Simple PT rating question

[nsubulysses]

I am curious how this works and how I should understand this spec in transformer ratings.

FOr instance, say a transformer has an 800V CT @ 535mA. I don't understand how this equates because if you have 4 power tubes that are being cranked full volume, it would seem that they would draw anywhere from 130-200mA each when driven hard. I understand it's not continuous but it does exceed 535mA. Is this ok because it is not continuous? Or is it ok beacuse the current is drawn from PS reservoir rather than directly through the high voltage secondary winding??

I am asking because I'm curious what would be necessary with 2, 4, and 6 power tube amps and how I should guage this when looking at different PT specs

[Old Tele man]

Remember, in Push-Pull, as 2 of the 4 tubes are conducting heavily, the other 2 are deep into cutoff, so average current is what's important.

[Dave H]

...if you have 4 power tubes that are being cranked full volume, it would seem that they would draw anywhere from 130-200mA each when driven hard.

Looking at the 6L6GC data sheet for the 55W class AB amplifier it says Maximum-Signal Plate Current for two tubes is 210mA. i.e. 420mA for 4 tubes.

[nsubulysses]

I appreciate the answers. If 210mA is maximum signal plate current for two tubes at 1.8% distortion I wonder what it does when the amp is cranked into heavy power amp distortion.

Now it gets me thinking further. How can a two power tube amp with only 450V on the plates and 400V screens do 55w? Is it because OT primary impedance is 5.6K?

I have a two power tube amp I can think of as example which has 495V plate voltage and 490V screen voltage and does 55W but with 4K OT primary impedance

[Dave H]

If 210mA is maximum signal plate current for two tubes at 1.8% distortion I wonder what it does when the amp is cranked into heavy power amp distortion.

How can a two power tube amp with only 450V on the plates and 400V screens do 55w?

210mA is for a full power continuous sine wave. I doubt it will average much more than that with a spiky guitar signal even overdriven.

I think they get 55W by using regulated power supplies for the plate and screen voltages.

[loudthud]

A fast and cheap approximation of full square wave current draw is 2x(clean pwr at clipping in Watts)/B+. If you are driving a resistive load like a cheap attenuator, you might want to add another 10%. But if you are driving speakers, that number is close enough for Rock and Roll. A power supply with that rating will run cool and be reliable. Most guitar amps have a power supply with an actual rating of 75% to 50% that number although they can output more current for brief times without immediate failure. They run a little on the warm side in guitar amps in actual use.

[bob p]

210mA is for a full power continuous sine wave.

When I look at how much supply current is needed for driving the power tubes I like to add the plate current and the screen current when the total isn't already listed on the data sheet. Using the GE data sheet snippet in post 3, that'd be 210 + 23 = 233mA for a pair of tubes.

Regarding the OP's question about sizing a transformer: is your question based on how to measure current in an existing amplifier, or how to select an appropriately sized transformer for a scratch build? I thought it was the later.

[nsubulysses]

yes I am asking how to understand this spec in regards to choosing appropriate transformer for a build. Of course it would help in repair scenarios too because then I could just read the specs and pick a transformer, rather than just buying a "Fender Twin Reverb" or whatever replacement from whatever manufacturer and not knowing what it actually is.

I guess the easy thing would be to crank the amp into a load and see what the current draw is on a meter. And crank the square wave as LT suggests for good max current draw scenario

There's a few places that come to mind where I could measure this. I wonder what would be best. Would it be in series with one of the OT primary leads that goes to two power tubes or in series with B+ between B+ and OT primary CT? Or perhaps in series between rectifier and B+ node for the whole shebang (where lots of standby switches are placed)?

[bob p]

spec'ing a PT used to be a lot harder than it is today. In the old days you might be flying blind, trying to use tube data sheets to determine what sort of generic Hammond to use as a replacement if you couldn't get an OEM part. Today you've got ready access on the internet to OEM replacement transformers from F/M/V, as well as third party iron that's specifically spec'd as replacement iron for F/M/V amps.

That Hammond is now making direct replacements for FMV type amps further simplifies the job of picking the right Hammond -- they even offer detailed spec sheets that tell you the exact specs of their Twin Reverb repro iron. That pretty much eliminates the guesswork.

