Years ago, I owned an SET amp with 26 tube driving 300b into Push-pull OPTs. It was designed and built by Paul Hogan. To this day it is the best sounding amp I’ve ever owned. Unfortunately, I sold it in 2001.
So, how do I make PP OPT work in a SET amp?
So, how do I make PP OPT work in a SET amp?
Did it by any chance have a pair of large plate chokes for the 300Bs? If so, it was probably a parafeed design.
I think you mean John Hogan, and yes, he used ancient PP transformers in a parafeed design. The 300b's load is a plate choke, with a cap from the plate to the output transformers to keep DC current off the primary. This is not hard to do--all the Bottlehead 300B designs use parafeed. But duplicating John's amps would be difficult only because he used very specific parts and presumably tuned everything to his liking. But it could be done. ;-)
http://glowinthedarkaudio.com/john-hogan-fiprimer-300b.html
http://glowinthedarkaudio.com/john-hogan-fiprimer-300b.html
Last edited:
I heard several of his amps and he said they were series feed. Here are a couple of photos of one of his amps and I don't see any large plate chokes.
ray
ray
There are two small chokes wrapped in tape on top, on the right. Another version I saw had them tacked on the back of the amp. They could also be power supply chokes, but it's impossible to tell what's going on in there. The output transformers also appear to be different sizes. ;-) Amazing that they sound like anything at all.
To use a PP OPT into SE, you must take off the Idc magnetizing current circulating through the primary. That can be done via parafeed connection, coupling the transformer via a DC blocking capacitor. Another options would be to add reverse DC current to a winding in order to cancel the real primary Idc, but that requires a CCS and wastes power.
Note that some PP OPTs may give a mediocre frequency response when connected in SE mode. Capacitance between windings will be differently distributed.
Note that some PP OPTs may give a mediocre frequency response when connected in SE mode. Capacitance between windings will be differently distributed.
I know the PP transformers are the point of this thread, but I feel obliged to refer to the Fi Primer, written by the inestimable jc morrison. Wayback machine brings it back: http://web.archive.org/web/20020202235915/http://tinpan.fortunecity.com/saints/668/primer/
You have 3 solutions:
1) get as much info as you can about the OPT and see if it makes sense to undo the core and put it back with an airgap to work like a standard SE OPT
2) para-feed
3) Compensated drive like in the Klimo Beltaine suggested by @waltube (i.e. one half of the transformer loads the 300B SE output stage and the other half has a "passive" tube, like the EL34 in the Beltaine, that compensates the DC current of the 300B and thus preventing saturation.
I've done my share of sloppy projects but I'm struggling to understand the duct tape thing going on here.
Maybe a PP xfmr could be used to good effect in an SE amp. The center tap gets the B+ voltage; the tube gets one end, and the other end goes to a constant current sink (CCS) set to the tube's quiescent current value in order to cancel out DC core magnetization inevitable is such designs. The CCS, having a very high internal impedance, would not load the signal much. Much more magnetic swing in the core would be made available to the amplifier with this approach.
The output transformers are the same size but one is rotated. John actually sent me a pair of them which I have used in my attempts at a PP amp.
I saw the two parafeed chokes but they seem to be labeled 12H which seems very small. I also can't find a suitable parafeed cap. The smallest caps seem to be 100 uf or 0.22 uf.
ray
Oh, I see about the OPTs. No, I don't see any parafeed caps either. Those must be filter chokes. I guess they are series feed. Can't imagine the bandwidth is anything to write home about, but who knows? I suppose they must sound pretty good. I'll never feel bad about a homebrew build again. ;-)
Eric Barbour's "Single ended glory for $100" article in Glass Audio v7n3 (1995) used P-P output transformers. They operate near saturation, of course, producing plenty of second harmonic distortion. But many find that to sound pleasant if it's not driven too hard.
In 2019 I believe it was I gave my solution "a simple tube circuit transforming a pp output into a single ended one" which can be found here under that heading. My idea was using an extra tube (pl 519) which is floating drawing the same current as the output tubes ( 2x vt4c) but flowing in the opposite direction. In case the opposing currents are equal there is no voltage across the primary and the core is not magnetized. Of course a penthode is used so as not to load the ac source too much. This way the whole primary could be used instead of only one half as others proposed. It worked but I didn't take into account this was a static "solution" and not a dynamic one. How that could be solved can be found there. Anyway, I learned a lot from the comments.
I suggest again to search the diagram of Klimo Beltaine and try to replicate working in the real world testing on the desk the circuit.
Also simple to do because you can replace the 300B with an EL34 triode strapped, then check the other parts of the circuit.
The EF184 is cheap and easy to find.
Walter
Also simple to do because you can replace the 300B with an EL34 triode strapped, then check the other parts of the circuit.
The EF184 is cheap and easy to find.
Walter