|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
141.239.172.61
and by that i mean there is a lot of interest outside of the usual hard core audio guys. I was wondering about the output transformer for single ended parafeed operation. I have devoted quite a lot of money for gapped single ended trans and if i wanted to build a para se amp, do i need to get some different opts. Is it possible to utilize pp opts and simply not use the center tap? thank you for your insight. Dak
Edits: 11/21/16Follow Ups:
Push-pull OPTs will work in parafeed, but usually not that well. I have done this myself, and heard a few that others have made. The main issue in my limited experience is high frequency resonances due to the distribution of capacitances. P-P transformers are designed in the expectation that the center tap will be at AC ground, whereas SET transformers assume the B+ input is at AC ground. Kind of like putting car tires on motorcycles.
Gapped transformers have much less inductance than the same windings with interleaved core laminations, but the interleaved-core inductance does vary with level and frequency. The lower inductance may limit the bass extension, but you can do a more precise calculation with a known constant inductance, and still see some of the parafeed advantages over series feed. Used in parallel feed, an airgapped transformer can handle 5-10 times as much power before running into saturation problems (low frequencies at high levels).
"The main issue in my limited experience is high frequency resonances due to the distribution of capacitances. P-P transformers are designed in the expectation that the center tap will be at AC ground"I would truly like to understand this.
A question or two if you please.
WRT the capacitance that is expecting the primary CT to be at AC ground,
is it,
primary winding to secondary winding capacitance?
primary to core/case capacitance?
primary winding to itself capacitance?
or something else?
Thank you Paul.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 11/22/16
I was afraid someone would ask that :^) It's hard to explain, which is why I didn't.
I am talking about the capacitances in the primary winding. The primary is usually in sections which are in series, interleaved with secondary windings which are basically at AC ground. Each primary section has its own self-capacitance, and a capacitance between it and each of the adjacent secondary windings and/or core. The inter-winding capacitance is adjusted by the thickness and permittivity of the dielectric material between them. A thicker insulation reduces the inter-winding capacitance, but increases the leakage inductance.
If you draw out the circuit, say for four primary winding sections (not uncommon in high-end P-P transformers), you will see that it's a bear to analyze. You do not need a thick insulator adjacent to the center tap, since it is already at AC ground, but that same insulator has serious requirements if the AC ground is moved to one end.
I once worked on a project with Magnequest Mike, in the course of which I have seen the drawings (which Mike now owns) for one of the best golden-age P-P transformers. Each insulation segment has a different material and a different thickness, specifically to manage the high frequency leakage/parasitic capacitances to avoid HF resonances. High frequency resonances in the primary winding wreak havoc with global feedback, so golden-age P-P transformers had to take these issues seriously.
For another example, there was some years ago a popular large SE transformer from another manufacturer which did not manage these capacitances well; it had a serious problem in the high frequencies and was fairly quickly redesigned.
Hope that helps! There is a useful section on transformer design in the RDH4, and Reuben Lee's "Electronic Transformers and Circuits" is a classic reference. The first edition has some information on parallel feed, which was abandoned in the second edition. FWIW.
Paul you said "I once worked on a project with Magnequest Mike, in the course of which I have seen the drawings (which Mike now owns) for one of the best golden-age P-P transformers. Each insulation segment has a different material and a different thickness, specifically to manage the high frequency leakage/parasitic capacitances to avoid HF resonances. High frequency resonances in the primary winding wreak havoc with global feedback, so golden-age P-P transformers had to take these issues seriously."
Does this mean that if an output trans like an Acro TO 450 is used as a SE Parafeed OPT the HF resonances are NOT a problem?
"Does this mean that if an output trans like an Acro TO 450 is used as a SE Parafeed OPT the HF resonances are NOT a problem?"No, it doesn't mean that at all.
The transformer is designed and the problems are dealt with but it is assumed that the CT of the primary will be connected to a AC ground point. (the last cap in the power supply is a AC ground point)
When used as intended the transformer's design deals with the problems.
When not used as designed those problems will show up again.
The above is my understanding of what Paul said.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Edits: 11/24/16
Excellent post. A keeper.
Builder of MagneQuest & Peerless transformers since 1989
"The primary is usually in sections which are in series, interleaved with secondary windings which are basically at AC ground."
