What i really want is the lowest distortion EL84 amp possible. Lundahl seems the most likely candidate. I also want UL taps at 25% as the Dynaco Z565. I cannot find a match on the Lundahl web site. Has anyone here every seen anything like that from Lundahl? Otherwise I will probably start with a Triode Z565-48.
-david BTW
-david BTW
Transcendar also makes a 565 repro (Link) and Magnequest did as well, though not sure if you can still get them from MQ.
Yup I have found a few 565 repos out there, some with 4 and 8 ohm secondaries and comments I have found are very good. But my personal experience with transformers is that they have a HUGE effect on the sound quality of a tube amplifier. So I am trying to find the best I can afford, not the cheapest.
Many years ago I used Plitron for power transformers in some commercial products I designed. I have not considered them, but I will. Thanks for the suggestion. I have looked at Sowter as well. But am still gathering info.
One important point, is that early dynaco research shows that the EL84 tubes prefer a 25% ultra linear tap, and many of the off-the-shelf transformers are higher from 33% to 40%. The paper "A New Look at an Old Friend" http://www.tronola.com/A_New_Look_At_An_Old_Friend.pdf describes and has influenced my thoughts.
One important point, is that early dynaco research shows that the EL84 tubes prefer a 25% ultra linear tap, and many of the off-the-shelf transformers are higher from 33% to 40%. The paper "A New Look at an Old Friend" http://www.tronola.com/A_New_Look_At_An_Old_Friend.pdf describes and has influenced my thoughts.
Last edited:
Member
Joined 2009
Paid Member
Primary windings are considered to be the experts for the Leak design.
Any high quality 6V6 transformer is suitable for EL84.
Any high quality 6V6 transformer is suitable for EL84.
This is way too low for EL84.
That's typically 6L6.
For lowest distortion you need to go fixed bias, higher A-A load, not cathode bias, (as David C. Gillespie has shown so well), as well as regulation of the bias line AND preamp/driver HT.
Transcendar are outstanding stuff.
8K Push-Pull with 25% Ultra Linear Taps to 4 & 8 ohms 17.5 Watts
Frequency Response: 10 Hz to 95 kHz +/- 1 dB @ 1 Watt
If you check, you will find many of the very best 6V6 transformers are 8-12K A-A.
(eg.DB110), and don't use ultra linear at all.
6k plate to plate primary is too low for a push pull EL84 in Pentode mode, and too low for a push pull EL84 in Ultra Linear mode.
But . . .
An EL84 in triode mode has a plate resistance of ~ 1500 Ohms.
A 6k Ohm plate to plate transformer has 1500 Ohms plate to Center Tap.
With that, an EL84 triode wired push pull in AB1:
When one EL84 cuts off, then 1500 Ohm plate resistance drives 1500 Ohms of the output transformer 1/2 winding.
The damping factor at that point is near unity (1).
An EL84 in Ultra Linear mode that has only a 25% UL tap, will have plate resistance about
3 to 5 times the triode plate resistance of the same tube in triode mode.
The damping is far less than unity (<<1). The 25% tap may be optimum, but only with Lots of negative feedback.
It seems better for EL84 to use 8k or 10k plate to plate for UL.
An EL84 in pentode mode has a plate resistance of 38k Ohms. You will need Lots and Lots of negative feedback to get the damping factor anywhere close to unity.
But . . .
An EL84 in triode mode has a plate resistance of ~ 1500 Ohms.
A 6k Ohm plate to plate transformer has 1500 Ohms plate to Center Tap.
With that, an EL84 triode wired push pull in AB1:
When one EL84 cuts off, then 1500 Ohm plate resistance drives 1500 Ohms of the output transformer 1/2 winding.
The damping factor at that point is near unity (1).
An EL84 in Ultra Linear mode that has only a 25% UL tap, will have plate resistance about
3 to 5 times the triode plate resistance of the same tube in triode mode.
The damping is far less than unity (<<1). The 25% tap may be optimum, but only with Lots of negative feedback.
It seems better for EL84 to use 8k or 10k plate to plate for UL.
An EL84 in pentode mode has a plate resistance of 38k Ohms. You will need Lots and Lots of negative feedback to get the damping factor anywhere close to unity.
I simply cannot understand this infatuation with damping factor.
It's largely irrelevant, and read this to understand why your reasoning is flawed...
Fact is, connected as a triode the EL84 loses practically all attraction,- its gain.
That's why singly they are a waste of time,- especially if you want to use non optimal A-A loads, hence why the better solution which such small glass is PPP or simply TV tubes.
Another fact, there are much much more linear better more modern valves out there which can be wired parallel push pull.
Later cathode materials were way more efficient, so they could get twice the cathode current using the same heater.
