Transformer Specifications

General discussion area for tube amps.

Moderators: pompeiisneaks, Colossal

User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Transformer Specifications

Post by roberto »

Hi all,
I'm collecting some info about power and output transformers on different amps because I've found a trafo maker near me and this will let me save some money so try different configurations and find what i really need for my amps.

By now I've found classical Dagnall specifications:

Dagnall C1998 38x38mm on EI M6 core:
- 16-0 ohm secondary (parallel): 120 turns x 0.76mm AWG21
- primary: 2 x 620 turns on eight layers x 0.45mm AWG26
- 8-16 ohm secondary: 35 turns x 1.18mm AWG18
- 4-8 ohm secondary: 25 turns x 1.18mm AWG18
- 0-4 ohm secondary: 60 turns x 1.60mm AWG15

Dagnall C2668 on 38x40mm on EI M6 core (500-650pF primary to secondary capacitance):
- 0-4 ohm secondary: 60 turns x 1.76mm AWG13
- 1/2 primary: 620 turns on four layers x 0.45mm AWG26
- 4-8-16 ohm secondary: 25+35 turns x 1.18mm AWG18
- 1/2 primary: 620 turns on four layers x 0.45mm AWG26
- 0-16 ohm parallel secondary: 120 turns on two layers x 0.75mm AWG21

Dagnall C3070 32x60mm on EI M6 core:
- 1/4 primary: 310 turns on two layers x 0.45mm AWG26 (B+ to half A2)
- 0-2 ohm secondary: 40 turns x 1.76mm AWG13
- 1/2 primary: 620 turns on four layers x 0.45mm AWG26 (B+ to A1)
- 2-4 ohm secondary: 20 turns x 1.76mm AWG13
- 4-8 ohm secondary: 25 turns x 1.18mm AWG18
- 8-16 ohm secondary: 35 turns x 1.18mm AWG18
- 1/4 primary: 310 turns on two layers x 0.45mm AWG26 (half A2 to A2)


Soldano SLO100 OT is similar to C3070, except is 32x50mm and has an higher primary impedance.


Now about Trainwreck specs, where I need more hints:
http://ampgarage.com/forum/download/file.php?id=10100

Moose wrote:The vintage spec is identical to the Pacific in Glenn's amp -- on M19/26 steel, same lams, same wind, same tap, etc. The difference is noticable in the feel of the amp, especially in the ease of sustain, as well as in the midrange focus and distortion. It's smoother vs. the crunchier distortion on the original M27 version.
Moose wrote:I did a touch of experimenting and that's close to the difference in the 5200 and 6600 ohm taps. Even with the M27 steel, that was the difference with the M19 giving you more bass whump at the attack, but the higher impedance smoothing things a touch. In M19 you can get some of that bassier sound of the stancor by using a liverpool tranny -- it's not so much MORE bass, but there is a change in character on the low end. Hard to describe as I didn't A/B, I actually swapped trannies on my personal amp.
Moose wrote:M6 is not the best steel if you want to chase Franchesca tone. I've got a stancor style in M6 here and it's clear, sounds bitchen with KT66, but the El34 sound is too sizzly and it doesn't thicken up when pushed.
M6 is a grain-aligned steel that is particularly good at resisting saturation, which makes it a technically great steel for audio transformers -- if you're a hi-fi engineer who is looking for clean. To get better saturation performance than M6 -- i.e. more flux with less time in saturation when overdriven, meaning more linear phase and frequency response and less harmonic distortion -- you have to actually go to much more expensive materials like nickle.
However, you can't get the Franchesca tone without the saturation and accompanying alinear response. So, if you're cloning an express circuit, M6 is bad.
VacuumVoodoo wrote:P=50W, FPBW: 70-12000kHz, Raa =8k, ZL= 4,8,16 Ohm

If you've done 15% of your homework you will also specify max primary IDC. Another 15% of homework will tell you to specify what power tubes will be used and their operating conditions i.e B+ in addition to I(bias)
Next 10% of homework will teach you to specify how windings should be interleaved, choose core material, what diameters of winding wire to use, how many turns per section etc. If I understand correctly what you're saying you already know this.

The remaining 60% you'll have to "discover" on your own. Hint: just as you can draw load lines for tubes and graphically analyze distortion characteristics of a particular gain stage or push-pull pair so can you draw load lines for magnetic circuit i.e. a transformer on the set of hysteresis curves for the core you intend to use and analyze distortion characteristics in analogous way.

