Specing a power transformer

[OverDriven]

I want to make sure I have this right before purchasing a power transformer. Say we have an amp with 2x6L6GC and 4x12AX7.

6L6GC:
Max dissipation: 30 watts
Heater current: .9 amps
Plate voltage being used: 430 volts

12AX7:
Max dissipation: 1.1 watts
Heater current: 300mA
Plate voltage being used: 350 volts

In order to spec a good power transformer for this build, the following math is used:

6L6:

30 divided by 430 = 93mA x 2 = 186 mA

12AX7:

1.1 divided by 350 = 3mA x 4 = 12mA

So we have a current draw on the HT winding of 198mA. Lets add 50% for safety. That comes to 297mA.


Now for heaters:

6L6:

.9mA x 2 = 1.8A

12AX7:

300mA x 4 = 1.2A

That makes a total of 3A on the filament secondary. Again, lets add 50% for safety. So we have 4.5A.

So this power transformer would look something like this:

320-0-320 @ 300mA HT secondary
and a 6.3 volt secondary @ 4.5A

Is all of this correct?

[R.G.]

How much power goes out to your speakers?

Presumably you want some of that, too.

[Tubis]

How much power goes out to your speakers?

Presumably you want some of that, too.

I dont understand? The current trough the tube is the current trough the primary(OT). If you get more current the tube will burn(or the fuse)(or the primary winding).

[MWJB]

What kind of rectifier are you going to use?

Yeah, you're in the ball park with PT secondary voltage (assuming fixed bias). PT current handling, brand of tube and bias conditions will greatly affect final B+, so think in terms of +/-10v? I'd go a little higher, say 325 to 330VAC to ensure plate voltage doesn't drop too much if you want a lot of plate current?

300mA might be overkill, 200mA would be good, 150mA minimum. Add together anticipated plate currents, add 10mA per power tube screen, add 10mA per 12AX7. E.g. 40+40+10+10+10+10+10+10= 140mA...plus some excess.

As much current as poss on the 6.3VAC winding is good, 4.5A is fine, you may find a PT with up to 6A?

Are you really going to run those 12AX7s at 350vdc on the plates?

The OT transforms your high impedance, high voltage, low current from the tubes into low impedance, low voltage high current from the speakers. The PT discussed will allow the amp to make 40 odd W, if the OT will.

[Steve Conner]

R.G. has a point. A lot of people don't seem to realise that the plate dissipation of the tube is not the same as the power drawn from the B+, or the power delivered to the speakers, or the maximum plate dissipation on the tube datasheet.

Let's take an example: One of my homebuilt amps has two KT88s idling at about 45mA each, on a B+ of 475V. 90mA at 475V is a power draw of 43W, of which 21.5W is dissipated in each tube, and none delivered to the load.

When I crank it to full power using a signal generator and dummy load, the current drawn from the B+ goes to somewhere over 300mA. That's 145W drawn from the B+ and I measure something like 80W delivered to the load, therefore the rest must be dissipated by the tubes, which gives us 32.5W per tube. (and an efficiency of about 50%, which is about as good as you can expect from a tube amp with the "latest" beam tetrode technology.)

None of these figures has much to do with the 40W plate dissipation rating of a KT88.

FWIW, solid-state amps aren't that much more efficient: maybe 65% at full power. The difference is in the idle current: you can run transistors at very low idle current and still get no audible distortion. Tubes always make crappy-sounding crossover distortion when they're biased cold.

[Tubis]

Still, all of the current goes trough the tubes. there is no other way for it to go so if more current is to be drawn at the same Voltage, that will be developed over the tube. The load takes nothing in form of current. There is no current connection between the primary and secondary of the OT. The primary takes the voltage and there will be lower voltage across the tube. Talking about signal here not, idle.

[MWJB]

"R.G. has a point. A lot of people don't seem to realise that the plate dissipation of the tube is not the same as the power drawn from the B+, or the power delivered to the speakers, or the maximum plate dissipation on the tube datasheet." Very true, a common novice error is to assume that more idle current/plate dissipation W will result in more W AC output, after a point you will just lose headroom and make more heat.

"Tubes always make crappy-sounding crossover distortion when they're biased cold." Or when they're biased OK and under severe overdrive. It's only crappy if you don't like it, harp players in particular often use relatively cold bias - I've seen guitar players plug into harp players amps & jump through hoops over the "tone", suggestions that they were "wrong" fell on incredulous, unbelieving ears. Bias by ear, & monitor carefully once you have established that you're not going to burn anything up.

[OverDriven]

How much power goes out to your speakers?

Presumably you want some of that, too.

Can you elaborate a bit in somewhat simple terms? I know I'm a newbie and these are probably stupid questions, but I'm trying my best to learn.

I dont understand? The current trough the tube is the current trough the primary(OT). If you get more current the tube will burn(or the fuse)(or the primary winding).

This is what I thought (I think)

What kind of rectifier are you going to use?

Yeah, you're in the ball park with PT secondary voltage (assuming fixed bias). PT current handling, brand of tube and bias conditions will greatly affect final B+, so think in terms of +/-10v? I'd go a little higher, say 325 to 330VAC to ensure plate voltage doesn't drop too much if you want a lot of plate current?

