801a maximised class A2 excursion.

What people are working on at the moment
Post Reply
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#76 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

When it’s possible again, maybe I’ll bring it to try on youre speakers.

The huge Tribute are all I can find thta’s anywhere near 10k and one thing clear as day; all these valves look bad at 2.5k my only other SE transformers.This is a shame because it all began with me wanting to just hook up a little SE amp and Ive ended up building a boatanchor. But I suppose that is destiny.
Last edited by Paul Barker on Mon May 11, 2020 9:37 am, edited 1 time in total.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#77 Re: 801a maximised class A2 excursion.

Post by Nick »

Don't worry about the typo's Paul we normally get the gist of what you are saying.

I am with Steve on this one, If your speaker needs 20W then you probably want to build a push pull amp either valve or solid state, if it needs 50W then solid state. In both those cases the speaker will be the limiting factor even if you make a 30W 833 amp. And at that point Physics limits what you can get from the transformer in terms of bandwidth.

And the other option is what’s Marks doing and use multiple amps for multiple drivers, again, doesn't require doing the heavy lifting with SET's.

Not to say what you are doing is not worth doing, I think it is, but I would suggest you focus your energy on a problem that needs solving (getting the most from a valve), not one that doesn't (getting lots of power from a single valve).
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#78 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

yes to all that.

Actually having refreshed some sleepy brain cells all of this is coming back to me, and my probable direction beyond this project is a Matrix amp.

I’m so anoyed that at vsac i was stood next to Stephie and sorry my mind is older and ive forgotten the name of the man who writes Tubecad. It seemed such a great conversation but I didn’t join in because I didn’t know Stephie was Steve. I just recognised this person’s brilliance but just needed that nudge to engage in converstaion and lost the moment.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
steve s
Shed dweller
Posts: 2829
Joined: Wed May 30, 2007 6:19 pm
Location: east yorks

#79 Re: 801a maximised class A2 excursion.

Post by steve s »

I have a pair of 4.2k, they look good transformers, but unknown maker, your welcome to have them Paul. It would also be great for you to try on my speakers. Once we are through this of course.
Sorry my tributes are all in use at the moment. I've been busy since Christmas.. I'm hoping pieter will wind me another pair.. But that takes a long time.

I too have to watch my spelling and grammer, every once in a while an unedited post slips through, but Im slow to type and now read/correct several times before posting. I like to think I'm gifted in other ways.. But in what I don't know.
The tube manual is quite like a telephone book. The number of it perfect. It is useful to make it possible to speak with a girl. But we can't see her beautiful face from the telephone number
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#80 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

Thank’s Steve I appreciate and accept your’e offer. I had a beautiful small but sweet sounding 4.5k se transformers but I’ve lost them. They were the best bar none Ive heard at low level. Ho hum!

So here’s the calc’s for the 211. Im using the curves on Thomas Mayer’s Tube of the Month on the 211. Because the +20v to the -60v grid lines fit and make all the current and voltage measurements simple on theose curves Ive filled it in acording to that which brings tiny bit over power output, seems you can’t easily hold the 211 down to 801a levels. It seems to in the evnt have more gain than 811-10 so the Svetlana 10 is optimistic the RCA 10 is on the money. But 811-10 is near enough. I can’t fathom what the 2nd Harmonic truely is, I’m guessing we interpret the negative to mean zero 2nd harmonic, I have to think that because the curves look so linear. Same with the 3rd and 4th they show positive but usually it shows as a negative figure. I suspect that if the formula showed db it would be easier to interpret where the distortion is so low that it flips the polarity of the calculator. Open to other views.

Image

Anyhow, on paper the 211 quality romps ahead of the other two from the get go.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#81 Re: 801a maximised class A2 excursion.

Post by Nick »

Just a guess, but I would think the +ve and -ve distortion just means if they are expanding or contracting. The non linearity can be either less output at a given input voltage level (contracting) or more voltage for the given level in (expanding). Both are distortion.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#82 Re: 801a maximised class A2 excursion.

Post by Nick »

You are being a bit gentle on the 211 though at 500v 40ma.
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#83 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

Nick wrote: Mon May 11, 2020 11:45 am You are being a bit gentle on the 211 though at 500v 40ma.
Very true but the starting point of all the valves in the project is the 801a operating point to compare them all at that. Then to see whta they’ll od in their own element.

Here I’ve backed off from +20v grid to -120v grid and although current is still low for a 211 I’ve drawn the loadline extremely conservative because where it intersects the 75 Watt curve you’d have to be pushing ac up to 2,200v . Now it’s OK to cross the max dissipation currve or venture to cut a portion of the bulge, because the infringement is only during that part of the signwave, it’s acceptable. But how far can that infringement go. Watch the colour of he anode at loud passages I suppose!

Image

So next I’ll make a judgement how far to push up quiescent voltage and establish how much happier it is in terms of distortion. Might not be discernable, but expecting less distortion.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#84 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

Ok so enough of theory. I stumbled ta the 211 because it transpires the characteristic curves found on Thomas Mayer’s block have confused me. I think there’s something wrong with those curves at the higher voltage, in the sense that the max power line out to the right gets crossed by youre loadline, but as this crossing takes place at ac swing you won’t ever reach I realise we can ignore it. But Im much more comfortable with the grid lines from the main characteristic curves issued by RCA. but those are not suffiently in depth to make possible acurate distortion measurements. A lot of filling in the gaps, you’re 2nd isn’t hard to get acurate enough but whtas going on with 3rd and 4th I for one can’t see and to get it acurate enough to make a comparisson in our situation here isn’t on the cards.

