Voltage doublers and DC filament supplies.

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

sircletus

Well-known member
Joined
Nov 19, 2008
Messages
165
Howdy again, all.  General question regarding the above.  I've got a tube project I'm working on.  I'm fleshing out the schematic, and will ultimately be putting this on PCBs.  That said, DC-heated tubes is going to be the way to go.  Power supply-wise, I had a few things in mind:

1.)  Hammond 370CAX (500V B+, 5V @ 2.5A, 6.3V @ 2.0A) as power trafo.  I was going to use the 5V secondary for indicator LEDs, and rectify and voltage-double the 6.3V secondary for heaters.

2.)  Same trafo as above, only buy a separate, purpose-built trafo just to spit out 12.6V for the heaters.  Hammond makes a suitable one for about $20 from Digikey.

Any opinions on which would be the "better" approach?  And any good references for the voltage doubler?  I have a general understanding of how they work and I've seen them implemented in various circuits, but I'm unsure of the "proper" way of approaching one myself.
 
AC heating can be better than DC.  DC heating needs to be designed correctly to be low noise.  Voltage doubling might be nosier than well done AC.  THE PS DESIGN DETAILS count.  I suggest you read all you can find on DC supplies to understand what matters and how to design one.
 
Put the 5V and 6V in series for 11VAC, go from there.

The 0.6V-1.0V drop per diode is distressing and wasteful with a mere 6V, slightly less so at 11V.

How many how hungry heaters? With DC heat you usually want higher voltage, run heaters in series. Peavey tends to run 24V or 30V heater strings, for good reason.

You can find a way to light LEDs a lot easier than you can heat bottles.
 
PRR said:
Put the 5V and 6V in series for 11VAC, go from there.

The 0.6V-1.0V drop per diode is distressing and wasteful with a mere 6V, slightly less so at 11V.

How many how hungry heaters? With DC heat you usually want higher voltage, run heaters in series. Peavey tends to run 24V or 30V heater strings, for good reason.

You can find a way to light LEDs a lot easier than you can heat bottles.

Funny you should mention that, PRR.  My original thought was to run the 5V and 6.3V secondaries in series, but I figured 11.3V probably wasn't high enough, and I'm a bit worried about current handling.  I'm talking about 8 tubes: 4 x 12AX7, 2 x 12BH7, 2 x 6AQ5.  That's right around 2A @ 12.6V (The Hammond is rated at 2.5A on the 6.3V secondary, and 2A on the 5V).  If I had taken the 5V + 6.3V approach, my thought was to regulate the DC up slightly with a 2-Amp-rated 78S12.

It's the number of tubes and the current they're drawing that's got me leaning toward a separate filament transformer with a higher current rating.

Why the recommendation to run heaters in series?  What are the advantages of doing that?

FWIW, I did a PCB layout for a "proper" EA Ampex 351 about a year ago with 12V DC on the heaters in parallel and it's quiet as a church mouse and works like a champ.

Any more thoughts?  I guess I'd like to know which solution is:

a.) Most efficient
    -but-
b.) Still reliable


Sidenote:  Yes, two LA-2As in one box, yes I know there's a forum for that, but I'm more interested in doing this from the ground up as a learning exercise.  The 351 was easy 'cause we had a box of original transformers to work with and it was really just about a new PCB design.
 
im with gus. low voltage ac heaters work better. i reckon the valve gets hotter. and there are good easy enough ways of suppressing any hum. the best ive found is to first cut off the heater wiring earth reference. isolate the heater circuit from earth/chassis/ground. the circuit will buzz like crazy . then fit 100 ohm wirewound pot across the ac wiring. then apply a dc bias (only 'potential' volts) of about 30 - 50 volts to the wiper. (pays to check valve data for maximum heater-cathode voltage spec) get that bias by dividing h.t with 270K . &.. 47K to ground with 1uf-100v across the 47k.  adjust the pot for for minimum hum and magic. hum gone. its especially effective if the  pot is wired directly across the highest gain amp stage that is bleeding the most hum. i think the theory is -  that applying the dc potential means that the 50 hz signal current wont flow across from the heater element onto the cathode structure if the element always looks positive in respect to it.  try it maybe. ive had good success with this very old school tricky method.  
 
Back
Top