problem with dc filament supply

[spy]

I tried it again at 5Vdc and all the highs dropped down! Those heaters need their 6.3V!

Will a low drop regulator help? I have 7.5V at the input. Or just get rid of it?

[J M Fahey]

Personally I would drop any regulator, and use:
6.3AC > Bridge > Cap (say, 4700uF) > filtering/dropping Resistor (a couple Ohms, 5 or 10W) > Cap (same as first)
Use a dropping resistor value calculated to get around 6.3V at filaments.
Resistors do not have a "minimum voltage drop" as regulators.
Regulators need at least 3 (or 4) V across them to work, but a resistor will happily drop 500 mV , 1V, 2V, whatever you need.
No diode junctions inside

[J M Fahey]

Personally I would drop any regulator, and use:
6.3AC > Bridge > Cap (say, 4700uF) > filtering/dropping Resistor (a couple Ohms, 5 or 10W) > Cap (same as first)
Use a dropping resistor value calculated to get around 6.3V at filaments.
Resistors do not have a "minimum voltage drop" as regulators.
Regulators need at least 3 (or 4) V across them to work, but a resistor will happily drop 500 mV , 1V, 2V, whatever you need.
No diode junctions inside

[spy]

I took it off!

The psu already has 3x 4700u in parallel then the regulator and then one more 4700u! I replaced the regulator with a 2 ohm 2W resistor and the filaments are 6.1V, I think it's perfect! Ripple around 80mVpp with load.

[J M Fahey]

Cool!! That's the idea !!!

[olddawg]

I took it off!

The psu already has 3x 4700u in parallel then the regulator and then one more 4700u! I replaced the regulator with a 2 ohm 2W resistor and the filaments are 6.1V, I think it's perfect! Ripple around 80mVpp with load.

Couldn't you just use a forward biased diode (or a couple) to drop .7v increments, or even better a Zener diode and resistor to ground? You could even make that adjustable if you wanted.

[spy]

It's already done and the amp is in his headshell!

Yes, you could add a diode or two to drop the voltage but with a resistor you have a pi-filter and helps a little with the ripple.