Re: Help: output transformer

[TubeGarden]

Hi RATs!

Never mind. Follow advice from people who have wound output transformers and
succeded.

Theory is not much use in reality :)

Happy Ears!
Al


Alan J. Marcy
Phoenix, AZ

PWC/mystic/Earhead

[Patrick Turner]

TubeGarden wrote:

> Hi RATs!
>
> Never mind. Follow advice from people who have wound output transformers and
> succeded.
>
> Theory is not much use in reality :)
>

Gees Al, without some idea about the theory, one wouldn't get too far.
But once the theory has been accepted as fact, you end up with a set of
equations
from which to design how many turns one needs, insulation thickness, wire size,
interleaving layout, and so on. More equations give the final response.
But what trips up many folks who try diy is the actual practical skill
needed for a good tranny, ie, winding something with nice flat layers of
fine wire, which will withstand HV testing when finished, and last for 100
years,
because there are NO crossed over turns in the wind up, and the wax/varnish
impregnation
is evenly spread right through the tranny, and the terminations are all nice and
neat.
One needs a winding lathe to do a good job, and diyers have that hurdle to get
over,
as well as finding a supplier of iron and bobbins,
although I have made plenty of my own bobbins....

Patrick Turner.

>
> Happy Ears!
> Al
>
> Alan J. Marcy
> Phoenix, AZ
>
> PWC/mystic/Earhead

[Mikkel C. Simonsen]

TubeGarden wrote:
>
> Hi RATs!
>
> Yup, theory is useful, but, only if presented in light of practical
> application. Trying to learn how to wind a good output transformer from someone
> who states clearly up front thay have never done it is simply asking to learn
> everything the hard way.

I may not have made any, but I'm quite sure the German guys who wrote
the original article and book have. All I have done is translated the
stuff into English, so I don't see why you shouldn't be able to make
good transformers from those formulas... You may not be able to make the
worlds best transformers the first time around, but I think that will be
the case no matter what formulas you use.

Best regards,

Mikkel C. Simonsen

[Mikkel C. Simonsen]

·à¤l¤f wrote:
>
> How to get 0.8 Ohm 3.2 Ohm 7.4 Ohm and 12.5 Ohm using four 0.8 Ohm
> secondaries in:
> http://stiftsbogtrykkeriet.dk/~mcs/Transformer.html

I fixed the drawing, so it now looks like it did originally. It should
make a bit more sense now :-)

Best regards,

Mikkel C. Simonsen

[TubeGarden]

Hi Mikkel and RATs,

OK. Sorry. I misread the intro :(

I met the guy who handwound silver outputs for the one-electron EL34 PSET in
layers which all run the same direction, with outside loops back to the
beginning of the next layer, which allows closer spacing of layers and gives
better performance. He seemed normal enough :)

I think it is called Z winding, as opposed to W winding? Anyway, John Atwood
said it sounds fantastic. I didn't get to hear it. Meeting the guy who did it
was fun. Next time I hope to hear it. If there is a next time.

Happy Ears!
Al

"If it more complex than two wires and a wire nut, I hire it out to a pro. For
wire nuts with two wires, I am a pro ;)"


Alan J. Marcy
Phoenix, AZ

PWC/mystic/Earhead

[Patrick Turner]

TubeGarden wrote:

> Hi Mikkel and RATs,
>
> OK. Sorry. I misread the intro :(
>
> I met the guy who handwound silver outputs for the one-electron EL34 PSET in
> layers which all run the same direction, with outside loops back to the
> beginning of the next layer, which allows closer spacing of layers and gives
> better performance. He seemed normal enough :)

Huh? the layout of the consecutive primary layers only reduces the
shunt capacitance to the secondary, if the layers
are arranged so the highest signal voltage coming from a group of say four layers,

is arranged to minimise C.
The benefit is quite minimal.

>
> I think it is called Z winding, as opposed to W winding? Anyway, John Atwood
> said it sounds fantastic. I didn't get to hear it. Meeting the guy who did it
> was fun. Next time I hope to hear it. If there is a next time.

One of the main things common with good sounding OPTs
is the minimization of the leakage inductance, and stray capacitances,
widening of BW, and the use of GOSS to keep the iron caused distortion lower than
the tube distortions, and to keep away from saturation, and all these things are
compromised by the mass market makers.
Silver in OPT wire isn't a big deal. Cobalt or amorphous cores
are very overated, and usually claims about great sonic performances
due to over zealous winding technique, exotic materials used, are all marketeering
BS
unless the R,C, and L quantities are optimised.

A common mistake is that some winders place the P sections too close to the S
sections, and sure they get the leakage down low, but the shunt C is high,
and its BOTH that need to be minimised.
The RDH4 has all the rules within to get a decent OPT.
My experience after winding so many OPTs leads me to believe the
RDH4 should not be ignored, ever.

However, one can move beyond what is in the great old book,
as I have, to make OPTs with sections used for CFB windings,
and the placement and geometry of such do affect the stability
and the outcome in terms of measured BW, thd, etc.
Ususally, the greater care taken to minimise thd, AND NFB, the better the sound,
with due respect to output impedance, so I see no benefits
to having a triode amp with a crummy OPT, and 26 dB of NFB.

look after the tubes, give them nice iron,
the tubes look after your ears!

>
> Happy Ears!
> Al

Patrick Turner.

>
>
> "If it more complex than two wires and a wire nut, I hire it out to a pro. For
> wire nuts with two wires, I am a pro ;)"
>
> Alan J. Marcy
> Phoenix, AZ
>
> PWC/mystic/Earhead

[·à¤l¤f]

It will be my first pair of home made output transformers.
Thank you very much.

"Mikkel C. Simonsen" > ¦b¶l¥ó
¤¤¼¶¼g...
> ·à¤l¤f wrote:
> >
> > How to get 0.8 Ohm 3.2 Ohm 7.4 Ohm and 12.5 Ohm using four 0.8 Ohm
> > secondaries in:
> > http://stiftsbogtrykkeriet.dk/~mcs/Transformer.html
>
> I fixed the drawing, so it now looks like it did originally. It should
> make a bit more sense now :-)
>
> Best regards,
>
> Mikkel C. Simonsen