Question (again) about using PP transformers for parafeed

[Tre]

I know this has been covered before but....

What is the down side of using a push pull output transformer as a parafeed output transformer?

PP transformers don't have a gap and some use nickel in their cores.

I remember someone saying something about the capacitance around the ground point (center tap) of the primary but why would that matter? The center tap would not be connected to anything.

I would like to try parafeed but I would like to use transformers I already have. If there will be a large downside of doing this, it won't tell me much about parafeed.


Thanks, Tre'

[Paul Joppa]

The art of designing broadband transformers is largely balancing the winding capacitances and leakage inductances so that they produce a smooth and extended treble response. Mis-using a push-pull transformer as a SE parafeed will generally produce a lumpy treble response with unexpected resonances. Since capacitance to ground is unimportant where the signal voltage is small, the designer is free to have large capacitances to ground, but when those are at half the signal voltage they will create problems well below the normal high frequency extension point.

I don't have measurements, but I do have experience. I did this myself when I was first experimenting with SE designs and did not understand transformer details like these. I used an excellent Stereo-70 output with a 2A3 in parafeed. It was quite disappointing! I replaced it with a $6 line transformer from Radio Shack that probably weighed 5% as much but greatly improved the sound. That was in fact the genesis of the first Paramour, but it's also a transformer lesson I've not forgotten.

I do encourage you to try it - there is nothing so valuable as your own direct experience. Just don't use the experiment to judge parallel feed.

[Tre]

Thanks for the response. Here's the part that I don't understand.

"Since capacitance to ground is unimportant where the signal voltage is small, the designer is free to have large capacitances to ground, but when those are at half the signal voltage they will create problems well below the normal high frequency extension point."

If I insulate the mounting of the transformer and it is not touching ground anywhere, is the capacitance still a problem?

Tre'

[Tre]

"The art of designing broadband transformers is largely balancing the winding capacitances and leakage inductances so that they produce a smooth and extended treble response. Mis-using a push-pull transformer as a SE parafeed will generally produce a lumpy treble response with unexpected resonances. Since capacitance to ground is unimportant where the signal voltage is small, the designer is free to have large capacitances to ground, but when those [capacitances to ground] are at half the signal voltage they will create problems well below the normal high frequency extension point."

Sorry, but I have to ask again.

1. What is meant here by "ground"?

2. If the case and core of the transformer are not grounded then how can there be capacitances to "ground" at the half signal voltage point?

I'm not trying to argue, I'm trying to understand.

Thanks again, Tre'

[Paul Joppa]

Good points. Two relevant things are generally grounded - the case and core of the transformer, and the secondary windings. The secondary is interleaved with the primary so the primary is adjacent to grounds at several places.

Those points are grounded primarily for safety reasons, and the core secondarily to help control hum pickup.

I know, that's not much detail, but this is a forum post, not a text book! I can recommend the discussion of transformer design in the Radiotron Designer's Handbook 4th edition, aka RDH4 or "the red bible", for much more information and details.

[Grainger49]

Wow, I Googled it and the first link was to a PDF file.  1/2 a Tera bite.  I saved one.

Thanks PJ!