Strange idea for an output stage - comments requested

[Fenris]

So I had an idea the other day while looking at parafeed designs and reading the thread on using power toroids as OPTs. I've got a pretty big junk box an have been looking for a project to clear alot out of it. Starting with a standard Williamson style front end - plate follower with split load phase splitter, nothing exciting here. Then a 6L6GC in parafeed with a CCS. I figured I'd add a second output tube in paralell, but a dual primary power toroid would give me a little more flexibility in wiring. I was thinking that running the tubes and transformer windings in opposite phases from each other would help with overall linearity. Has anyone tried something like this or see any pitfalls? I'm thinking it still has overall low power output compared to a P-P amp. If there's a major problem, I'd just go with a more standard P-P arrangement, but I figure I'd ask the experts first. Thanks.

PS. - B+ of around 360v.

[Iain McNeill]

I'm a sand-guy so forgive me if I'm missing something here. If the primary windings were ground referenced rather than cathode referenced wouldn't this be equivalent to a PP output, assuming zero PSU impedance?

[smoking-amp]

It will sound like P-P. Some possible advantage as far as PSRR, since the primaries are ground referenced instead of B+ reference. And the xfmr will be DC balanced. But those CCS's on the outputs are going to burn up power. And the output tubes have to run in class A too. I think this may be close to summing the disadvantages of SE parafeed with the disadvantages of P-P.

I think the caps to the primaries could be dispensed with if the primaries are just series connected between the plates. And the CCS's could be dispensed with by using a separate center tapped inductor to B+. But then it loses the PSRR advantage. Although P-P largely cancels out the Power Supply effects to first order anyway.

[TheGimp]

Para-Feed push pull. Should work.

Try it and tell us how it sounds.

[ChrisA]

You are using a power transformer for output? I doubt there is any interweave in the coils. What is their spec for leakage inductance? Power transformers are not designed to work past 60Hz.

[analog_sa]

Originally Posted by ChrisA

Power transformers are not designed to work past 60Hz.

Toroids work just fine. That's why they shouldn't be used in power supplies.

[rdf]

An easier option might be to put the CCSs on the cathodes bypassed with a cap. They make it trivial to null static DC in the toroid primary.

Edit: CCSs in the plate circuit means losing half the B+ to permit Vp swing.

[planet10]

Originally Posted by analog_sa

Toroids work just fine. That's why they shouldn't be used in power supplies.

[Geek]

Originally Posted by ChrisA

You are using a power transformer for output? ..... Power transformers are not designed to work past 60Hz.

They rock in guitar amps though

Cheers!

[ystkkmkr]

http://www.anc-tv.ne.jp/~suzuki3/amp...300Bpp_Amp.pdf

This schematic will be of your interest. You can find it on a Japanese DIYer's website. As the website says in Japanese (sorry), the original design of this SE-PP switchable class-A 300b amp was actually published by Mr. Takaaki Uzike (or Ujike) in a couple of issues of RADIO GIJYUTSU magazine (a monthly Japanese DIY magazine like MJ) in 2000. In the articles, he says he developed the circuit and assembled his 300b amp, in order to make the best use of Permalloy PP output transformers, and named this output stage configuration "floating OPT", which I think is equivalent to "parafeed PP". Mr. Uzike used a pair of Tango NY-45-5 (discontinued).

Yoshi