Specing a power transformer

[MWJB]

We can asume that Overdriven is not going to be running those 12AX7 at 350 on the plate, with a 430v B+, it's extraordinarily unlikely.

"Sometimes you have a 220k resistor that vill make the current smaller." or a larger cathode resistor than say 1.5K, which is pretty common in amps with 220K or larger plate resistors.

[Tubis]

Not 350 on the plate but on the B+ feeding the tubestage and that resistor. Then we calculate what in worst case can pass there to ground and that is if the resistor will be shortened to ground. No more current can be drawn here. With a 100k Plate resistor that is 3.5mA if there is a stabile powersupply.

[Alex/Tubewonder]

All these calculations are OK for an amp with a power stage that you don't expect to drive into distortion/saturation. If you're designing an amp that will be used at full power tube saturation you need to design the B+ supply so it can deliver peak plate current continuously, not average. Calculate HV windings accordingly.

"You need to pull 200VA from HV windings in a 50W amp? You're kidding me!" - said a very surprised manufacturer of highly regarded transformers.
I wasn't.

[MWJB]

OK but I'm not sure where this 350v B+ supply comes from, we don't know anything much else about Overdriven's design other than he wants 430v at the 6L6 plates & 4x12AX7 type tubes.

The way I would look at it is this:

Max B+ available is 430v from Overdriven's post (it will be less after choke & preamp dropping supply resistors, but we don't know how much until we know plate voltage being aimed for, role & type of the "12AX7 type" tubes in question).

Smallest likely plate resistor values are 100K (possibly 56K/47K but we'll stick with the much more common 100K), smallest likely cathode resistor is 820ohms/triode. Likely voltage at the 12AX7 plate in this scenario is in the 220v region, so current would likely be in the region of 2.1mA per triode at idle. Maybe 2.7mA or so for a 12AY, 3mA-ish for a 12AU?

Your short to ground scenario becomes 4.3mA, but obviously if this happened the amp would have to be switched off & fixed.

[Steve Conner]

I budget 1.5mA per preamp triode, which amounts to the same thing as ignoring them.

And in the light of the results I posted earlier, Alex's 200VA figure doesn't surprise me either. I calculated it from power tube cathode current, so that doesn't include any other part of the amp.

[MWJB]

Hammond 260K seems to fit the bill...perhaps we should have kicked off with a specific suggestion?

[Tubis]

I assume that hes not aiming for 350V on the plates. Thats very high and above the maximum. It will problaby work anyway but.......
And if he means 350V on the plate that means the tubestages all must have the same plate voltage. So i draw the conclusion that he meens the Preamp B+ at 350V. Why would anyone want 350V on the plates of the preamp?

[MWJB]

"Why would anyone want 350V on the plates of the preamp?" Cathode follower? PI tube? (essential caveat - I mean voltage to ground at the plate rather than from plate to cathode).

B+ supply at the PI could well be over 350v, of course from then on you can add a dropping stage from one tube to the next or even from triode to triode.

[Tubis]

"Why would anyone want 350V on the plates of the preamp?" Cathode follower? PI tube? (essential caveat - I mean voltage to ground at the plate rather than from plate to cathode).

B+ supply at the PI could well be over 350v, of course from then on you can add a dropping stage from one tube to the next or even from triode to triode.

I think you know what i mean here. Ofcourse you could have 350 on the plates when you have a big cathode resistor. Must i spell out everything? I think you understand what im saying. I calculate every stage separately Why calculate a cathode follower with a 56k resistor the same as a plate follower with 220/2k7.
Thats just.......

[MWJB]

I'm not being contrary I am just pointing out in terms that a novice who is designing an amp can understand (the point of this thread, remember) & what they may expect to find when they put a meter at the 12AX7 plates.

A 12AX7 in the PI could concievably have >350v from plate to cathode. Perhaps less likely in a 430v B+ amp.

We have different ways of working things out.

Typically, I'd start with anticipated power tube plate voltage, then what kind of voltage drop & plate voltage for the PI (voltage from plate to cathode & accross tail resistor will vary with values here - I know that YOU know this Tubis, I'm saying it for the benefit of Overdriven/anyone else who wants to learn to spec a PT/power supply), then work towards the input stage.

I would suspect, or hope, in the absence of further details, that Overdriven already has a basic circuit design in mind and ball park values for plates & cathodes.

[Tubis]

I got 145VA at full blast with 5,6k primary. Could that be right? Alex?

[Alex/Tubewonder]

I got 145VA at full blast with 5,6k primary. Could that be right? Alex?

How and what did you measure? Input signal: guitar, sine, square wave or bandwidth limited pink noise? B+, what tubes, what is "full blast"? Your amp at "full blast" could be mine just revving up
Not enough data to compute answer.
A 50W amp sucking 145VA off HV secondaries seems close to ballpark.

[Tubis]

How and what did you measure? Input signal: guitar, sine, square wave or bandwidth limited pink noise? B+, what tubes, what is "full blast"? Your amp at "full blast" could be mine just revving up
Not enough data to compute answer.
A 50W amp sucking 145VA off HV secondaries seems close to ballpark.

I`m afraid to say how i calculated so show me how you would do it
Well sine wave into the P-tubes and the 450 B+ at a bandwith of 80 to 16kzH. But the outputtransformer will ofcourse put a stop to the bandwidth when one reaches its limit Full blast is of course the headroom limit. Squared clipping maybe?
I never do much output stage overdrive as you know. Us metalheads want it all in the preamp

Hm. maybe pink noise? Sounds more like a shredding guitar.

[loudthud]

I do it something like this. If the tube manual says a given set of tubes and OT will do 50W, it will do something like 60W at hard clipping. When that power amp is clipping a square wave, it will deliver about 120W (double the sine wave watts) to the load and the efficiency will be about 90%. So it will draw about 132W B+. 145W is not totally unreasonable.

132W/430V=306mA

Here's where a little experience comes in handy. In this case a 300mA transformer will see its max only at full clipping and will coast along the rest of the time. A 250mA transformer will get the job done and probably never fail. A 200mA transformer is going to run hot and might fail at a gig. So depending on what's available, I'd choose the 250mA or 300mA.

[Tubis]

I do it something like this. If the tube manual says a given set of tubes and OT will do 50W, it will do something like 60W at hard clipping. When that power amp is clipping a square wave, it will deliver about 120W (double the sine wave watts) to the load and the efficiency will be about 90%. So it will draw about 132W B+. 145W is not totally unreasonable.

132W/430V=306mA

Here's where a little experience comes in handy. In this case a 300mA transformer will see its max only at full clipping and will coast along the rest of the time. A 250mA transformer will get the job done and probably never fail. A 200mA transformer is going to run hot and might fail at a gig. So depending on what's available, I'd choose the 250mA or 300mA.

I just calculated the worst case swing over the primary. I think that will never be obtained so your way i think is more close to real life