BIG tube amplifier, two qbl5/3500 tubes

[tonskulus]

I have some nice transmitter tubes (two brand new qbl3/3500 tetrodes etc.) and I was thinking if it is possible to use these as hige audio hi-fi amplifier?

Class A - lots of heat! But not too hard to built..No need for center tapped output transformer.

Class AB - Enough output power. Much better efficiency. Needs huge center tapped output xfrmr..

What comes to powersupply, I have xfrmrs like 2kV/0.1A....pair of 6kV/1A pigs. So that is not a problem.

Filament supply - Seems to be difficult as those tubes requires 6V/32A each..And it should be well filtered DC? big rectifier diodes and huge capacitor bank. Propably switching mode psu will do the job.

Anyway, I was considering to use my 2x100W amplifier for driving these tubes via coupling transformer. So there is no other tube stages, only power stage.

Any good ideas for this kind of amplifier? I know lots of people using RF power amp tubes as hifi amplifier finals but those are using much smaller power tubes.

[SY]

Before going any further, let's clear up something. When you say class A has "No need for center tapped output transformer," you must be referring to single-ended operation. Indeed, there is no need for a center tap, but... there IS a need for a gapped core, and that means that for a given power, that transformer will need to be much bigger and more expensive than the equivalent center-tapped transformer for push-pull (regardless of class of operation).

[tonskulus]

Well, that was only grazy idea so I know that there is much to learn about. I have made many smaller tubeamps like 1..100W output/4x EL34 but this is a way different.

So, I need gapped core output transformer..hmm, I have no idea. Why is that? I haven't seen before airgapped output transformers..

[SY]

There are some work-arounds to the gapping requirement, but they carry their own baggage.

In any case, to supply the plate, DC must flow through the transformer. The magnetic field from the DC will take the core toward saturation. In a push-pull design, the DC-induced field is canceled out by the current in the two haves flowing in the opposite direction. This is not the case for SE, so the gapped core is generally used.

[bear]

SY,

We'll go parafeed...

But, I'd not play around with extreme high voltage unless you have significant experience with it.

HIGH VOLTAGE AT MODEST CURRENTS IS INSTANTLY DEADLY!!

HIGH VOLTAGE AT MODEST CURRENTS IS INSTANTLY DEADLY!!

And then there is the question of where you'd get an output xfmr with a high enough primary Z that has any freq response worth a damn, and a size that is comensurate??

There's probably a reason that this isn't done much...

I'd take a really hard look at plate curves, (assuming a pentode - strapped into triode), with reduced plate voltage and run in class A, and see how low you can go with the B+ and how low you can get the output Z to fall... then you have half a shot, imho...

_-_-bear

[Bandersnatch]

Hey-Hey!!!,
I saw an output for a SE TL304. The 304 runs 25A of 5V for its filament. The gap and size of the output TX was reduced by cancelling a lot of the idle flux with the DC for the filaments in a separate winding. A 32A filament should do even better...
cheers,
Douglas

[Miles Prower]

Originally posted by tonskulus
I have some nice transmitter tubes (two brand new qbl3/3500 tetrodes etc.) and I was thinking if it is possible to use these as hige audio hi-fi amplifier?

Sure, anything's possible, but that's not really the question. Have you checked the specs on that? Why would you ever need 7200W of audio power? Let's not forget that, at the stated efficiency of 65%, you will be needing at least 11KW of DC input power, and that's not accounting for efficiency losses on the AC side.

Are you planning on having your power company install your own, personal, three phase substation? Because you're sure gonna need one.

I'd say FUGGEDABOUDIT, sell those to the VHF contest fanatics with a burning desire to win contests, way too much time and money on their hands, and no conscience, who already build illegal VHF rigs for winning contests that provide nothing more than braggin' rights.

You could use the proceeds to build a few really decent -- and practical -- HiFi rigs.

[Miles Prower]

Originally posted by bear
I'd take a really hard look at plate curves, (assuming a pentode - strapped into triode), with reduced plate voltage and run in class A, and see how low you can go with the B+ and how low you can get the output Z to fall... then you have half a shot, imho...

That won't work either. The Vsgsg max is 800V. It's a true tetrode, not a pentode or beam tube either. For any Vpk below 1000V, the plate curves kink badly. 800V would put that right in the middle of a neagtive resistance characteristic that's far from linear, and would result in uncontrollable instability.

[Mark A. Gulbrandsen]

Check with the Tubelab dude. He's fairly impossible to get ahold of as he's very busy launching Space Shuttles or something like that but he might be of some help to you. Check this out!

Mark

[rdf]

Originally posted by Miles Prower
...at the stated efficiency of 65%, you will be needing at least 11KW of DC input power, and that's not accounting for efficiency losses on the AC side....

On the bright side a rectified 3-phase power supply would be easier to make ripple free.