A better LDO heater regulator
 

Putting 6.3vac through a bridge and filter caps doesn't give enough voltage for a normal LM317 regulator to work properly. I'm trying to power two 6DJ8s for a phono stage and even with a 5600uf filter cap the ripple is around 200mv. So I started looking for regulators and found this one

http://www.electronic-circuits-diagr...liesckt3.shtml

None of the parts look too exotic, in fact I believe a scrapped computer SMPS would yield most of them (sans crowbar protection). Anyone care to comment?

Two things - the LM1084/5/6 has a a lower drop out and higher current rating than the rather ancient LM317. There are many many other good LDO linear regulators, and some encapsulated SMPS drop-in equivalents as well (which run at 100s of kHz and will dissapate far less).

I'd avoid the 555 version of the circuit as it'll introduce all sort of c**p, either directly onto V+ or via EMI.

Dude,

IMO, you're banging your head against the wall without cause. "Full wave" voltage double the 6.3 VAC using Schottky diodes ("noiseless" and low forward drop). That gives you 16+ VDC. Regulate with a low cost 7812. Wire the 6922 heaters in series.

Re: A better LDO heater regulator
 

aditionall 15V/1A xformer is certainly too big and expensive for your new toy?
along with one Graetz,cap ,two LM317 (one per tube as CCS) and few resistors and small caps.................
:devilr:

ps-use doubler with 6V3 xfomer only if your mains voltage is rock stable.

Ah but the point is to use parts I already have :) The doubler schottky idea sounds good. The winding is good for 2 amps so no worries there. I will also put a 1N914 in series with the ground leg on the 7812, it should output 12.6v then.

The schematic is from RJM Audio

http://www.geocities.com/rjm003.geo/.../diy_pho3.html

Re: A better LDO heater regulator
 

The other suggestions are good ones. But another simple method is to cascade a few RC sections with something like 0.2R per section. Play with PSUD and see what you get. Something like 20 mV ripple should be no problem. Use shottkys.

Sheldon

That's exactly what you should not do. Using a 12.6 V. O/P puts you back into an inadequate head room situation. The drop out bugaboo will resurface.

12 V., "on the dot", is inside of a 5% deviation from nominal, which is well within tolerances. 6922 data sheet here.