So should the low end of the secondary be grounded for the reason that the transformer was designed with that in mind? (apart from safety concerns)
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
It is hard to predict the behavior of a component when used outside the expectations of the designer. No general rule is possible.
I do not recommend ignoring safety concerns, and - in my humble opinion - no competent designer would ignore them.
Years ago, Mike made mention of someday rewinding a particular 'Q' series transformer (can no longer recall the number...268?/265?).
This transformer was so difficult to wind he said that if he pulled it off, that would be his Swan song.
Was the transformer in question the blueprint you were looking at?
Thank you for sharing that superb explanation, Paul.
The Mind has No Firewall~ U.S. Army War College.
That would be the Peerless S-268-Q. Indeed, it is a complex transformer to make. The test sheet for this blueprint shows that it was able to pass a clean 30KHZ squarewave. I'll have to pull the blueprint to cite some of the other cool features of this design.I'd still like to make it my swan song! Other than the Peerless TL-404 autoformer and some of the very early Freed small signal transformers... the 268 would be my most challenging undertaking. Peerless only ran (by memory) the 268 in two very short runs due to the difficulty of their manufacture.
MSL
Builder of MagneQuest & Peerless transformers since 1989
Edits: 11/23/16 11/23/16 11/23/16 11/23/16
Likely the 268...the 265 is relatively simple. The 271( which is more closely related to the 268) is substantially more complex than the 265...:)
cheers,
Douglas
Friend, I would not hurt thee for the world...but thou art standing where I am about to shoot.
I once built what I thought would be a simple plate choke for a high power transmitter covering 3.5 - 30 MHz. This was a single layer choke, probably less than 100 turns, and not only did I section the winding, but the core was non-magnetic. It took three tries to eliminate all the series RF resonances. Wind a tranformer with a primary and secondary and thousands of turns? Not me!
--------------------------
Buy Chinese. Bury freedom.
.
Have Fun and Enjoy the Music
"Still Working the Problem"
I have purchased several pairs of SE OPTs for amp projects and they are all gapped. So, i was wondering about their suitability for a parafeed circuit. It was not clear from the schematics that i read about the specs for the output trans. If the se opt has the proper primary impedance. regards, Dak
The gap is to avoid core saturation from DC. If you don't put DC through the core (parafeed) there will be no saturation to avoid. The downside is that gapped SE transformers have to be comparatively bigger with more turns and thus more expense compared to a non-gapped opt designed for SE parafeed for the same power output.
Yes, i know what the gap in an SE opt is for. But, i wondered if i could use the opt that i have on hand which are SE and PP instead of shelling out more cash for a specifically designed parafeed opt. Which of course, i would rather avoid. tyvm
Did you read my subject line?
Might I suggest that you take one of your gaped SE xfmrs and restack them without a gap (solid core). The inductance will go way up and they will work well as PF xfmrs.
It seems that when they varnish the unit it glues all the lams together. Also i spent a lot of money initially and don't want to ruin them. Maybe i will play with some cheap el84 unit first. thank you for suggesting.
Pulling lams out of a finished transformer which has been waxed or varnished is, as you've suggested, a very, very difficult undertaking.
Almost always you will not get back in the same number of lams as you've yanked out when stacking alternately (say 1x1).
If you bend or tweak the lam in getting it "unglued" and pulled out... it should go in the trash can. Lams (especially high perm materials) are very strain sensitive and their magnetic properties subject to degradation if bent or tweaked.
If you have a nice SE airgapped output... keep it... don't monkey with it.
For many projects you can get nice parafeeds for not a bankroll of money.
MSL
Builder of MagneQuest & Peerless transformers since 1989
Just as i suspected. In the past i have tried to "straighten out" a power trans which had crooked lams because it must have been dropped. I could not get to straighten out the lams. And i really tried hard. I won't describe the process but decidedly caveman style. Even after that i am using the trans and it works great but there is a little buzz when the current use is high.
.
are good places for parafeed info.
Paul Joppa is the man
I'm listening to SE 45 active parafeed amp with Altec 70V line OPT's tonight.
You can pick them up on EBay from time to time.
You have some homework to do. Spend your time away from the forum and read read read
Plenty of information out there including using torriod power transformer for output!!! NO DC = many options
You can use PP O/P "iron", as no "standing" DC is present.
IIRC, MagneQuest uses the same coil in an air gapped model and a parafeed model. The parafeed model is rated for higher power.
Eli D.
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: