Which PT?

[Russ]

Howdy,

I'm designing/building a 2xEL34, class AB, SS rectifier based amp and I'm trying to pick a power xformer, but I'm unsure about the current requirements. I'm considering using a Hammond 272FX with 300-0-300vac @ 150ma.

I figured the 272FX will put out around 430 plate volts and with 25w/EL34 dissipation: 25w/430v = 58ma per EL34. And including 4 12AX7 preamp tubes at 3ma/triode max => 24ma. So I've got max current draw at about 140ma: 2 x 58 + 24 = 140ma...

...but most of the schematics of similar amps I've seen use a 272HX (200ma) or even a 272JX (250ma). These seem like overkill, 150ma seems like enough - am I off here, what am I missing?

[stokes]

A good margin of safety would be to double the amp rating.If you anticipate 140ma draw,I would go at least 200ma, 250ma would be better.

[Russ]

Is my calculation of 140ma correct, because if it is, why bother with all that extra current capacity, cost, weight, space etc. Under what conditions will the amp ever draw more than 140ma?

[stokes]

It is common in tube amps to use components of twice the anticipated current draw.Your calculations are for the amp at idle,put a signal thru it and things change.If you have a problem with a power tube drawing more current before it blows and your tranny isnt up to the extra current it could go south as well.If you have a leaky cap in the B+ rail it could,over time,fry the tranny.Transformers,both power and output are the heart and soul of a tube amp,its just foolish to try to cheap it here. "...but most of the schematics of similar amps I've seen use a 272HX (200ma) or even a 272JX (250ma). These seem like overkill, 150ma seems like enough - am I off here, what am I missing?"These schematics you speak of here are the result of much R&D by engineers and the like,who likely know much more than we do.There is a good reason they use what they do.

[Russ]

Russ

Thanks for the responses.
Actually my calculation were for max current, not idle current. I was assuming max dissipation (25w) for each EL34. Biasing at 70% dissipation would work out to about 105ma at idle.

So is it true that the output tubes will conduct whatever the power supply can give them? If the amp is cranked at 10:

With an undersized or boderline capacity PT, the tubes draw their max current and the PT gets stressed/fried.

With an oversized PT, the PT can handle it but the tubes draw beyond their max current and they get stressed/fried.

The more I think about this, the more confused I'm getting. What effect DOES the PT current capacity have on the response & sound of an amp?

[Russ]

Russ

Thanks for the responses.
Actually my calculation were for max current, not idle current. I was assuming max dissipation (25w) for each EL34. Biasing at 70% dissipation would work out to about 105ma at idle.

So is it true that the output tubes will conduct whatever the power supply can give them? If the amp is cranked at 10:

With an undersized or boderline capacity PT, the tubes draw their max current and the PT gets stressed/fried.

With an oversized PT, the PT can handle it but the tubes draw beyond their max current and they get stressed/fried.

The more I think about this, the more confused I'm getting. What effect DOES the PT current capacity have on the response & sound of an amp?

[stokes]

"With an undersized or boderline capacity PT, the tubes draw their max current and the PT gets stressed/fried." This is correct.
"With an oversized PT, the PT can handle it but the tubes draw beyond their max current and they get stressed/fried." This is not correct.With an "oversized PT" your tubes will not draw "beyond their max and get stressed/fried".Your tubes will only draw what they will draw,not beyond...unless there is a problem causing them to draw more.i.e. internal short or say a screen resistor,merely having a power supply that is stiffer than you "need" is not going to stress your tubes.On the other hand,having an under rated or marginal PT can cause a lot more sag,from trying to keep up with the demand,which will eventually burn out the PT.And remember if your tubes are starving for current the PT cant deliver,they are more likely to be stressed here than with an over rated PT.Okay,if your idle current is 105ma,then the 200ma PT would be almost 2x the expected idle.This is a safety margin.Like I said earlier these schems you are looking at were designed by people with more know how than us.

[Russ]

I don't think it's true that "tubes will only draw what they will draw,not beyond..."
If you look at el34 graphs for plate voltage vs current, an el34 will conduct way beyond its dissipation if the grid voltage is low enough. As an example, at 400v plate volts, an el34 will conducts about 200ma with -20 grid volts. This is 80 watts of power - way beyond its 25w capacity. It will conduct even more as grid volts goes to zero. For a fixed plate voltage, tube current draw depends on grid voltage.

If this is right then I'm back to my original thought about undersized vs oversized PT.

It seems like the PT, or actually power supply, current capacity should match
what's need to let the tubes reach their max dissipation. At this point the PT is at max capacity, the tubes are at max, and the signal staturates/clips into
distortion. PT and tubes both suffer equally. I'm thinking that PT/power supply current capacity is key to getting an output stage to overdrive properly. Whadaya think?

[stokes]

Okay,if your tubes are idling at 40ma's as the current draw increases your plate volts will drop accordingly,by the time you reach your stated 200ma's you wont have 400v on the plates any more.I dont know where you are coming from saying if your grids have -20 grid volts you will draw 200ma's at 400 plate volts.If you are drawing 200ma's thru an EL34 at 400 volts something is wrong and that tube or tubes will meltdown.I dont know where you are getting these numbers from,but something aint right.When I am biasing a pair of EL34's or any other power tube,I start with -55v on the grids and work down to around -40 or so to bring the current up.Having an over rated power supply will not make your circuit draw more.Note the word DRAW.Your circuit draws the current it needs,it is not "force fed" current by the power supply as you seem to think.If I use a tranny that can supply 1000ma's,it doesnt mean that I am going to force my circuit to draw more than it is designed to.The extra current rating is a safety margin.Look at this,when you first turn the amp on,it draws mega current momentarilly,till the caps charge up,this can be seen if you plug into a light bulb in series current limiter.I cant give you an exact number,but it is likely close to twice the normal idle draw,as can be seen when that bulb burns full bright for a second or two.You sound hell bent on using an under rated PT,I said it before and I'll say it once more,the schematics you look at that you think are using too big a PT were designed by people who know more than you or I,but if you think an under rated tranny is the way to go,go for it,you will be able to keep your coffee warm on that tranny-till it dies.

[R ski]

Pt requirements

Long time ago I followed your assumption, using industrial control transformers
for the plate supply. Not enough head room (current) will whimp out, when
pushed. This is not good, for two reasons. First, for clean head room, the
amp will suffer. Second, whem max gain and single notes are played, the note
has a spongy sound, not always desirable.

As far as tube life, the biasing will govern the conduction state. Lowering
the source, doesn't insure longer life.http://music-electronics-forum.com/i...es/biggrin.gif