• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

Power transformer: How Big?

I need a little help sorting through information in a data sheet to pick a power transformer. Actually I already built the amp, I just want to be sure that I made a good choice with this power transformer.

I have 4 KT88s at 50mA quiescent and various class A voltage and current gain stages that I estimate at 70mA max. My total zero signal current will be less than 270mA.

According to the KT88 data sheet, they can dissipate up to around 150mA at max signal in the configuration that I am using them. (ultralinear with 560V B+) This must mean that that they dissipate this at the peak of the wave. It can't mean that this is an average value, since we would be way over the plate dissipation.

Working backwards from the max signal Pa+g2 of 33W, that gives us about 59mA per power tube +70mA for 306mA at max signal.

The power transformer that I bought is a Hammond 278CX which puts out 465mA. I figure that is plenty. Did I make any bad assumptions here?

What rules of thumb if any do you guys use when selecting a power transformer as far as current rating is concerned?

Thanks in advance.
 
The power should be 2X the idle current of the outputs, plus whatever current the driver/preamp tubes will eat up. The 400-0-400, 465 mA Hammond trannie will work, but it's a tad low on the current requirement. It may stress out (and distort) at high volume levels. I would get something at least 500 mA. Alden at Heyboer built me a 400-0-400, 500 mA transformer for only $125. Believe me it's worth it....these guys have been in business since 1957!
 
I am going to build a second unit, so I may just give that a try. I would really prefer something with a 120V primary so that when I plug it into a 120V outlet I don't get overly high output on the secondary. The 115V primaries of all the Hammond line kind of bug me.

I just wanted to get an idea of how much I should shoot for. I got an 800mA transformer quoted from Electra-Print to see how much it would cost and it was a bit expensive, so I'll probably give them a chance to quote a smaller unit as well.
 
I could do dual mono blocks. The Hammonds are certainly cheap enough. It kind of bugs me that they buzz real loud when I pull about half the rated current. I was really thinking about getting one custom made for the second unit to see if it buzzes less. Since I am using SS rectification I could switch from CT full wave rectification to bridge rectification, get rid of taps. I don't need filament windings as I am using a separate 10V transformer for regulated DC filaments.

I was hoping a couple of more people would weigh in on good targets for current output like TubeHead Johnny did. (Thanks TubeHead Johnny)
 
You don't say what filter topology you are using; if it's a critical inductance choke input you can use a higher fraction of the PT rated current, but with C input filter the peak ripple current can be quite a lot higher than the DC load current.. This may be a source of your buzzing. For C input the PT might need to be sized something like 2X or more, depending on the DCR of the PT and ESR of the 1st capacitor.

Michael
 
I am using C input filters for both amps.

The first amp (a Fisher X-101-B) that I used this PT on runs about 230mA quiescent and has about 180 Ohms in series with the first cap(70uF) to drop some voltage(since the Hammonds output voltage is high due to the 115V primary) and solid state rectification. The previous transformer(Fisher original) was totally silent. I had to add more series resistance(voltage too high) and eliminate the tube rectifier(no 5V winding) for the new transformer.

The second amp is my own creation that dumps current directly from the diodes into the 400uF cap. Huge peak currents here.

The funny thing is that the second amp doesn't buzz that much more than the first. I'll admit that I'm abusing the second transformer a bit, but the first one probably shouldn't buzz like it does.
 
I think TubehadJohnny is right. It seems to me that your 400 mA might be on the edge. I am building KT88 PP in triode mode, class-A, with SS bridge + CLC filtering and I am using 250 VA transformer with 1 A for B+. When I was making this decision I was googling and it seems to me that this is a typical transformer sizing for this kind of amp. It may be a bit oversized, but I think that supply should be always oversized ... Very good transformer that you can buy for these purposes is Lundahl LL1649.
 
dejanm said:
I think TubehadJohnny is right. It seems to me that your 400 mA might be on the edge. I am building KT88 PP in triode mode, class-A, with SS bridge + CLC filtering and I am using 250 VA transformer with 1 A for B+. When I was making this decision I was googling and it seems to me that this is a typical transformer sizing for this kind of amp. It may be a bit oversized, but I think that supply should be always oversized ... Very good transformer that you can buy for these purposes is Lundahl LL1649.

I think you might be doing a bit of a miscalculation there... the transformer you are specifying is smaller, not larger. 250VA (187.5 VA for the B+) vs. 454VA rated capacity of the hammond. It's also barely over half the rated voltage. 230V vs 400V. I do not see at all where this would be a better power supply for a high power amp. Lower B+ and lower VA capacity? Did we miss something? Did you mean to use 2? I am not stating that the hammond line is better, the one I have buzzes loudly too and it irks me, but in this case, I cannot see that the specified Lundhal would be a better choice unless you are using 2.. and even then the lowish B+ would be an issue (at least for me).

I also wanted to address the 33W Plate + Grid dissipation number above... 6550 and KT88s are 35W Plate dissipation tubes, some specs up to 42W or 42W max. That is not including the 6W screen grid dissipation. So it looks to me like you could be driving your tubes even harder.

Then to the 115V Hammond primaries... I do wonder if this just isn't a myth/misprint these days... my AC usually is 120-123V - I just checked it and it's at 122.8VAC. That's with 2 Fluke digital meters. Now, on the recent model hammond 200 series I have, the voltage out was 2% over spec on the B+ and 1% over on the heater line. That's in circuit under load. That to me is excellent. Dare I say, that's fare more precision than many of the circuits built by some DIY will ever see in terms of matching the "original spec or plan?" In fact, my only gripe about their line (and it appears to be a common one) is the buzz noise they make.

To the original question, I cannot fathom anyone saying you couldn't make a great audio amp with that power transformer, it's one of the biggest stock, economy/regular transformers you can buy. You are running the heater secondary over spec, maybe consider a separate filament transformer? Or just leave it as is.. you aren't pushing that as hard as many amp manufacturers have done before...

Would mono blocks be better? Monoblocks are always better... but is that really an answer to the question of, "will this transformer work?"
 
Cycline3,

On the KT88 amp, I am running heaters off of a separate 10V transformer(regulated DC). The HV winding is the only one being used on the Hammond. Forgot to mention that.

According to the data sheet, in the configuration that I am running the tube, I should be drawing 306mA from a 465mA transformer with unused filament windings. I don't see anything wrong with this. TubeHead Johnny suggested Power tube idle current x 2 plus class A stage current usage as a rule of thumb. This adds up to 470mA. That's darn close to 465mA, so I think I am fine with this transformer, especially since I am not using the filament winding.

Really my basic question is this: Will this transformer run too hot if I play loud music for several hours?
 
Cycline3 said:


I think you might be doing a bit of a miscalculation there... the transformer you are specifying is smaller, not larger. 250VA (187.5 VA for the B+) vs. 454VA rated capacity of the hammond. It's also barely over half the rated voltage. 230V vs 400V. I do not see at all where this would be a better power supply for a high power amp. Lower B+ and lower VA capacity? Did we miss something? Did you mean to use 2? I am not stating that the hammond line is better, the one I have buzzes loudly too and it irks me, but in this case, I cannot see that the specified Lundhal would be a better choice unless you are using 2.. and even then the lowish B+ would be an issue (at least for me).


I missed the fact that the desired B+ is 560V. Sorry for that. I am working with 320 V ... And yes I am having two of them, but one would be sufficient either.
 
SpreadSpectrum said:
Really my basic question is this: Will this transformer run too hot if I play loud music for several hours?

I personally don't think so. It may get quite warm, but that is the nature of the beast, isn't it? The transformer is rated at 465ma continuous, and as long as you are under that for an average value, you are fine.

People sometimes look at specs and think it's like a line in the sand, one that cannot be crossed. But the reality is, the transformer will supply 1ma, it will supply 465ma and every value in between. It will also supply MORE than 465ma, far more in fact, up until a point where it will self destruct from heat.

The manufacturer rates a transformer with a rating that you can run continuously that won't damage the transformer. Again, as long as you are under that you are fine.

As for the 2X rule, I do like that in terms of audiophile idealism and reserve current; however, it simply isn't a true requirement.

You can always test your B+ as well, run the amp with no input and gradually up it to max output and see if your B+ is sagging or changing. Assuming you have large filter caps, I have a feeling for normal listening levels (even with loud peaks) that is NOT happening.