That 300B Fi Primer amp again...

[Pedro]

While the PT15 takes shape I'm fiddling with the 300B now that its got its EH 300BGG's in.

After taking various measurements Ive come to the conclusion that it runs too hot anyway.

Ive replaced the 880R cathode power resistors with 1K's to see what effect it had on idling.

I now have the following :

Plates approx 452v and 460v
Cathodes 1 @ 71v across 1K cathode resistor and the other at 63v across 1K cathode resistor. So, Im dissipating about 27w on one side and 25W on the other. The valves seem very hot indeed to me.

Amp is sounding good but I wondered if I should back off the anode voltages somehow. (Most 300B amps Ive come across have approx 400v on the plates. )

If I do , whats the best way to drop some voltage when Im already using a Cap input power supply. The chassis is pretty tight so can I do it without adding chokes ?
I've read about using zener in mains TX centre tap but I'm worried about adding nasty zener hash to the signal. Also not sure what zener ratings to use ?

Any help ?

TIA
Pete

[NickG]

Raise the cathode voltage with extra cathode resistors, bipass them with caps, and reference the grid to the top of the extra resistors.

Hope that makes sense...

[Pedro]

yes indeed, great stuff - thanks Nick
btw, does it matter what value of bypass cap to ground ?

Rgds
Pete

[NickG]

I would try and aim for at last as large time constant as the real cathode cap.

The valve will see too small a cap as a poorly regulated supply. You could do the same in the supply to the OPTX, a resistor to drop voltage and a cap to smooth. The cap again becoming the last cap in your power supply. The advantage to doing it below the cathode is you can use a lower voltage cap.

This is assuming cathode bias of course.

[Pedro]

OK,

I have typical network of 1K and 100uf from the cathode to gnd at present, and about 65ma idling current. ( original cathode R was 880R )

I'm thinking of using 470R and 200uf to "maintain" same sort of RC time constant, and get about 30v additional drop

If I get you correctly - I just tap the grid resistor at the junction between the original and additional cathode resistors , with the new resistors going to ground. So, cathode to gnd ends up aprox 95v but grid sees -65 respect to cathode, and anode to cathode drops appropriately.

Just out of interest - is it feasible to have the bypass caps running from cathode to ground right around the tapped cathode resistor network ?
I realise I would have to use approx 60uf to maintain original 100ms time constant

cheers
Pete

[NickG]

Don't see why not. Anyone?

[James D]

Should be fine as the objective is to remove the signal current from the resistor chain leaving dc only through the resistor chain and as reference for the grid.

James

[Pedro]

thanks Nick and James,

while I lay my hands on some more caps, will it hurt to have an RC time constant of 147ms vs the original designed 88ms ?

i.e. in meantime I will use the existing 100u caps across the combined 1K and 470R resistor string. (watching for the increased voltage requirement across the cap of course )
cheers
Pete

apologies for all these questions - I'm learning more than ever before here.

[Pedro]

update...

had the soldering iron out for the last hour and made the necessary changes to grid tap the cathode resistor and drop 30v or so across the extra cathode resistor...

now have 101v cathode to ground. anodes at 450v.
32v on grid so effective grid bias at -69v respect to cathode.
349v plate to cathode across the EHGG300B's.

All sounding good......many thanks for the help

back to the PT15 amp now....

cheers
Pete

[Pedro]

hi again.....listening , and reading round SE theory last night I came across the ultrapath or WE connection for the first time.

I have a measured 9mv hum at my speakers from the 300B SE. Its OK, but it bugs me really.

Would fitting suitable caps between cathodes and output TX help. Glass Audio article suggests it would. I see that injecting out of phase PS noise will help cancel some out.

I have 100uf bypass caps on the 300B's - is there any rule of thumb for the cap value ?

I'd like to try this out. Amps are never finished ( even when youve started on another better one )

cheers
Pete