Some of them are in the 1USD a pop bracket.... like the 6AV5 used to be but you can still get 6EZ5 for 2-3USD NOS and xGE5 which looks like a 2A3 triode wired...
As they approach high loads then have an anode impedance more like 6k, eg. the GEC Axxxx which in PPP look more like a small KT88 than an EL84.
As a PPP triode if you insist in going there, they are down at 400 ohms, and even the xGE5 is down at 800....
Using PPP pentodes will net you easily 30W (I have managed 50!), with incredibly good linearity & low distortion at 6K A-A load.
Just saying....EL84 is old hat.
It's largely irrelevant, and read this to understand why your reasoning is flawed...
Fact is, connected as a triode the EL84 loses practically all attraction,- its gain.
That's why singly they are a waste of time,- especially if you want to use non optimal A-A loads, hence why the better solution which such small glass is PPP or simply TV tubes.
Another fact, there are much much more linear better more modern valves out there which can be wired parallel push pull.
Later cathode materials were way more efficient, so they could get twice the cathode current using the same heater.
Some of them are in the 1USD a pop bracket.... like the 6AV5 used to be but you can still get 6EZ5 for 2-3USD NOS and xGE5 which looks like a 2A3 triode wired...
As they approach high loads then have an anode impedance more like 6k, eg. the GEC Axxxx which in PPP look more like a small KT88 than an EL84.
As a PPP triode if you insist in going there, they are down at 400 ohms, and even the xGE5 is down at 800....
Using PPP pentodes will net you easily 30W (I have managed 50!), with incredibly good linearity & low distortion at 6K A-A load.
Just saying....EL84 is old hat.
6vheater,
Build us a push pull pentode amplifier, and do not use any negative feedback.
You will get lots of gain, and lots of power.
Then bring the amplifier to one of our audio crawls, and let us listen to it on several different speakers.
You will find out that damping factor IS a factor in the sound.
Pentodes require the highest amount of negative feedback to get a reasonable damping factor.
A voice of the theater does not require lots of power when it is in a medium sized room.
Nor does near field listening with a medium efficiency loudspeaker.
Next amp . . .
Build us a push pull pentode amplifier, and do not use any negative feedback.
You will get lots of gain, and lots of power.
Then bring the amplifier to one of our audio crawls, and let us listen to it on several different speakers.
You will find out that damping factor IS a factor in the sound.
Pentodes require the highest amount of negative feedback to get a reasonable damping factor.
A voice of the theater does not require lots of power when it is in a medium sized room.
Nor does near field listening with a medium efficiency loudspeaker.
Next amp . . .
Last edited:
Don't understand half of that, it sounds like DOGMA.
I don't need to crawl,- I can see and hear linearity when it's obvious.
Perhaps you never ever worked with PPP pentodes. (NOT beam tetrodes).
Your opinions are not at all shared by Jack E, who is an expert on damping, IMD and NFB.
NFB makes the sound horrible & lifeless, just like most of those MACs, and lots of the "modern amps" with supposedly fantastic sound.
Not only do they sound boring and dry, but they also blow up easily.
(Watched a EL34 amp on test where the grids glowed WHITE HOT borderline melting at only 1/4 claimed power output).
J.E. has some highly unorthodox designs using pentodes that simple have nothing in common with what you claim, inc an autoformer phase splitter.
Perhaps you should look them up.
I produced some PPP pentode amplifiers which sound outstanding & still have low distortion at only -1dB from clip.
FYI, there are not all that many decent TRUE pentodes out there, least of all nice modern ones post 1950.
Well, I was only ventilating the findings published in a book for audio enthusiasts (circa 1960) by Philips, a rather large multinational holding several electronics patents, among which the penthode valve.
A "crawl" is a gathering in case you didn't know.
In my experience and testing, open loop amplifiers don't particularly sound good, especially pentode/tetrode. A beefy triode can make it work, but that same amp will sound MUCH better with ~8db of gNFB in my testing. Flatter frequency response, less distortion, and much better control of the drivers on low frequencies (<40Hz). Oh, and square waves that don't look like shark teeth!
IMHO if you're going to go pentode, you might as well go transistor.
I suppose now we have to talk about history, and that maybe involves a bit of personal experience?
Philips did nothing, it was mostly down to their British juggernaut Mullard, who employed masses of people, and a sizeable office in Mitcham.
It was Mullard did the EL84, and pushed the little thing as a modern alternative to the 6V6.
(a 2.8W heater beam tetrode which specifically was invented to get around Philips' stranglehold on the pentode patent.)
The 6V6 is an excellent low power consumption transmitter valve for mobile use, and very low screen diss.
They are not known for melting, even when grossly overloaded, and originating from the 1930s.
They are still made by the zillion today.
The 6BW6 was the EL84 in noval, while the tiny Axxx4 was GEC's answer to it in a much much better tiny little 7 pin pack.