300mA might be overkill, 200mA would be good, 150mA minimum. Add together anticipated plate currents, add 10mA per power tube screen, add 10mA per 12AX7. E.g. 40+40+10+10+10+10+10+10= 140mA...plus some excess.

As much current as poss on the 6.3VAC winding is good, 4.5A is fine, you may find a PT with up to 6A?

Are you really going to run those 12AX7s at 350vdc on the plates?

The OT transforms your high impedance, high voltage, low current from the tubes into low impedance, low voltage high current from the speakers. The PT discussed will allow the amp to make 40 odd W, if the OT will.

I kind of follow you, but when you give the 40+40+10+10+10+10+10+10, I'm assuming that you are saying that each power tube will only consume 40ma each. I don't quite understand how that is. I thought it would be much higher.

[Enzo]

10ma for each 12AX7 triode is pretty generous, I think. That would be quite a voltage drop across the typical 100k plate load resistor, don't you think? Even for both triodes in one tube together. Not to mention across the resistors in the B+ string.

[OverDriven]

10ma for each 12AX7 triode is pretty generous, I think. That would be quite a voltage drop across the typical 100k plate load resistor, don't you think? Even for both triodes in one tube together. Not to mention across the resistors in the B+ string.

Yeah I think 10ma is generous too. You did make me realize that I made a big mistake though. I gave 3ma per preamp tube...I somehow didn't take into account that there are two triodes in there. So 6ma per 12ax7 tube would be a good figure, right?

[Enzo]

A .lot more reasonable than 10, but I usually expect about 1ma or 2. On average.

Look at some examples, cathode voltage over cathode resistor wil tell you the current expected.

[MWJB]

"10ma for each 12AX7 triode is pretty generous, I think." We're building in a little (OK a big) margin, if your B+ current rating is too low it will pull down B+ voltage and all other secondary voltages when large demands are placed on it.

Overdriven - Why are you bothered about "only" allowing 40mA for the power tubes, but concerned about another 16mA total for the preamps?

You will find that a Fender style PT with adequate heater current handling is easy to find with 150mA B+ secondary. If we cut corners and get 'sensible' on screen current and preamp current requirements you are going to end up with a Deluxe Reverb PT (120mA)...workable for what you need & fine if it's in a DR that you're using with 6L6s (and not going loopy with plate current) BUT if you're designing from scratch you may as well build in a margin for longevity and consistent running.

"I'm assuming that you are saying that each power tube will only consume 40ma each. I don't quite understand how that is. I thought it would be much higher." At idle yes 40mA is typically at the high end of the scale. When you're playing demands on the PT are transient (in Steve's earlier example with a signal generator - constant signal - & dummy load I wouldn't be surprised if voltage dropped accross the B+ rail and that current was also being drawn from other sources rather than just KT88 plates).

I am assuming that you are going to be running fixed bias? If so how much current were you intending to idle the tubes at, 30-35mA would be normal, 40 at a push. Note also that I suggest allowing some "excess". If you were going to run cathode bias then your idle current is more likely to be around the 50-55mA mark, but you are going to have to go with a higher B+ rating to allow for voltage drop accross the cathode resistor and large idle current demand on the B+ rail pulling down your B+ so that you actually end up with 430 on the plates, say 345-0-345VAC?

Bear in mind that in cathode bias current is not going to rise appreciably when playing, we want the dynamics of the power supply to sag under power demands. In fixed bias we need to allow for a nominal rise in current from idle when playing. Sure you could allow 70mA per tube, as per your original estimate BUT you will have trouble finding a PT that meets your other requirements with a 300mA B+ secondary, that's more like the current rating for a 4x6L6 PT, such a PT won't typically have a rectifier winding (a nice option for a 2x6L6 amp). It's also unlikely to have a 320 to 330VAC secondary?

Few current production tubes will actually take 30W plate dissipation, if you try and run them at 30W plate dissipation your amp won't be as loud as if you back the current off, tone may be garbled, you'll just turn more watts into heat not sound. Even in cathode bias you're more likely to bias to 22-23W, in fixed 12-16W is typical.

[MWJB]

A 12AX7 can dissipate a watt per section (OK, it most likely won't need to), that's 4mA per triode at 250v, 10mA is not unreasonable as allowable margin from a design point of view.

The examples you refer to will be at idle, if Overdriven wants to try other 9 pin tubes other than 12AX7, these will draw more current.

[Tubis]

Can you elaborate a bit in somewhat simple terms? I know I'm a newbie and these are probably stupid questions, but I'm trying my best to learn.



This is what I thought (I think)

Remember that the maximum current for the tube is higher than the calculation from your maximum anod dissapation. Start calculating how high the voltage must swing across the primary on the OT to get the desired maximum effect over the load(speaker). Thats why you should match the OT to the tubes. Differnt matchings for differnt parameters.
You always get a bit of transformer loss so what you calculate on the primary will be lower on the secondary of the OT
Beware if you start square clipping it that effect and drawn current will increase dramaticly. The datasheets are for the given distortion.
Old datasheets ar great.

[Tubis]

I will have to say that 5mA per tube all depends on the connection. If you have a 100k resistor at 350V that will be 3.5mA if the circuit se it as a short to ground and there is no dropping resistors along the B+ line .
Sometimes you have a 220k resistor that vill make the current smaller.
Calculate it from that.
Now you dont have many preamptubes here but say in a Marshall 6100 with 7 of them. To overcalculate will be money wasted.