When you will have done all this and discover how to match tube characteristics to magnetic properties of core material to design a transformer that together with your chosen power stage design will produce required distortions you will be able to give a manufacturer very exact and detailed specifications. 95% of manufacturers will then say you don't know diddly squat about transformer design.
http://www.mif.pg.gda.pl/homepages/tom/transfor.htm
http://www.cfd.tu-berlin.de/~panek/cykin/cykin1.zip
http://www.cfd.tu-berlin.de/~panek/cykin/cykin2.zip
http://www.cfd.tu-berlin.de/~panek/cykin/cykin3.zip
VacuumVoodoo wrote:You should be more concerned with laminations and isolating lacquer thickness, carbon and silicon content, cold vs. hot rolled steel, post roll heat annealing
User avatar
Colossal
Posts: 5038
Joined: Sat Oct 20, 2007 9:04 pm
Location: Moving through Kashmir

Re: Transformer Specifications

Post by Colossal »

Very cool writeup Roberto. Thanks for posting that.
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

Hope it helps also other people, and that someone can help me to complete the specs for trainwreck OTs and PT as well.

I've some doubt on C3070 too. It could have different layers and secondary should be 1.5mm from 0 to 16 ohm.
It should have 820 turns on primary instead of 1240, and 80 turns of secondary instead of 120. This is because the core is 32x60 and this way it has the same induction than 38x38 with 1240 turns.
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

Some more specs for C2668
You do not have the required permissions to view the files attached to this post.
User avatar
VacuumVoodoo
Posts: 924
Joined: Fri Feb 17, 2006 6:27 pm
Location: Goteborg, Sweden
Contact:

Re: Transformer Specifications

Post by VacuumVoodoo »

Very "interesting" winding data for the Dagnalls. Are these winding figures taken from factory specs or reverse engineered?
When calculating number of turns for a given impedance ratio one must factor in transformers power transfer efficiency (95% is pretty common) and voltage loss on primary winding resistance.
These corrections usually are made by slightly increasing secondary (or reducing primary) number of turns by a couple % over calculated number based on "ideal loss-free" transformer assumption.
It seems to me Dagnall calculated one set of "generic" windings for a mathematically ideal transformer and used the same number of turns on different core sizes.
This gives different primary inductance for each core size and different low end cut off frequencies.
IMO a rather poor job, hardly a "sound design" - but what the heck, it's for a guitar amp so who would care? But that's where 2-3 turns out of 120 can make a significant difference.
Material specs regarding layering and isolation etc look ok except too low isolation and HiPot test voltage.
Last edited by VacuumVoodoo on Wed Sep 19, 2012 1:01 pm, edited 1 time in total.
Aleksander Niemand
------------------------
Life's a party but you get invited only once...
affiliation:TUBEWONDER AMPS
Zagray!-review
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

Hi Aleksander and thank you,
two are the men I was weayting, you and Larry Novosibir.
Got it, I've to divide the number of secondary windings by the efficiency of the trafo: EG 120/0.95= 126 windings.

Specs are obtained from forum digging, net searching and some retroengineering. C3070 should be ok now, and also C1998 (820 primary and 80 secondary).

About the iron, I've obtained some basic general uningegneristic concepts:
- M6 works like 6L6GC, stays cleaner up to a point where it starts to add more harmonics (like more NFB)
- M27 works like EL34, more vintage oriented, cruncier, better on saturation (like less NFB)
- M19 and M26 works like EL84, smoother and adds a bit more of a bassy

Do you have some more hints to work on?
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

I think someone can find it useful:
http://www.dissident-audio.com/AutoInde ... os/Papers/
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

BudP wrote:The difference in sonic’s is interesting. An amorphous core OPT for PP will be very clear, have great transient response and excellent tonality. However, they are somewhat one dimensional in that most of the fine gradient detail that provides internal note structure and makes transient noises into musical events, has been stripped away by the lack of E Field coupling and the higher dielectric constant materials used in the coils, essentially half of the information must be lost to avoid frequency response peaks.

An M3 commercial core, properly constructed for audio purposes, will sound slightly softer than an Amorphous core transformer. There is after all more distortion, and when comparing 99.997% correct signals (Amorphous) with 99.993% correct signals (M3), this is what you should expect. However, there can be a great deal more internal information made available, for all musical signals, because you do not have to throw away half of the signal and the antenna event is actually encouraged during it's E Field portion.

