of the valve (the gradient of the curves) has. Ultra-linear operation moves you from the beam tetrode condition towards the triode condition. What tap percentage you go for sets the positioning between these extremes.

The valve's own anode impedance (not to be confused with the load impedance presented by the output transformer) is a lot lower for triodes than for beam tetrodes. This has several effects; damping the Q of any resonances, and reducing the output Z presented to the speakers as well as dictating different optimum-fit load lines.

Look at loudspeaker impedance/frequency curves, and you'll see the magnitude of the point G-J was making. it is not unusual for high quality speakers to exhibit dips to below half their nominal impedance value (below a quarter in one speaker that comes to mind) and can climb or peak to several times their nominal impedance.

This makes a mockery of any pretence of precision in choosing a load line, instead you're trying to find a compromise which is not much worse at one extreme than the other. On the whole, valve amplifiers are more problematic with too-high impedance loads than with too-low. So the choice is best slewed to suit that.

At about this point you should see that there is a lot of influence of the loudspeaker on amplifier

It gets worse.

Speakers don't look purely resistive, and so actual load lines cease to be lines and become ellipses. So you now have to do the sums, plot the ellipses for what the poor valves might be the victims of, and be prepared to change the turns ratio of the output transformer a bit to best keep things within the valve's operating area, considering HT supply voltage etc.

But plotting a straight-forward load line on a set of valve characteristics is the right starting point. Then you hit it with the complications. You won't be able to handle everything so you have to choose your compromise by looking at what you won't be able to handle.

with beam tetrodes and power pentodes, (the terms were bandied around by firms dancing around each other's patents) the things are a lot more similar than the different names suggest, electron flow is formed into sheet-shaped beams, so we have to consider the electron optics of the beams. Electrons naturally repel each other and beams tend to de-focus. High perveance tetrodes use tight focusing to make the beams sharper and improve the efficiency of anode current versus screen current. This is typical of later valve designs. The EL34 is often said to be a 'harder' valve than the 6L6/807 and they weren't talking about the vacuum. Maybe an extreme example is the 4CX250B RF power tetrode. It's a bit of a bad lad with some rather bad habits. Used as a linear amplifier it shows dramatic reduction in linearity if the screen supply presented to it is high impedance. For linear behaviour, you really need a well regulated screen supply. Then the worst nasty of high perveance tubes kicks in, driven hard, they can give negative screen current, so that screen regulator has to be able to dump current to ground... this means shunt regulator rather than series. To make the screen supply low Z at the RF frequencies these little beasties operated at, you needed special valve holders with built-in ceramic screen decoupling capacitors, and the screen was brought out as a metal ring all the way round the skirt of the valve.

By comparison audio beam tetrodes are cuddly, furry, and make good pets.

But their extreme cousin illustrates some of the measures which still make the sane end of the family behave better. Low-Z screen supplies, screen regulation. And the ability to take a bit of reversed screen current if overdriven or the speaker Z goes high. The audio valves are used with load variations that would cause drama with the thin-skinned RF nutjobs.

The audio valves don't have the ultra-low inductance screen connections of the rF tetrodes, so the option of such intimate decoupling isn't avilable, but you still don't want them oscillating at RF, so you have to go the stopper resistor route. You take a hit in linearity, so you want to keep stopper values low.

So now let's nip over to the world of loudspeakers....

THe speakers screw up amplifier design by ruining all their nice load lines, so it's time for the amplifiers to get their own back. Interaction is a two-way street.

The basic principle of a moving coil speaker is that the coil resistance is low and that it is driven by a low impedance electrical cource. This seems like a recipe for out of control large currents. It is. But then the cone starts to move, and by moving in the stationary field of the permanent magnet (or wound field) the coil generates back EMF, opposing the voltage from the amplifier. In the idealised case of zero copper resistance and a true voltage source drive, the cone accelerates to the velocity where it fully opposes the drive voltage. Cone velocity creates airflow, and that makes pressure against the acoustic load impedance. So speakers try to look like voltage to pressure converters. This acts to flatten their frequency response.

But the coil resistance isn't zero at DC, which spoils the ideal a bit, but we really don't want a high output impedance amplifier to make it much worse.

But the speaker designer makes it worse. He can't get the full frequency range as flat as he'd like with one driver, so along come multiple drivers and crossovers. These are diplexing (triplexing...) filters and they are strongly influenced by the source impedance presented by the amplifier. The designer has to make some assumptions and some downright guesses and come up with the best compromise he can. But loudspeakers want low drive impedances.

Back to the amplifier. Pentodes and beam tetrode anode curves are a lot flatter than triodes. This means that the impedances the anodes present to the output transformer are a lot higher with the beam tets than with the triodes. The output transformer faithfully scales this and presents it to the loudspeaker.

So the designer uses feedback to calm things down. Negative feedback not only flattens gain, reduces noise and mends split ends, it also reduces output impedance and this reduces the speaker's dependence on the amplifier.

Here is the root cause of why some people, by using their ears, have decided they don't want any feedback, it opens the game up to a lot more interdependence of amplifier and speaker, and they can spend the years of their lives looking for better combinations that least upset each other.

Valves hav high-ish to high anode impedances, so they really need output transformers... they keep cables at HT away from toddlers as well. Unfortunately they exhibit growing phase shift at increasing frequency, and the feedback loop has to kept stable until the loop has run out of gain to be unstable with. This puts a strong limit on how much feedback can be employed, and that limits how much the output impedance presented to the speaker can be reduced by. Here we go round the mulberry bush.....

One last thought, Quad, when they designed the QuadII chose an unusual output topology, and had to pay for it with a more complex output transformer design. They must have been thinking of their loudspeaker developments with their famously reactive load impedances and elliptical load lines. They certainly can be seen reflected in some design choices in the 303

David
__________________
Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done
Radio Wrangler is offline  
Closed Thread

Thread Tools



All times are GMT +1. The time now is 12:37 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.