Class A output transformer - UK Vintage Radio Repair and Restoration Discussion Forum

vinrads · 18th Oct 2020, 10:41 am

I have just been reading up on class A valve output stages ,and was wondering does the o/p tx have to be designed for class A, Mick.

Herald1360 · 18th Oct 2020, 11:56 am

Yes and no....

A single ended class A output transformer needs to be designed to handle dc magnetisation of the core, a push pull one can just be the same as any other class of p-p transformer AFAIK.

Graham G3ZVT · 18th Oct 2020, 11:58 am

I suppose on one level the answer must be yes to deal with the static saturation current, but in practice wouldn't a p-p transformer have a centre-tap?

I have a large p-p transformer with lots of primary and secondary taps that I have used over the years in p-p & single-ended amps as well as a choke and modulation transformer, all quite successfully.

I would warn you to read only appropriate "vintage" sources, there's a lot of PX4 robbing "S.E.T." audiophoolery around these days!

GrimJosef · 18th Oct 2020, 12:09 pm

All single-ended audio amps have to be Class A but, as Graham says, it's not the 'Class A-ness' that determines the output transformer design. It's the need to handle a DC magnetic field.

Push-pull output transformers don't need to be specifically designed for Class A. All Class AB amps run in Class A when the signal level is low, after all.

Cheers,

GJ

Graham G3ZVT · 18th Oct 2020, 4:21 pm

Tell you what, I've been looking for the transformer I mentioned for a year or more. Mention it here this morning, and this afternoon it turns up!

Click image for larger version

Name: DSCF0749.jpg
Views: 122
Size: 58.7 KB
ID: 218244

Click image for larger version

Name: DSCF0752.jpg
Views: 115
Size: 53.8 KB
ID: 218245

Cheers Mick!

Radio Wrangler · 18th Oct 2020, 4:27 pm

It is the single-endedness that gives un-cancelled DC quiescent current. Single-ended amplifiers are normally set up for class A (class B and C would sound terrible)

But RF amplifiers are sometimes single-ended in classes A, B, and C. In the non-linear classes, subsequent filtering completes the missing half of the waveform. However, RF amplifiers don't need the amount of inductance it takes to handle audio and that eases the saturation issue.

Push-pull amplifiers cancel the DC components of the anode currents of the two valves, even in class-B. There is no net progressive growth of magnetisation.

You can use a plain (gapless) single ended transformer, but you do lose power capability because you have to restrict the bias current you can set, as well as the signal amplitude. Using a gapped core to throttle back the DC flux component saves you from saturation, but reduces the inductance, so to get the bass end back, you need more turns and that needs a bigger core to combat saturation to the same extent. Not quite lose-lose, but definitely not win-win. But then some people think the sun shines out of a single-ended amplifier's anode.....

David

GrimJosef · 18th Oct 2020, 5:02 pm

Quote:

Originally Posted by rambo1152

... I have a large p-p transformer with lots of primary and secondary taps ...

Ah yes, the OP transformer you need when you really haven't made your mind up

!

I have its big brother. This would give the 'tube rollers' something to think about.

Cheers,

GJ

Radio Wrangler · 18th Oct 2020, 6:36 pm

Now that could also get them going trying different ultra-linear ratios as well.

David

G6Tanuki · 20th Oct 2020, 7:16 pm

Core-saturation is indeed the big problem in *single-ended* Class-A amplifiers. But in a Class-A push-pull amp the symmetrical static currents of the two halves of the push-pull arrangement neutralise each-other so the transformer has an easier life.

[It gets fun again when you're using a push-pull amplifier (of whatever class) to anode-modulate a RF power-amplifier for full-carrier AM transmission; though the currents in the primary-windings balance each-other out you have to deal with the static DC current flowing through the secondary which needs to exist to support the standing-carrier. Woden produced a good range of transformers [their 'UM' series] which were characterised to handle this].

All things considered, I've re-wound old 'wall-wart' transformers with 100+100 turns of 22-gauge wire and used them for 'single-ended' Class-A audio amps - an OC35 or NKT404 can happily deliver 6 or 7 Watts of audio in such a configuration.

Cobaltblue · 21st Oct 2020, 3:52 pm

Quote:

Originally Posted by G6Tanuki

.

All things considered, I've re-wound old 'wall-wart' transformers with 100+100 turns of 22-gauge wire and used them for 'single-ended' Class-A audio amps - an OC35 or NKT404 can happily deliver 6 or 7 Watts of audio in such a configuration.

Did you find you needed to Gap the transformer for this?

3 decades ago I spent some time experimenting with this sort of arrangement and unless I used the transformer from an old car radio (i.e. one built for the job) I got disappointing results.

Cheers

Mike T

G6Tanuki · 21st Oct 2020, 6:47 pm

Yes, when there was 70mA or so of DC flowing through the reworked 'modulation transformer' it was worthwhile to dump the classic interleave of the E and I lams and fit them with what-amounted-to-a-magnetic-gap to lessen saturation-nasties.

I also played around with the idea of the secondary of the mod.trans not passing the PA anode-current and instead feeding modulation via a capacitor to the PA anode which was DC-fed via a single-winding AF choke. Though in theory this made sense I never really found it worthwhile - it just made the resultant thing bulkier.

Saturation in the modulation-transformer, with associated limiting of the audio-waveform - could be seen as a form of 'brute force' speech-processing!

