Push Pull to a single ended transformer

[LooseChange]

I know I've seen this somewhere before but can't find it.
Can a push pull transformer be used as a single ended transformer. I'd like to switch from one to another.
My experience says that a push pull with one tube pulled or if half the phase inverter is not working the amp gets distorted quickly and sounds bad.
Thanks!

[triode]

It is not optimum by any means. A SE transformer needs to be air gapped (most of the time) and is built differently.
A PP tranny is also designed to have near zero DC current from plate lead to plate lead (e.g. one is pushing, one
is pulling)... where a SE transformer is designed to handle all of that current all of the time.

Shoehorning does not work well with iron shoes.

[VMS]

Do you mean something like this:

http://ampbooks.com/home/amp-technology/amp-patent-7173488/

[tubenit]

LC,

If you look in the TUT series, there is a Soma84 amp built by KOC & it has a simulated SE feature to it even though it is a P/P amp.    If you can't find that schematic, let me know and I can probably draw up what he did in that area.

With respect, Tubenit

[HotBluePlates]

Building on what triode said:

Because the SE transformer must handle unbalanced d.c., the core must be larger for the same power output to the same low frequency limit. Push-pull transformers by comparison can have smaller cores because they only have to handle flux due to signal current.

Note that the Hammond 1650F (25w) is 4 pounds; the Hammond 125GSE (25w) is 6 pounds.

The 125GSE is rated down to 20Hz, where the 1650F is rated down to 30Hz. However, that is close enough for a reasonable comparison.

The SE unit is 50% larger for the same power handling capability.

[LooseChange]

Thanks guys.
But doesn't Mesa do this with a couple of their amps... I think like the 5-50?

[Boots Deville]

Would the SE design's larger core, air gaps be detrimental to PP operation? 

I don't know what Mesa is doing, but maybe they are using an air-gapped OT with a center tap primary so it can "go both ways".

[kagliostro]

Soma84 .................................

HERE:  http://www.mediafire.com/file/6621tr20ij2m51m/SOMA84.pdf

Kagliostro

[stingray_65]

Would the SE design's larger core, air gaps be detrimental to PP operation? 

I don't know what Mesa is doing, but maybe they are using an air-gapped OT with a center tap primary so it can "go both ways".

I'm curious about this, but its been my life experience that things that do double duty often compromise so that both work. usually I'm of the opinion that they compromise the best features, but that may be my piss poor outlook I get now and then! 

[phsyconoodler]

I tried using a large 15 watt PP OT on a champ and it sounded ok but when I finally sprung for a proper SE tranny it sounded worlds better.
 HUGE difference.

[jeff]

 I think this is what's going on. What it sounds like is that the DC current is the problem with SE transformers. This is why they need to be air gapped and made bigger.

 The primary is using elecrticity to make an electromagnetic field and the secondary is using that electromagnetic field to make electricity. A SE transformer is designed to take into account the magnetic field created by the DC current but a PP is not. It's not a problem with PP transformers running PP because the field created by the DC through one side of the primary is counteracted by the field created by other side which is out of phase.

 I don't think you can just wire a PP transformer as a SE, you need that other tube. The schematic is not just wiring a PP transformer as a SE transformer. Even though the other tube has no AC signal, it's still on. One tube is operating SE and the other tube is canceling out the problem caused from using the PP as a SE.

That's how I see it. Maybe someone with more technical knowledge can explain it better(or tell me I'm way off base).

[HotBluePlates]

But doesn't Mesa do this with a couple of their amps... I think like the 5-50?

If you design/build a push-pull amp, you can have a low-power quasi-single-ended option by turning off the drive to one side of the push-pull output.

The key here is to still have the non-driven side idling through the OT, so that its d.c. offsets the d.c. applied by the idle condition of the other side. That minimizes the standing flux in the OT, and allows more signal current before saturation.

Mesa's "patented system" is generally a patent on the switching to accomplish the push-pull to single- ended change. Other have used similar approaches; for example, London Power offers a continuous range between push-pull and quasi-single-ended by inserting a pot between the phase inverter output and power tube grid on one side of the output stage (much like a post-phase inverter master volume, but impacting only, say, the "push side" instead of both sides).

What it sounds like is that the DC current is the problem with SE transformers. This is why they need to be air gapped and made bigger.

That's how I see it. Maybe someone with more technical knowledge can explain it better(or tell me I'm way off base).

Exactly.

Simple terms:
Say you and I are able to carry 50 pounds each without serious strain. However, you (single-ended) also have a backpack with 25 pounds in it; I (push-pull) am unencumbered. You can accept an additional 25 pounds, while I can accept an additional 50 pounds. It looks like I'm doing more useful work, but really, we're both working as hard. My (push-pull) effort is going into all the extra weight I'm carrying.

A lump of iron can only handle so much magnetic flux before it reaches its limit. Single-ended "wastes" some of that lump's capacity because the idle current magnetizes the core somewhat before you even apply a signal. So you need a bigger lump to handle the same power, compared to an output stage with no unbalanced d.c.

There are other differences (such as an air gap) which make it undesirable to simply take a push-pull OT and run it at lower power as a single-ended transformer.

If you have an emergency situation, you can use a push-pull OT in a single-ended circuit. It will make sound. It will not make as much sound, with as little distortion, at a given low frequency as you could get using the right part.