Toroidal I/P & O/P Transformer Design Questions

[Kid Squid]

Hwyl Lads,

Right , got a question, with regard to designing Toroidal I/P and O/P transformers.

How do you do it ?? :wink:

Seriously tho, one of my mates, works in a toroid transformer factory, and I'd love him to make a few transformers of the 600 ohm to 10K c/t , and 10Kc/t to 600 ohm types..

He has no real idea, of what i'm on about, so i thought, i'd ask .

I've been reading up on designing transformers, but the E/I type.

Also, just found this, http://www.plitron.com/audio_se.asp

Toroidal SE OPT's , how are they doing that ??, Obviously, with an E/I core type, you just 'gap' the laminations, to suit the DC current,
how are they doing this with a toroid - do they just make the core larger, than it needs to be - so that it saturates at a higher level ??
Questions, questions.......
:grin:

Steve :thumb:

 

[Kid Squid]

Just found these articles as well - looks like some good reading... :wink:

http://www.plitron.com/audio_articles.asp

http://www.rfcafe.com/references/design_data/transformer_design_data.htm

http://www.amplimo.nl/

Steve :thumb:

 

[rotation]

Hi, a few years ago i wanted to try the same thing because i had a book (Modern High-end Valve Amplifiers: Based on Toroidal Output Transformers) publihsed in relation to Plitron, where some theory and applications for this transformers were described. Book didn't tell much except some math (not helpfull, most important things for diy missing) and apps, when i searched google i couldn't find anything either. It looks like only Plitron makes this kind of transformers. I guess there must be a good reason that production of audio toroids is so limited.
I find this idea very interesting, so if you will be able to find anything more please report here.

Miha

 

[zebra50]

FWIW, Oktava used toroidal o/p trannies in some of their older ribbon mics, (ML11 for certain). And those mics sound very good.

 

[zebra50]

http://microphonium.blogspot.com/2006/08/more-on-oktava-transformers.html

http://microphonium.blogspot.com/2006/08/many-oktava-transformers.html

 

[BBob]

Yes, these are very efficient. Hats off to the engineer who got such high output! I did not think they contributed much to the sound quality and were fairly neutral. We once had piles of them available but they were all given away or who knows what was done with them.

Bob

http://microphonium.blogspot.com :sam:

 

[emrr]

Western Electric made a number of toroidal audio transformers, starting in the '30's for sure, and maybe even earlier.

 

[PRR]

> one of my mates, works in a toroid transformer factory, and I'd love him to make a few transformers of the 600 ohm to 10K c/t , and 10Kc/t to 600 ohm types.

10K:600 is 4:1 voltage ratio.

At the bottom of the audio band, a "230VAC 50Hz" winding is ~~about~~ several-K impedance and will stand about 60V at low distortion.

So your first step is to take a stock known-build 230V:57V (or anything close) power transformer, and measure the audio performance. You want a test rig with signal generator, source resistance, transformer link, load resistance, AC voltmeter; and you probably want an "actual use situation" to judge "sound".

I suspect that a stock 230:48 20VA PT with 10K source and infinite load will be -3dB near 40Hz, which may be acceptable, but we usually like the -3dB near 20Hz. If loaded in 600 ohms and wound "for line safety", secondary well separated from primary, it could be drooping by 10KHz due to stray inductance. Or it could be much better. Or the 10K side could be droopong due to stray capacitance.

To improve the bass, double the number of turns. You could simply say "460V:110V 50Hz". Double the turns in the same space means half the wire area, so 4 times the resistance. Resistance, for the audio application, was probably "very low" before, now it may be highish. This can be checked quickly: an ohm meter reading should be around 5%-10% of the nominal audio impedance. The "600" side should be 30 to 60 ohms. Since we don't do Power Transfer no more, higher could be acceptable.

If stray inductance droops the highs, get the secondary more intimate with the primary. Wall-power regulations force large separation and 1,000+ test voltage, you don't need that.

This tranny will be larger than an optimized audio tranny. You can try a smaller core. But this won't save a lot of money, and will increase copper resistance.

 

[zebra50]

[quote author="BBob"]Yes, these are very efficient. Hats off to the engineer who got such high output! I did not think they contributed much to the sound quality and were fairly neutral.[/quote]

Hi Bob - nice to see you here & good to speak to you this week - looking forward to those green rossies! :wink:

Octava have been using toroids for some decades now. There are a couple of pictures of the ML11 transformer and some discussion of its operation HERE

Stewart

 

[BBob]

Aha!