18th Oct 2020, 10:41 am	#1
vinrads Dekatron Join Date: Feb 2009 Location: Chesterfield, Derbyshire, UK. Posts: 3,759	Class A output transformer I have just been reading up on class A valve output stages ,and was wondering does the o/p tx have to be designed for class A, Mick.

18th Oct 2020, 11:56 am	#2
Herald1360 Dekatron Join Date: Feb 2007 Location: Leominster, Herefordshire, UK. Posts: 16,535	Re: Class A output transformer Yes and no.... A single ended class A output transformer needs to be designed to handle dc magnetisation of the core, a push pull one can just be the same as any other class of p-p transformer AFAIK. __________________ ....__________ ....\|____\|\|__\|__\_____ .=.\| _---\__\|__\|_---_\|. .........O..Chris....O

18th Oct 2020, 11:58 am	#3
Graham G3ZVT Dekatron Join Date: May 2010 Location: Greater Manchester, UK. Posts: 18,711	Re: Class A output transformer I suppose on one level the answer must be yes to deal with the static saturation current, but in practice wouldn't a p-p transformer have a centre-tap? I have a large p-p transformer with lots of primary and secondary taps that I have used over the years in p-p & single-ended amps as well as a choke and modulation transformer, all quite successfully. I would warn you to read only appropriate "vintage" sources, there's a lot of PX4 robbing "S.E.T." audiophoolery around these days! __________________ -- Graham. G3ZVT

18th Oct 2020, 12:09 pm	#4
GrimJosef Dekatron Join Date: Sep 2007 Location: Oxfordshire, UK. Posts: 4,311	Re: Class A output transformer All single-ended audio amps have to be Class A but, as Graham says, it's not the 'Class A-ness' that determines the output transformer design. It's the need to handle a DC magnetic field. Push-pull output transformers don't need to be specifically designed for Class A. All Class AB amps run in Class A when the signal level is low, after all. Cheers, GJ __________________ http://www.ampregen.com

18th Oct 2020, 4:21 pm	#5
Graham G3ZVT Dekatron Join Date: May 2010 Location: Greater Manchester, UK. Posts: 18,711	Re: Class A output transformer Tell you what, I've been looking for the transformer I mentioned for a year or more. Mention it here this morning, and this afternoon it turns up! Cheers Mick! __________________ -- Graham. G3ZVT

18th Oct 2020, 4:27 pm	#6
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,868	Re: Class A output transformer It is the single-endedness that gives un-cancelled DC quiescent current. Single-ended amplifiers are normally set up for class A (class B and C would sound terrible) But RF amplifiers are sometimes single-ended in classes A, B, and C. In the non-linear classes, subsequent filtering completes the missing half of the waveform. However, RF amplifiers don't need the amount of inductance it takes to handle audio and that eases the saturation issue. Push-pull amplifiers cancel the DC components of the anode currents of the two valves, even in class-B. There is no net progressive growth of magnetisation. You can use a plain (gapless) single ended transformer, but you do lose power capability because you have to restrict the bias current you can set, as well as the signal amplitude. Using a gapped core to throttle back the DC flux component saves you from saturation, but reduces the inductance, so to get the bass end back, you need more turns and that needs a bigger core to combat saturation to the same extent. Not quite lose-lose, but definitely not win-win. But then some people think the sun shines out of a single-ended amplifier's anode..... David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

18th Oct 2020, 6:36 pm	#8
Radio Wrangler Moderator Join Date: Mar 2012 Location: Fife, Scotland, UK. Posts: 22,868	Re: Class A output transformer Now that could also get them going trying different ultra-linear ratios as well. David __________________ Can't afford the volcanic island yet, but the plans for my monorail and the goons' uniforms are done

20th Oct 2020, 7:16 pm	#9
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 13,996	Re: Class A output transformer Core-saturation is indeed the big problem in single-ended Class-A amplifiers. But in a Class-A push-pull amp the symmetrical static currents of the two halves of the push-pull arrangement neutralise each-other so the transformer has an easier life. [It gets fun again when you're using a push-pull amplifier (of whatever class) to anode-modulate a RF power-amplifier for full-carrier AM transmission; though the currents in the primary-windings balance each-other out you have to deal with the static DC current flowing through the secondary which needs to exist to support the standing-carrier. Woden produced a good range of transformers [their 'UM' series] which were characterised to handle this]. All things considered, I've re-wound old 'wall-wart' transformers with 100+100 turns of 22-gauge wire and used them for 'single-ended' Class-A audio amps - an OC35 or NKT404 can happily deliver 6 or 7 Watts of audio in such a configuration.

21st Oct 2020, 6:47 pm	#11
G6Tanuki Dekatron Join Date: Apr 2012 Location: Wiltshire, UK. Posts: 13,996	Re: Class A output transformer Yes, when there was 70mA or so of DC flowing through the reworked 'modulation transformer' it was worthwhile to dump the classic interleave of the E and I lams and fit them with what-amounted-to-a-magnetic-gap to lessen saturation-nasties. I also played around with the idea of the secondary of the mod.trans not passing the PA anode-current and instead feeding modulation via a capacitor to the PA anode which was DC-fed via a single-winding AF choke. Though in theory this made sense I never really found it worthwhile - it just made the resultant thing bulkier. Saturation in the modulation-transformer, with associated limiting of the audio-waveform - could be seen as a form of 'brute force' speech-processing!