PR was not invented yesterday.
So, in short if you believe anything from Philips, you will believe anything.
Luckily the US & their brilliant Sylvania had other ideas.
Having found the 4.8W heater EL84 thing was plenty unreliable, no good for TX on HF, no good for mobile & prone to melting the glass, they attempted to upgrade the thing several times by pretending to make a better version called the 7189A etc.
(Scott luckily got wise to this and got into the 7591act, as a seriously upgraded 6V6 which is much much better valve. period).
Scott used the EL84 for ages then sensibly made the beautiful later stereomasters with proper valves and whopping transformers. I wonder why?
In fact late Westinghouse 6V6 look identical to early 7591, but the latter a 6.3W heater instead of the former's 2.5W.
If you want to compare the poor short life EL84 against the 7591, which often are still going strong 40-50yrs after, you have it all in a nutshell.
Luckily the Russians copied the EL84 successfully and made the 6P14P-EV, because television had been invented and they needed to cram the things into tiny spaces but keep them cool eg. (EL86/ECL86).
Russian TV wasn't exactly hi-tech.
An octal valve was never going to be popular any more, so the US, made the novar 7868 (pointless because couldn't compete with that NOVAR base), and finally the 6GM5 which there is no doubt is a serious quantum leap better, all crammed into an unlikely size small bottle....*
Sadly like betamax/VHS fiasco, the world had decided the 1953 EL84 developed from the earlier EL3/41, was better or easier to make than the thoroughly modern and far better, far more rugged 1960s 7591.
The 7591 continued on in US TVs in another form into the 1970s in a superb 12 pin 19W Pa upgraded form*,it's much bigger, and better in every way, but practically everyone is ignorant of that, and want to carry on with the unreliable overheating old 12W Philips thingy.
Ask Scott, Fisher, Eico, in fact look at this!
I know what I prefer.
Btw my pentode square waves look pretty impressive, so where do you get the "dogma" from?
Books written by Philips?
Philips did nothing, it was mostly down to their British juggernaut Mullard, who employed masses of people, and a sizeable office in Mitcham.
It was Mullard did the EL84, and pushed the little thing as a modern alternative to the 6V6.
(a 2.8W heater beam tetrode which specifically was invented to get around Philips' stranglehold on the pentode patent.)
The 6V6 is an excellent low power consumption transmitter valve for mobile use, and very low screen diss.
They are not known for melting, even when grossly overloaded, and originating from the 1930s.
They are still made by the zillion today.
The 6BW6 was the EL84 in noval, while the tiny Axxx4 was GEC's answer to it in a much much better tiny little 7 pin pack.
PR was not invented yesterday.
So, in short if you believe anything from Philips, you will believe anything.
Luckily the US & their brilliant Sylvania had other ideas.
Having found the 4.8W heater EL84 thing was plenty unreliable, no good for TX on HF, no good for mobile & prone to melting the glass, they attempted to upgrade the thing several times by pretending to make a better version called the 7189A etc.
(Scott luckily got wise to this and got into the 7591act, as a seriously upgraded 6V6 which is much much better valve. period).
Scott used the EL84 for ages then sensibly made the beautiful later stereomasters with proper valves and whopping transformers. I wonder why?
In fact late Westinghouse 6V6 look identical to early 7591, but the latter a 6.3W heater instead of the former's 2.5W.
If you want to compare the poor short life EL84 against the 7591, which often are still going strong 40-50yrs after, you have it all in a nutshell.
Luckily the Russians copied the EL84 successfully and made the 6P14P-EV, because television had been invented and they needed to cram the things into tiny spaces but keep them cool eg. (EL86/ECL86).
Russian TV wasn't exactly hi-tech.
An octal valve was never going to be popular any more, so the US, made the novar 7868 (pointless because couldn't compete with that NOVAR base), and finally the 6GM5 which there is no doubt is a serious quantum leap better, all crammed into an unlikely size small bottle....*
Sadly like betamax/VHS fiasco, the world had decided the 1953 EL84 developed from the earlier EL3/41, was better or easier to make than the thoroughly modern and far better, far more rugged 1960s 7591.
The 7591 continued on in US TVs in another form into the 1970s in a superb 12 pin 19W Pa upgraded form*,it's much bigger, and better in every way, but practically everyone is ignorant of that, and want to carry on with the unreliable overheating old 12W Philips thingy.
Ask Scott, Fisher, Eico, in fact look at this!
I know what I prefer.
Btw my pentode square waves look pretty impressive, so where do you get the "dogma" from?
Books written by Philips?
Attachments
Last edited:
Short life EL84? There are still many table radios from the 1950ies in use with their original set of tubes, including their EL84's. And still they're going strong.
Any reputation of being short living (if so...) most probably results from notable guitar amplifiers that run them way out of ratings.
Best regards!
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.