In addition the M3 and M6 core are basically only providing a ferrous bounding box for the coil antenna event above 400 Hz. The material still responds but the losses are so high, with extended frequencies, that it neither adds nor subtracts to the signal passing from coil to coil, so long as that core construction issue has been dealt with. Neither amorphous core not M3 core is "better", they have differing strengths, and when both are used properly either will provide superb audio, within the limitations found above, and their pricing will be about the same too. An M6 core unit, with more distortion and only slightly less information, without quite the refinement of either of the other materials, is a much less expensive alternative, with very few drawbacks.
http://www.diyaudio.com/forums/tubes-va ... ost1915988
User avatar
VacuumVoodoo
Posts: 924
Joined: Fri Feb 17, 2006 6:27 pm
Location: Goteborg, Sweden
Contact:

Re: Transformer Specifications

Post by VacuumVoodoo »

First 2 paragraphs read like something I could have written after having over dosed on peyote. Great marketing BS and I'm sure it helps his sales Aside from it, Onetics transformers are pretty decent but no better or worse than any other decent products.
Try to find my dissertation on using paper soaked in fish liver oil for interlayer isolation in output transformers, it's a good one.
There are some serious knowledgeable guys on dyiaudio but also a large number technically illiterate hard core audiophools.
Aleksander Niemand
------------------------
Life's a party but you get invited only once...
affiliation:TUBEWONDER AMPS
Zagray!-review
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

:lol:
As to paper vs. plastic bobins etc. In my D.I.Y. output transformer days I came up with a recipe for great sounding OT. I cannibalized PTs from russian TV sets for cores, made my own bobbins from cardboard and used baking paper soaked in fish liver oil rich with Omega-3 fats for interwinding & interlayer isolation. The sound this resulted in was heavy, fat and very aromatic. That last attribute was not always appreciated by the audience but somehow the fish & chips street vendors near the venue often reported significantly increased sales after gigs where amps with my OTs were used. Now you know how the band name Country Joe & the Fish came to.
http://www.powerscaling.com/phpbb3/view ... 9152#p9148

PS
I've almosf finished the book found on the previous link, some serious question may appear into my mind, then promptly FWD to you.
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

I've not "worked" on this topic for a few years, but now I would like to build a 450 W Bass amp with an octet of GU50 with plates at 850-900 V and screens at 250-300 V.

These are the specifications of the PT, I already asked to have it toroidal:
320V 0,1A Preamp+PI
210V 1A Ub Rail 3
210V 1A Ub Rail 2
210V 1,2A Ub Rail 1 + G2 supply
24V 0,2A MosFET Stage supply
200V 0,1A Bias and PI CCS suppy
6,3-0-6,3V 7A Heater GU50
6,3-0-6,3V 3A Heater Preamp +PI

While for the OT I would like to go toroidal as well. What I would like to try is a UL configuration with CFB.
Due to the limitations of the screens, it would need a separate winding for it (to be supplied at 250-300V instead of the 850-900V of the plates). I would like to try a CFB configuration as well, with around 13% of feedback from it.

The toroidal transformers maker I have found near my home has no audio skills, so needs some guidelines from me.
I've seen that the optimal load for an octet of GU50 at 850V B+ and 250V screens is 2k. Secondaries will be 4 and 8 Ohms.

So let's suppose the following secondaries:
294 turns for 8 Ohm tap
208 turns for 4 Ohm tap

The primary should have 4656 turns, divided this way:
4050 turns for the plates (87% of 4656 turns) for 800mA circa
606 turns for the cathodes (13% of 4656 turns) for 800mA circa
1396 turns for the screens (30% of 4656 turns) for 100mA circa (from the datasheet I see Ig is Ia/10)

This is theoretically, then it comes reality and the ratios need to be altered a bit. But I'll open a thread only for it.

My main question is: how to interleave the transformer in case we have three separate primaries with different purposes like here? I would like how this basic design can be improved (p = plates; s = secondary; g = screens; k = cathodes):

1/3 p 1350t
1/4 s 74t
1/2 g 698t
1/2 k 303t
1/4 s 74t
1/3 p 1350t
1/4 s 74t
1/2 k 303t
1/2 g 698t
1/4 s 74t
1/3 p 1350t

Thank you very much for your help.
User avatar
JazzGuitarGimp
Posts: 2355
Joined: Mon Jul 23, 2012 4:54 pm
Location: Northern CA

Re: Transformer Specifications

Post by JazzGuitarGimp »

Transformer nuts and bolts stuff. I wish I could help, but I have no useful knowledge. But I am looking very forward to reading along. Thank you, roberto.

Lou
Lou Rossi Designs
Printed Circuit Design & Layout,
and Schematic Capture
User avatar
xtian
Posts: 6990
Joined: Mon Apr 19, 2010 8:15 pm
Location: NorCal
Contact:

Re: Transformer Specifications

Post by xtian »

Hopefully you can get the attention of JM Fahey.
I build and repair tube amps. http://amps.monkeymatic.com
User avatar
roberto
Posts: 1841
Joined: Tue Nov 27, 2007 4:45 pm
Location: Italy

Re: Transformer Specifications

Post by roberto »

Thank you Lou and Xtian, let's see if there will be some feedback.
Post Reply