Odd/Crazy output transformer design? - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 13th July 2010, 08:27 PM   #1
Matt BH is offline Matt BH  United Kingdom
diyAudio Member
 
Join Date: Jul 2009
Location: Cheltenham
Default Odd/Crazy output transformer design?

Hello,
I have seen these trannys wound by John Lenard Burnett:

Lenard Audio - Education - Valve Amps

Scroll down to the Pi wound ones.

I have had an idea based on the already existing planar transformers used in switch mode supplies but using pair of iron c-cores.

If I had a load of flexible PCBs made with the windings on either side connected with vias and insulated either side then stacked up P-S-P-S etc. Then connected to form the whole winding for P and S.

This should make a very high bandwidth tranny with almost ideal coupling that could be configured for all sorts eg. UL, CFB etc. The insulation of flexible pcbs is capable of 3KV continuous so no probs with insulation. interlayer capacitance could be problem though. Thicker insulation?

What do you guys think? Is it feasable? I know it would be expensive for one or two but for lots it would be minimal. The transformers could be considered identical due to the high tolerance of production and the balance for a push pull tranny would be perfect.

A bit like this:

http://archive.electronicdesign.com/.../figure_02.gif

Cheers Matt.
  Reply With Quote
Old 13th July 2010, 09:43 PM   #2
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
I believe I have seen a patent using the same idea. Try a patent search online for audio output transformer. I'm not aware of anyone using this to construct transformers though. Patent might be expired too. Seems likely that you would end up with too much insulation and not enough copper in such a setup, and an enormous connectivity problem between all the layers.

The ultimate output transformer is also quite easy too make using a toroid winder machine with progressive wind capability. No interleaving required, no magic or mumbo-jumbo OT voodoo, simple as making a peanut butter sandwich. You can set the machine up (typically programmed control), and return when its all done. No labor. Just a single progressive wind layer for each winding across the full core with good insulation between windings. (well, one does have to change the wire spools and insulation spools on the winder between the windings) If you are concerned about DC balance with a toroid core, all that is required is a thin film coating on the steel lamination before its wound up into the toroid form to make a distributed gap. (the flux has to cross between lamination layers)
A SE design can be made using a thicker separator film.

All those audio OT winders out there just have the wrong stuff. And they're stuck with paying it off. But make sure you don't buy a toroid OT wound using random wind technique. Those old winder machines were used for 60 Hz power xfmrs, and will make a rotten audio OT.
__________________
I want a Huvr-Board!!

Last edited by smoking-amp; 13th July 2010 at 10:01 PM.
  Reply With Quote
Old 13th July 2010, 10:44 PM   #3
Matt BH is offline Matt BH  United Kingdom
diyAudio Member
 
Join Date: Jul 2009
Location: Cheltenham
Thanks for the great reply Smoking Amp.

I did a patent search and cant find any use in audio. You may have better access in the US for this.

I completely agree that connecting each layer is a pain. I have worked a way around this in design only

You would be supprised on the amount of copper that can be stuffed in. I have done a fair amount of research before posting ie. finding out whats available and designing layers etc.

I reckon and this is by no means spot on that approx 10-15% more copper can be used compared to a normal output tranny. Of course this is based on the usual 3mm margins to the core. I started with a track width rated for 300mA in a six layer board ie. buried. My first design barely filled 40% of the window and this was derated to account for the current etc. This is for a 3K3 ppp 807 type design.

Torroid output trannys are not ideal no matter how they are wound. Sorry to be pessimistic about this but its true. I have heard/read rumors of a distributed gap in the Plitron/Van Der Veen types but I dont believe it. Perhaps George could give comments on his 400W trannys and their ability to take out of balance tube currents.

I have also used a pair of admittedly mains torroids in a 6080 amp very much along the lines of Shoog. I found them to be very perculier about DC offset in PP. With CCS forcing class A they sound great. For SE forget it. Apparently there are makers of SE torroid transformers, perhaps they have some crazy LASER that can cut it so fine

Not that I care too much for SE. A pair of GM-70 would be OK by me

My idea is to address all these problems. With a C or EI core it is rellatively easy to arange a gap. This has advantages in PP as well as SE. Especially with the more modern cores where a lowering of permiability is not as bad as with a non GOSS core. Clearly an EI core cant take advantage of this but the simplicity in production may well win it.

You hint at insulation and capacitive problems. This is the only downfall I can see. And as the first design, even though it was derated only filled 40% I dont see it as a problem.

Please guys especially Bud. P. Come in on this as I need to be shot down in flames

Cheers Matt.
  Reply With Quote
Old 13th July 2010, 11:39 PM   #4
diyAudio Member
 
Join Date: Nov 2005
I don't think you can create a distributed gap by spacing a closed tape core. The physics of it doesn't make sense, and if it worked you'd probably see it a lot, which I personally haven't. Got equations?

I'm finding it hard to imagine getting enough turns on a planar transformer to be useful with anything but something like a bunch of 6C33C or some inverted grid experiement.
  Reply With Quote
Old 14th July 2010, 12:56 AM   #5
diyAudio Member
 
smoking-amp's Avatar
 
Join Date: Dec 2001
Location: Hickory, NC
For a tape wound core, the flux has to jump from layer to layer to form closed loops. The gap then consists of the spacing between lam layers, but with the whole turn as length by the tape width (very large gap area). The whole turn acts as a lot of conventional gap in parallel (just like resistors in parallel), so its hard to get much equivalent gap without putting in a fair amount of spacer. This can cause your steel cross-sectional area to suffer with all that spacer in there (so requiring more turns or a bigger core). So its more effective to just cut a gap in the core in one place (a saw) for SE cores where a larger gap is required. Its always done that way for SE type cores for practical purposes.