Want a 100W Twin Reverb PT? Just read the power specs off of the Hammond 290FX: 335-0-335 @ 518mA, 6.5VCT@5.5A.

https://www.hammfg.com/part/290FX

That compares favorably with our calculated value of 2 * (210 + 23) = 466mA for the power tubes, leaving 518-466=52mA to accommodate the preamp tubes.

Regarding your question about where to measure: I'm thinking where you decide to take the measurement depends on exactly what you're trying to measure -- whole shebang or an individual device.

[Dave H]

When I look at how much supply current is needed for driving the power tubes I like to add the plate current and the screen current when the total isn't already listed on the data sheet. Using the GE data sheet snippet in post 3, that'd be 210 + 23 = 233mA for a pair of tubes.

yes, we need calculate the total amplifier current including the PI and pre-amp current but that only gives us the total DC current. To size the PT we really need the AC rms current which is easier to simulate than calculate. Hammond has a handy crib sheet to make it easier to guesstimate the rms current when the DC current is known but it's only an approximation. I have some sims somewhere for my 18W amp. I'll see if I can find them.

Hammond Design Guide For Rectifier Use.pdf

[bob p]

Dave, that's a handy document! thanks for linking it. My copy is on the inside of the hard copy hammond transformer catalog. i use it all the time. i tried looking for the online version sometime last week but didn't find it.

as far as simulation goes, here's a thumbs-up for PSUD. i used to use it all the time when i had windows, but when i moved to linux trying to run a windows program turned out to be more trouble than it was worth.

now that hammond is associating amp names with their iron i've learned to get by without using PSUD, by picking something that i know will be close. if i'm building an oddball amp that requires me to use the other types of hammond iron, or if i'm going to repurpose some junk drawer piece of iron that's not got proper specs on it, then i'll do a complete simulation using the transformer winding impedances. to do this i have to bite the bullet and put together a temporary windows box to run PSUD. i need to find a better solution...

anyone know of something that works for linux? i'm not a spice user.

[cjenrick]

stick AC current meter in B+ line , crank up amp with stomp box included, measure max current.

power transformer ratings can be all over the place, they can play games with temperature rise, regulation, duty cycle,

Hammond is probably the least conservative of the bunch, Triad is usually pretty good, pull a transformer out of a piece of HP tube equipment if you want something that runs very cool.

[eschertron]

Would I be out of line to propose that a PT with a spec like 800V CT @ 535mA which results in a total HV power of 428VA shakes out this way:
The 428VA must be greater than the sum of
audio output power (nominal max output power of the amp, so say 200W)
tube idle dissipation (100% design limits won't be far off, so say 180W for 6x6L6)
preamp/PI power consumption (a couple tens of mA at 300..400vdc, so say 15W)
additional losses (heat - other than tube dissipation - and audible noise from the trannys, a couple % of the total, say 8W)
So total VA 428 >= 200 + 180 + 15 + 8 (403W)

The biggest load on the PT high voltage winding besides the desired output power is the heat losses of the power tubes. Estimating the losses from the tubes at the design limit seems to be a 'safe' guess. And this analysis presupposes worst-case operation; Amp cranked, continuous maximum power output. Note that I can't build this amp without doing a current/power/voltage analysis of each individual diode, resistor, etc., to ensure that their ratings aren't exceeded, but I can choose the PT to meet my design criteria.

edit: and of course, I've looked at that PT because the B+ from whatever rectifier I've chosen is the right voltage for my application. I'd guess that is the first design choice.

[cjenrick]

i believe that 800 V center tapped at 535 ma means VA = 400 x 525 ma as they predict two diodes on outer legs feeding cap with CT grounded?

so 400 x .53 = 212 VA ?

[Dave H]

i believe that 800 V center tapped at 535 ma means VA = 400 x 525 ma as they predict two diodes on outer legs feeding cap with CT grounded?

so 400 x .53 = 212 VA ?

From the Hammond catalogue it looks like they specify the current for the whole winding. If they spec 800V CT @ 535mA it's 428VA. Adding up the VAs for the 278CX (see below) I get 800 x 0.535 + 6.3 x 6 = 466VA which is close enough (2.5%) to its 454VA rating (some of the others are exact).