All of the schematics I have seen have a much lower voltage on the screens. Why is this necessary?
My question is about the motive to have guitar amp using tubes MOST designs do not use for guitar?
It's not in your case. If you buy a 12-cylinder Ferrari engine, would you be happy paying all that money if it were hamstrung to yield only 120HP in your car?
The KT88 were big, expensive imported tubes, capable of big power output. The data sheet claims the tubes can output 100w per pair or 200w per quartet, and with designs not typical in most guitar amps, you can certainly do that. But regarding the "typical guitar amp tubes" question, you
could make a 60w guitar amp with 6x 6V6's, but that's something you won't see because it's easier to use a pair of 6L6's to get close enough.
And since 4x 6L6 (for 100w) would probably have been easier to buy in the drug store and most people don't need a 200w guitar amp, you don't see KT88's much
in guitar amps. But that does not mean there is any reason you can't use a pair of KT88's for guitar.
So let's assume you were an amp maker and used KT88's to get every last bit of that 100w/pair, 200w/quartet. You would keep raising the plate voltage to increase output power, but you run into the problem of not being able to keep the screen at the same voltage as the plate. Say screen voltage stays at 450v while you raise the plate voltage. You might find you hit your 100w goal at something like 550v (for example).
It would be awkward to derive 550v and 450v supplies that were both reasonably constant, so an easier path is to make the screen voltage half the plate voltage. You could have a power transformer with a bridge rectifier and series input filter caps, with the 300v tapped from the lower filter cap and 600v from the upper filter cap. The lower screen voltage reduces plate current and offsets the increased plate voltage when the supply voltage was raised to allow the easy power supply setup.
Ultimately, the plate voltage (voltage swing) and screen voltage (current swing) are selected to attack each part of the equation for power, though economics often push them to be similar.
Anyway, copy the datasheet condition for fixed-bias and 70w output.
Would there be much different in cathode bias as opposed to fixed? If it is fixed, it seems like all the HiFi designs use dual bias. Don't know why. Anyone?
You'll need fixed bias, unless you want to start using uncomfortably-high supply voltages (which will increase costs as you need more power supply components, caps, etc). You should also see from the datasheet that for similar supply voltage you will get less output power than in fixed bias (mostly because of the change in plate-to-cathode voltage).
Genalex advised independent cathode resistors and independent bias adjusts so that each tube could be given exactly what it needed. I think they also wanted to make sure a fault in one tube didn't wipe out the whole output stage.
For you & me, this is an easy way to not need to worry about perfectly matched tubes, as the twin bias adjusts allows each tube to be dialled in to exactly the same idle current even if they're grossly unmatched. Power output will be reduced somewhat for
very mismatched tubes.