So now is the end of testing and from here on we’ll wait for the build and aural opinions. My only distortion meter states it calc’s THD to -20dB, whihc basiclaly is too limited to bother if I build a 801a/811-10 or 211 with 10% distortion I’ve forgoten much more than I knew before I knew anything. So until we’re at a gathering when Nick or Andrew can bring their better test equipment we’ll put all that junk aside and see whta the three options sound like.

As far as that is going the jeft cascode was an abject failure and I have no understanding of solid state so its probably too steep a leraning curve, so I probably won’t trouble shoot the jfet cascode at the moment. Basically the signwave looks crap even at 200mv input clips fully soon after, and only produces A of 11. A far cry from author’s promises.

I found a circuit of John Broskies using a different, more powerful Jafet source follower into a grounded grid 6sn7 and onwards. He ccs’s the tail made of of jfet source follower and 6sn at 10mA and shares that equally with individual auto bias set resistors. Quite neat.

My next move ; Ive ordered the Jfet’s and plan to od the calc’s to see if I ca. grounded grid the MH41 to get all my voltage gain for this project, pass out of the MH41 with a cf or source follower which I haven’t yet chosen. Not driving grid current so it doesn’t have to be massively high transconductance or power. Possibly 5687. Why I like the cf more is because of the levels, I want to get as much of the build didrect coupled as I can, not sure I can achieve that in entirety. But I’ll build stage after stage , think about the levelling and if it has to get a cap or transformer in one place, so be it.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
IslandPink
Amstrad Tower of Power
Posts: 10041
Joined: Tue May 29, 2007 7:01 pm
Location: Denbigh, N.Wales

#85 Re: 801a maximised class A2 excursion.

Post by IslandPink »

Presumably a good Spice model of the 801a should be accurate to at least 3rd order, Nick ?
"Once you find out ... the Circumstances ; then you can go out"
User avatar
Nick
Site Admin
Posts: 15706
Joined: Sun May 06, 2007 10:20 am
Location: West Yorkshire

#86 Re: 801a maximised class A2 excursion.

Post by Nick »

Yep, the spice model should be at least as good as the published curves. Assuming they use a triode model that can model grid current.

Paul, you are more than welcome to borrow the Hewlett Packard HP 3581A. Its probably the simplest to use to extract individual harmonic levels down to about -80dB
Whenever an honest man discovers that he's mistaken, he will either cease to be mistaken or he will cease to be honest.
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#87 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

Beautiful, thanks Nick, Id be as happy to wait until we’re together again at something if you could bring it and we just use it there. Whatever that future event is.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#88 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

IslandPink wrote: Wed May 13, 2020 12:53 pm Presumably a good Spice model of the 801a should be accurate to at least 3rd order, Nick ?
Well theres plenty of things my expression of humanity hasn’t grasped and one of those is spice. But when I’ve muddled on with my inate abilities however lacking they seem, I’ll publish the circuit values and maybe someone can spice it up for me?
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#89 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

It’s obvious to me that I’m out of my depth with fets and mosfets. So there shall follow a fairly lengthy pause while I observe. Just found a good Broskie page full of it.

So far I haven’t found that the solid state discrete component useage is as well supported by the knowledgeable suporting the mortal in leraning the ABC’s. When I became interested in tube audio design last century it was before google got to where all such material is hidden and covered over by a search engine that only finds things according to the revenue it precipitates a cascade of for Googles long term future.

So it’ll take me a while to get over this first failure. The easy way for me is to just use valves. But there are these devices out there which I want to explore harnessing if they support and possibly improve my efforts with valves.

The circuit I built exactly as drawn and powered conservatively with +30v is the first example with vaules I found that I fancied the amaising positives. But instead of geting elated after succeeding in a high dive in which I barely disturbed the water I belly flopped: LOL! Least I can laugh at myself. I can my weeknesses, because I have proved my strengths over a long period. When we get out of our comfort zone we find out more, even if we fail we learn a lot! So heres the failure. I used the devices asked for, 50ohm feedback resistor, 50k gate resistor (because the uathor states that the input capacitance is so great as to limit bandwidth to 10 khz, but below 50k gate resistor you’re up towards 50khz, in the event my build is so clearly distorting that bandwidth is the least of my problems, Hameg signal generator at the ready, no clena sinewave at any frequency so whats the point sweeping for bandwidth) 3k3 drain load +30v bench supply. Image

But plenty of hope on the horizon, Ive now found out John Broskie really went to tonw on Jfets and Mosfets. There is so much stuff he’s blogged on them I daren’t even say whta netx idea will be. Doesn’t mena I won’t just stick with jfet driving MH41 grounded grid. But the more powerful one Broskie uses.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
User avatar
Paul Barker
Social Sevices have been notified
Posts: 8863
Joined: Mon May 21, 2007 9:42 pm

#90 Re: 801a maximised class A2 excursion.

Post by Paul Barker »

This type of thing, but I’d handle the valve element and go for more A in this one VA stage, probably with mh41 but the CF I’d switch to 5687.
Image and I’d level the output probably with bipolar power supply with sufficient negative headroom to bias all of the output valves under consideration.

Failure paves the road to success.

As a positive I’ve found my power transformer that is up to the task, but at 500mA capability it’s a carry into Owsten on it’s own. If the pub survives.
"Two things are infinite, the universe and human stupidity, and I am not yet completely sure about the universe." – Albert Einstein
Post Reply