To be or not to be, UL or quad?

[Larry Lomax]

Well, my expectation is not to rise a flame about which configuraton is better, Haffler & Keroes and Mr. Quad allready did.
The real question is turning me mad, is this:

I'd like to make a PP 807 amp since I own a lot of this tubes original and allmost brand new.
I allready made the separate super power supply.
But....
What about the output transformer?
Yes, I made all calculation, I gathered the iron and I'm fanatic to wind it by hand by myself.
Yes, I have good reasons to do it.
But......
What I'm going to wind this OT? For Ultra Linear with separate windings for ultra linear feedback, or for Quad like kathode feedback?
As more as I read as more I get undecided, this question lead me to a stop in my project.
Can anyone help me with sound arguments to take this decision?
Has somebody made something like this before?
Yes, I know,
Why to use 807? Why not a better tube?
Why ultralinear? Why quad? Why not just SE triode?
Why not to buy a commercial made OT?
Why not etc.... All are good questions but I need an answer to my question!
Thanks a lot for your opinion, best regards.
Larry

 

[richwalters]

hi there.........apart from that wretched top cap and the UX base the 807 is an excellent tube (I like it).......... So many amps used this tube during the 1950's & 60'si n lieu of EL34's and KT66's. It would be a pity to ditch it. In quitar amps in tetrode mode it sounds better than EL34's.
Winding output transformers is a late night black art job......if you can master it then you are the master of amp design....the appalling reality is the high number of primary turns and care in assembly. If you use parallelled o/p stages then life becomes alot easier. No winding technique difference between UL & true P-P, just a matter where you want the % taps. If you get UL a bit out..i.e leakage inductance and capacitance not symmetrical in each section, one can end up with a darned good oscillator. If you speak to a winding/transformer house (who knows about how to wind audio transformers) they will wind it nicely for you....at a cost but it may not work properly. However they won't help you with hints on making a home made one. This should'nt
deter you from making one, providing you have the procedures.
I have winding details on a 20W version using EL34's but the principal is the same for larger designs.
I have no facility for copying this on computer but I can fax it readily if you want me to.
An 807 single-ended with 6K primary # roughly 75H to hit the
-1dB @ 30Hz ballpark Bmax at 1.35T requires alot of wire....I think too much to damp the treble. (only my opin). If you use parallelled SE o/p stage, (i.e anode Z halved) winding life becomes more tolerable with better bandwidth.

UL-P-P offers more power and bandwidth.. so many options..
Re-study the tubedata sheets on this tube.....they do give operating conditions......
Youv'e made a super PSU.......can we have a peek at it ?

rich

 

[audiousername]

There are a fair bunch of links to transformer winding in this post.

Don't be worried about using 807s.... just because they're less popular than say, EL34, does not mean they're bad valves

Since you're intending to wind your own - you could try using both UL and CFB, can't you?

 

[Brett]

807's are great tubes. And I like topcaps.
Personally, I'd either go tetrode or triode and leave UL, I think it's pretty poor sonically. SE triode will give very little power.
If you're going to wind your own OPTs (I think you're mad) then cathode feedback would be better than UL. Partial feedback can also sound excellent. I'd suggest you just buy OPTs; lots available in Europe (Lundahl, Sowter etc) and even Hammonds.

 

[Larry Lomax]

Thanks for all the good wise suggestions.
Yup, I'm mad, sick of DIY, I allready made a lot of OT, I own the know how to make them, to calculate the number of turn for the windings etc.... The threat was exciting in late 60's and that was a beatuful time on making tubes things.
But, after 35 years afterer, transistor, MOS, micropros, computers, robots and a lot fun stuff, I landed back to the old love, tubes audio.
And... yes making OT is a dark wizards, after dinner job. But, let me tell you, when you turn your amp on .... and it works... is a feeling like wen you first sun is born! (All those times nothing goes wrong). You can demostrate to yourself that you did it.
Nice opinion about 807, nobody likes this tubes here in my home town. At least 300B single ended or Futtermans are allowed by fine hearing people
But 807 is what i get. And I like a little extra power for modern music, after that I'm making an experiment.
How far I'm mad? A lot. I allready winded, for the power supply:
The main trasformer 220/600V 500W
The filament transformer 220/10V x4 250W
The pre-driver and bias transformer 220/180-200-250-300V/100V
The chokes for all the above PS sections since I'll use choke input filter for all rectifing sections.
The filament are all 4/6.3V stabilized by LM338 regulator, mounted on heat sink and driven by a stand by timer for warm up a 4V.
All the transformers and inductances have hand crafted brass made covers. I'll post a pic of it in the pics section.
Is this enugh to show how sik of DIY I'm?

But my real undecision still remains here. I'm ready to start winding, I allready made the core, I made it off corse. And I started with the convincement of making a UL. But sombody changed my mind, to kathode feedbak, esier to wind (less fun
), esier to polarize the tubes, but hard to drive....
And after, your, more than welcome, opinions, I still can't find a good reason to be decided for the UL or Quad. Both? I fear a waist of time. I can make just one section first, try one and try the other, making two OT, or wind both. But a good OT must be optimized and I cant'fit a lot of windings in it without making a lot of compromizes.

Thanks again for your comments.
Sincerely Larry.

 

[audiousername]

Nice opinion about 807, nobody likes this tubes here in my home town. At least 300B single ended or Futtermans are allowed by fine hearing people

Blegh... the 'fine hearing people' aren't really hearing at all

My foggy memory tells me that someone important and/or well-respected (possibly Norman Crowhurst) made a convincing argument that the combination of distributed load operation with cathode feedback taps was the best topology to use for the output stage... (I've never tried it - can't afford any OPTs which have those taps and too fundamentally lazy to wind my own )

If you have to choose between the two, cathode feedback may be the one to go with. It seems that this connection reduces distortion quite well - a reduction of 3dB in gain has been found experimentally in a 6528 SE amp to reduce 2nd and 3rd harmonics by around 6.5dB and 5th by some 13.7dB.

Another thing to consider is "partial feedback", as suggested by Brett. Basically hook up the valves in fixed screen mode to an ordinary OPT and use shunt feedback from the anodes to the grids of the output valve. This of course will decrease input impedance, so you'll need a driver capable of swinging some current, and with a high output impedance (to keep b large).

 

[richwalters]

Thanks for all the good wise suggestions.
Yup, I'm mad, sick of DIY, I allready made a lot of OT, I own the know how to make them, to calculate the number of turn for the windings etc.... The threat was exciting in late 60's and that was a beatuful time on making tubes things.

Hi there...It looks youve got a fair power tranny.....go for the big one....Why change ways of thinking ? I've botched up audio circuits over 40 years.....had some shocks but that's the kick out of it still staying alive and doing more.... DIY audio sounds way better than bought audio equipment. Never be put off by new challenges......this physics stuff never ends. I've got to the stage of winding any kind of mains tranny like you have but personally I avoid o/p transformers.

rich