Output Transformers

[MikeMurray]

Hey guys

So what's the deal with working out necessary primary impedence? I read somewhere that it's related to B+ but I'm not sure entirely. Can someone fill me in here.

Thanks
Mike

[limbe]

Yes,the primary impedance increase as the plate voltage rises if that is your question.
Of course there is a whole lot more to know about output transformers

[SamBooka]

Yeah.. but the way you put it is funny. Limbe is right.. but I dont think that was your question.

Take this with a grain of salt. A tube has an output impedance. This does change depending on the plate voltage. Think of the output impedance as the "ideal" load for the operating point of the tube. You can get this info from the characteristic curves on a data sheet for a tube ( Tubewizards book explains how to do this IIRC.. data sheets are all over the web .. or go to tdsl.duncanamps.com)...

Ok.. where were we? Ok.. say tube X at whatever voltage you are running it has an output impedance (is this the same as plate resistance?) of 5Kohms. Ok? Now you have a speaker that is 8ohm. If you had a speaker that was 5Kohms you could just plug it directly to the tube (ok.. maybe not .. but bear with me). But because there is such a big difference you need something to match one to the other. This would be an output transformer.

Now.. to your question. Transformers have a turns ratio. How many turns on the primary vs turns on the secondary. Let us say this number is 25 to 1 (25:1). Reflected impedance is the square of the ratio by the impedance presented.
In this case (25X25) X 8ohms = 625 X 8ohms = 5000ohms.

So with a transformer with a 25:1 ratio and an 8ohm speaker you are presenting a 5000ohm load to the tube.

This is a really dumbed down and incorrect way of looking at it. but that is all my brain can muster :)

[MikeMurray]

Thanks Sam. I understand the theory about how transformers are wound just figuring out the optimal primary impedance baffles me.

[celeste]

Oh, you are looking for optimal, that is easy. There is not one, especially with MI amps where distortion is the tone. The rule of thumb is to start with 2X plate impedance and work from there, higher and power and distortion goes down. Go lower on the impedance and you gain power and distortion raises. For a given operating condition, each impedance will have a unique distortion signature, which is best is a matter of taste. As an example, some people like PP 6V6's at 6.6K and some people prefer them at 8K, there is plenty of room for both to be right

[MikeMurray]

Great thanks :) That explains a ton. However I'm still a little unsure here.

Impedance <> resistance yes? Then how would I get that from here?

http://www.ampage.org/td/vtd5881.html

Thanks so much, really helps :)

Mike

[SamBooka]

Sorry.. try this
http://www.freewebs.com/valvewizard1...Gain_Stage.pdf

[limbe]

Excuse me for asking Sam,It was a good article about preamp stages,but why did you bring it into a discussion about output tubes and transformers?

[MikeMurray]

Hey guys

Bumping this thread..

Speccing my transformers atm. Anything in particular? The primary (Center tapped for push pull right?) and secondary(ies) obviously. Wattage as well.. What effect does this have really? Anyone tested?

Then would anyone care to fill me in on power transformers?

For example, 240v primary, 350v secondary @ 120mA, 6.3v @ 3A, 5v @ 3A, Single 50v line for biasing.

That correct? Also would there be any advantages/disadvantages to overspeccing the current handling? Besides weight :P

Thanks fellars

[celeste]

Depends on what you want to do with it.

Yes the primary has to be center tapped for a PP amp.
I like my secondaries to have 4,8,16 ohm taps so have a tap for most anything that is going to come at me, and I can add cathode feedback to the power tubes if I feel I need to.
Wattage, a lot depends on what the limiting factor is. Usually power and current are spoken of separately. When that is the case, is limited by how much iron you have and how much core distortion you are willing to put up with, and current is limited by the gauge of the wire in the primary.

winding interleaving is important, it is the primary and secondary windings being wound among each other, like 3 layers of primary, then one of secondary. Interleaving it a bit of a black art

Core material is important as well, in the US the usual choices are M6 and M27. M6 is more articulate, it magnetizes deeper and faster, has less "magnetic inertia" if you will. M27 has lower permeability so can run higher flux density. Which is better is a matter of taste

On your PT, spec your voltages loaded. Some winders quote specs for a loaded winding, and some for an unloaded winding and give regulation.

[MikeMurray]

Deluxe Reverb clone :D

Ok - planned to have those taps in anyway for safety sake.

What's core distortion? :| Gotta understand these SA folk aren't so techy... It's more like "get the job done".

Ok I see. Thanks for the tip about loaded volts.

Mike

[MikeMurray]

Originally Posted by MikeMurray

Hey guys

That correct? Also would there be any advantages/disadvantages to overspeccing the current handling? Besides weight :P

Thanks fellars

And a little bump ;)

[celeste]

Originally Posted by MikeMurray

And a little bump ;)

To up the current, you have to up the size of wire you wind the primary with, to understand why that can be bad, you have to think about the EI lamination the core is built up from. There are other kinds of cores, but we will only deal with EI for now. When you put an E with an I, they form 2 "windows" that all the windings have to pass through. if there is to much wire, you have to go to the next large lamination. That will likely increase the leakage inductance and inter winding capacitance. While that may or may not be a bad thing in terms of tone, it makes for a very different OT.

Core distortion is the result of magnetic flux not following current exactly. The flux kind of acts like it has inertia, minus any over shoot. The higher the flux density, the worse the difference is. There are also eddy currents induced in the core itself by the changing magnetic flux. Eddy currents also cause distortion. A great deal of effort is put into controlling those currents. Core steel has lots of silicon in it, it makes the steel less conductive. Many laminations are "grain oriented" That means as the steel is rolled into sheets, the crystals silicon are stretched in the direction of the rolling, or perpendicular to the rollers and forms a "grain". It is easier for electrons to flow parallel to the grain then across it, so the lamination are punched from the sheet so the grain interrupts the current flow in the core once it is assembled.

[MikeMurray]

So interesting!!! Never knew there was such an art to transformers!!