DC filaments for 6550s

[richwalters]

Have you got any gain reduction when applying the 510K ? With the GEC circuit such a high value resistor has little or no effect. Something still isn't right.
BTW.If the o/p tranny primary was wrong-way-round the amp would howl in protest with a 5K1.

richy

 

[troposnet]

yes i could do with the 5K1 but i know the sound will be dull (less treble)...pretty much same as 4K7. in fact anything less than 100k seems to have bad treble. i use those values to cut the hiss im hearing and very little gain reduction. the primary wiring is in phase so no howling noticed whichever value used. i need to strike a balance between hiss reduction and treble loss here.

i was wondering if i used 5K1 on the NFB and played with the 6550 grid resistor values, would i get back the treble i lost in the nfb? the grid is now at 10k. or maybe the damping cct between the AX and AU needs a tweek. what do ya think?

 

[richwalters]

As SY mentions,the o/p stage grid stopper values should have no effect on treble and playing with grid stoppers in an attempt to alter the treble is quite wrong.. Seems you have a serious component misplacement, have you double checked against the circuit diagram ?
I have a slightly modified GEC 88-50, the 5K1 is the correct value for 8 Ohm LS (20dB globbal nfb)with absolutely no loss of treble. In fact global nfb actually extends the top end range. Your input stage + phasesplitter should be pretty close to the enclosed jpg. I used a 6SN7 (better sounding) instead of ECC82 types.
richy