Theoretical OPT question

[weinstro]

Hello,

I have a theoretical question. Let's say that I wanted to purchase an inexpensive PP OPT, but wanted to avoid some of the winding imbalance issues (different Z) that seem to be associated with these.

Could I not simply use two smaller SE transformers, primaries connected in series to yield a center tap, and secondaries connected in parallel?

I recognize that I would probably have to juggle impedance and power ratings to yield something feasible.

Your thoughts appreciated!

[richwalters]

Originally posted by weinstro
Hello,


Could I not simply use two smaller SE transformers, primaries connected in series to yield a center tap, and secondaries connected in parallel?


Your thoughts appreciated!

Nup....Amperés 3 finger magnetic torque rule. For push pull You require a common core with BOTH primary halves wound opposite directions to cancel the magnetic flux created from each half. Remember what the phasesplitter does ?

ce tout

richj

[DougL]

Could I not simply use two smaller SE transformers, primaries connected in series to yield a center tap, and secondaries connected in parallel?

To me, the question isn't whether you can get sound from that arrangement, (I think you can) but what are you optimizing, and what are you degrading?
An SE transformer requires a relatively large air gap to function.
Since DC flux is not canceled, it requires a larger core for the same base response. This also degrades the high frequency response by adding more turns. Bottom line, 2 larger, more expensive SE transformers with more compromises substituting for 1 smaller push-pull transformer.
If I was going to propose an alternate solution, para-fed the transformer and use either plate chokes or CCS.
The CSS requires twice the voltage as the transformer or plate choke, and twice the voltage is about 4 times as costly power supply. This offends my Scots ancestry, but may not bother you. The plate chokes also add cost, and require an additional coupling cap.

YMMV

Doug

[Miles Prower]

Originally posted by weinstro
Could I not simply use two smaller SE transformers, primaries connected in series to yield a center tap, and secondaries connected in parallel?

That only works for RF power amps, and where you're using air core xfmrs. In that case, you don't have the DC magnetization problem, and since you're operating over a very narrow band, you can balance with trimmer capacitors. Even here, good balance and low harmonics aren't so important since you will always have one or more tuned circuits and/or bandpass filters between the amp and antenna.

For audio, that might be a "quick 'n' dirty" solution, but for sonic excellence, that dawg ain't gonna hunt. You can either go OTL (which has problems of its own) or bite the bullet and get a decent OPT.

[weinstro]

Originally posted by Miles Prower


That only works for RF power amps, and where you're using air core xfmrs. In that case, you don't have the DC magnetization problem, and since you're operating over a very narrow band, you can balance with trimmer capacitors. Even here, good balance and low harmonics aren't so important since you will always have one or more tuned circuits and/or bandpass filters between the amp and antenna.

For audio, that might be a "quick 'n' dirty" solution, but for sonic excellence, that dawg ain't gonna hunt. You can either go OTL (which has problems of its own) or bite the bullet and get a decent OPT.

Thanks everyone for the responses. In short, Bad Idea.