However, the distributed gap approach does not suffer from the field radiating out from a localized gap, making it less of a shielding problem. And P-P does not require much spacer to bring it in line (effective Mu) with the butt joint gaps of E-I. The formulas are quite simple: effective gap = spacer width * ( tape cross section area)/(turn length*tape width) You can see that the last (...)/(...) terms are a very small fraction, leading to the high effective permeability of non spacered toroids.

In any case, I don't think that Plitron has any effective gap in their P-P cores unless they have come out with something new. (A cut gap would be too large for P-P fixup) Without some gap, you have to contend with the critical DC balance.

It has also been remarked that the very large inductance of typical ungapped toroid P-P OTs makes for a lack of tube sound due to the lack of magnetizing current distortion at low frequencies. A distributed gap would fix this. (Well, assuming someone actually wanted to "fix" that "problem". Some, maybe many, would argue that large inductance and low distortion are virtues)
------------------------------------------------
A planar printed circuit transformer would need some vacuum impregnation with epoxy or something to improve the HV insulation properties for tube usage.

OH, I think I have also seen a patent on winding multiple windings on an xfmr using printed flat ribbon cable the width of the bobbin, just coiled up.They just interconnect the wire ends to get the turns ratio. Would have a lot of distributed capacitance maybe if they tried any sectioning/interleaving within the ribbon.
__________________
I want a Huvr-Board!!

Last edited by smoking-amp; 14th July 2010 at 01:08 AM.
  Reply With Quote
Old 14th July 2010, 01:31 AM   #6
diyAudio Member
 
Join Date: Nov 2005
I get the picture with the toroids, although it seems you'd double core cross sectional area before you'd begin to have a practical gap because of the extreme area. Probably not seen because the effects on wire requirement shut down the advantage.
  Reply With Quote
Old 14th July 2010, 05:31 AM   #7
JPS is offline JPS  India
diyAudio Member
 
JPS's Avatar
 
Join Date: Sep 2005
Location: New Delhi, India
Quote:
I have had an idea based on the already existing planar transformers used in switch mode supplies but using pair of iron c-cores.

If I had a load of flexible PCBs made with the windings on either side connected with vias and insulated either side then stacked up P-S-P-S etc. Then connected to form the whole winding for P and S.
I think it is possible to make OPT using same technique as used in Planner transformers. You are right you have to make a thick stack of flex PCBs and EI or C core made of silicon steel can be used in same way the ferrite core is used in smaller planner transformer which are made for SMPS.
regards
  Reply With Quote
Old 14th July 2010, 05:33 AM   #8
JPS is offline JPS  India
diyAudio Member
 
JPS's Avatar
 
Join Date: Sep 2005
Location: New Delhi, India
Quote:
Originally Posted by Matt.B.H. View Post
Hello,

I have had an idea based on the already existing planar transformers used in switch mode supplies but using pair of iron c-cores.

If I had a load of flexible PCBs made with the windings on either side connected with vias and insulated either side then stacked up P-S-P-S etc. Then connected to form the whole winding for P and S.
Cheers Matt.
Your idea is good I think it is possible to make OPT using same technique as used in Planner transformers. You are right you have to make a thick stack of flex PCBs and EI or C core made of silicon steel can be used in same way the ferrite core is used in smaller planner transformer which are made for SMPS.
regards
  Reply With Quote
Old 14th July 2010, 09:06 PM   #9
diyAudio Member
 
Join Date: Aug 2008
Quote:
Originally Posted by smoking-amp View Post
The ultimate output transformer is also quite easy too make using a toroid winder machine with progressive wind capability. No interleaving required, no magic or mumbo-jumbo OT voodoo, simple as making a peanut butter sandwich.
Unfortunately, interleaving IS required, albeit not so much interleaved sections between primary/secondary required (compared to EI/double C core).

Currently I have 2 over-sized 70W audio toroids with leakage inductance only 17 mH.

Additionally, toroid must have unusually large internal diameter in order to accommodate several layers of primary/secondary connected in parallel (on main, secondary is wound with wire d = 1.4 mm).
  Reply With Quote
Old 16th July 2010, 06:03 PM   #10
Matt BH is offline Matt BH  United Kingdom
diyAudio Member
 
Join Date: Jul 2009
Location: Cheltenham
Sorry guys been working away again

This seems to be going the way of toroids which was not my intention, merely a reply to Smoking Amp.

Yes torroids work but that was not the point of my OP.

Quote:
A planar printed circuit transformer would need some vacuum impregnation with epoxy or something to improve the HV insulation properties for tube usage
I agree that each layer would need to be stuck together to stop any audible ringing. The HV qualitys of flexible PCB are such that 1 though (your mill) will withstand 3KV.

Cheers Matt.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
PP Output Transformer design Hi_Q Tubes / Valves 12 28th October 2009 02:23 PM
Output transformer design for SS amp Tyimo Tubes / Valves 12 17th January 2007 01:02 PM
odd transformer design godfather_cz Parts 7 3rd May 2003 10:32 AM
Crazy transformer? Lisandro_P Everything Else 4 5th January 2003 04:54 PM
Output transformer makes odd whining noise tr3 Tubes / Valves 5 23rd June 2002 07:11 PM


New To Site? Need Help?

All times are GMT. The time now is 09:28 AM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2