Soon you shall have some very rare mics indeed. I rather think they are greenish blue. They certainly do have green in there.

The LC at the output of the old Oktava - know the values they used? I have tried that in the past to shape the output, but I got rather wide variations from piece to piece, cap tolerances being what they are.

Bob Crowley


http://microphonium.blogspot.com

 

[gyraf]

The nice small Beyer input transformers are toroidal also (at least the few I've seen taken apart)

For air-gapping you might consider "B-grade" core material - made from scrap strips that were left over from making "A-grade" (continuous steel band core) transformers. Many cheap Chinese power toroids are supposedly B-grade core, my local winder tells me - and can accept larger primary imbalances and DC-currents without misbehaving.

Downside is lower and somewhat unpredictable Al-value, so you must use more turns in your windings to be safe with inductance..

Jakob E.

 

[zebra50]

[quote author="BBob"]
The LC at the output of the old Oktava - know the values they used? I have tried that in the past to shape the output, but I got rather wide variations from piece to piece, cap tolerances being what they are.
[/quote]

Hi Bob,

This is the sketch I made when it arrived:

I should probably replace the electro, although I did some great vocals with it at the weekend.

Stewart

 

[rodabod]

[quote author="zebra50"]looking forward to those green rossies! :wink:[/quote]

Stop making me jealous! And wondering exactly what you are talking about at the same time.

I don't think the Oktava filter idea is so bad. It could possibly be quite handy on some of my darker ribbons. Fair point about parts tolerances, although with a wide enough boost it might not matter too much. Funny how Oktava seem to like to boost the HF on almost everything they do, whether it be treble resonators on their condenser and ribbon mics, or filters on the outputs.

 

[BBob]

Oktava do seem to have treble boost on their minds.

Fostex had two wonderful coin like treble discs (which you cannot see, but they are there) on each side of their fig 8 element on the M-S mic to give it the general tone of the cardioid element. The similarity in appearance to a steam cylinder is amusing.

http://microphonium.blogspot.com/2006/06/fostex-mid-side-stereo-mic.html

Bob Crowley

 

[zebra50]

[quote author="rodabod"]I don't think the Oktava filter idea is so bad. It could possibly be quite handy on some of my darker ribbons. Fair point about parts tolerances, although with a wide enough boost it might not matter too much.[/quote]

It does seem to work. I have quite a few old ribbons but the ML11 is the only one that gets any serious use. (I've ordered a pair of C&T Roswellites from Bob so all that will change very soon!)

The filter is in series, so it could be fun to build a little in-line XLR box with a suitable inductor and a range of cap values on a switch. Might be a very useful little box to have around.

I should probably replace the electro, although I did some great vocals with it at the weekend

Looking at the Oktava pic again, I think it is one of those old Soviet film caps. Probably why I didn't replace it.

Are we getting off-topic? Sorry Steve!

 

[BBob]

Hams gap power transformers and use them as push-pull class B modulation transformers by adding a center tap. Lots of power in that application, and gapping is an art. A good modulator might push out several hundred Watts with a pair of 810s and can modulate a couple of 4-400s in class C. That's some heavy iron.

I have not seen toroidal transformers used in plate mod (where the transformer carries the DC to the plate of the RF power tube) applications and assumed that the DC current of the plate voltage through the mod transformer would indeed preclude their use due to saturation. But is that correct? It takes a lot of energy to saturate some of the biggies.Dunno.

BTW that reminds me of Jim Brown's excellent paper:

http://www.audiosystemsgroup.com/RFI-Ham.pdf

Bob

http://microphonium.blogspot.com

 

[mcs]

[quote author="BBob"]I have not seen toroidal transformers used in plate mod (where the transformer carries the DC to the plate of the RF power tube) applications and assumed that the DC current of the plate voltage through the mod transformer would indeed preclude their use due to saturation. But is that correct? It takes a lot of energy to saturate some of the biggies.Dunno.[/quote]
For push-pull operation no gap is usually needed. The DC current in the two halfs of the primary cancel each other out. There may still be a small imbalance, and some toroids don't like that - others work fine.

Best regards,

Mikkel C. Simonsen

 

[BBob]

What about the secondary, where one end is connected to B+ supply, and the other right to the anode of the PA? There's no center tap. Typical currents of 800ma at 2 KV or more at full swing.

Bob

 

[zebra50]

Steve,
These guys are in Powys...

Audio trannies

 

[sonicmook56]

Didn't the Neve 8108 have a option for a toroidal on the output?

Might be a good model to look at for your research.