Basic heater Q

[SteveH]

All,
Ok - I cannot believe Im asking this...but....For rectifier tubes, say, an 816 (2.5v@2A), if I connect them in series, the draw would be 5v@2A right ? My logic is w=V*A.....

Right ?

Thanks !

[mesherm]

That is correct. In series you would need 5 volts @ 2 amps. In parallel you would need 2.5 volts 4 amps. Watts=volts x amps Volts=amps x ohms

[SteveH]

Thanks a bunch ! Im not quite sure where my mind is lately....

[TomMcNally]

Steve -

With directly heated rectifiers like that (with no cathode) you typically use the center tap of the filament transformer as the DC output. I'm not sure that would work right with two tubes in series.

Anyone else have any thoughts ?

... tom

[SteveH]

Tom,
Check this file out; it give connection schemas for all sorts of hookups:

http://www.mif.pg.gda.pl/homepages/frank/sheets/081/8/866A.pdf

Im not quite sure how to make it all work yet...Its actually going to be a B+ power supply for a preamp design I have. I figured I'd do something different, yet matching the 83 tube on my ST35...If you like, I can forward you the PSU designer file; but it isn't anything super special.

Thanks !
Steve

[erichayes]

Hi All,

Yep, it'll work; in fact, it's the preferred way to do it, even if you're running half-wave. The center tap should probably be connected to the filament midpoint, to act as an equalizer for any heater current differences. If the transformer isn't center tapped, you can pull the DC off of either end, or through a phantom center tap composed of two resistors in series across the winding (like a hum balance network for AF tubes). Directly heated rectifiers have been around for eons, but nobody's developed one with a center tapped filament. That fact would make me hesitant to use the filament midpoint to draw off the DC. I think I'd wire up a couple of 5Y3s, or whatever I had the most of, and do a little experimentation before I tried it with anything of value.

[TomMcNally]

I agree - there should be no problem even with the series filament string. I like those old data sheets - very cool. Be sure to let them warm up for 15 minutes to 1/2 hour the first time they are socketed, so the mercury can heat up properly. After that, 30 seconds is good.

One of the old radio stations I take care of has a mercury vapor tube warm-up jig on the shelf - just a socket and transformer on a piece of wood. We just donated the 60 year old transmitter to a museum. The backup now is a modern 1961 vintage ! It has mercury vapor sockets, with solid state replacements in the sockets. Yellow_Light_Colorz_PDT_21

[SteveH]

Gents,
Ok....I *think* I have an understanding now ?



Whereas :
Transformer - 274BX (375-0-375, 5v@3A)
V1, V2 - 816 (2.5v @2A)
V3, V4 - 0D3 (VR150)
L1 - L5- 2H, 50ohm
C1 - 3.3uF
C2, C4, C6 - 100uF
C3, C5 - 47uF
R1 - 1.25k ohm

With an 80mA load, it is kicking out 310.21v, with a flutter of 42.140nV in PSUd. If I can even get close to that in real life; this should be an excellent psu. Notice though that I had to use an 83 tube in PSUd as it doesn't have the specifications for an 816. However, Im pretty sure it will be close. Im designing this PSU to power my preamp and phonostage...

Thanks again,
Steve

EDITED, thanks to Tom :D
EDITED II - Update..Better/cheaper design :D

[TomMcNally]

Steve -

Your schematic isn't quite right - you want the center tap of the rectifier filament transformer to be the DC output that goes to the filter stage. Obviously, you'll put the filaments in series, not parallel. The center tap of the hight voltage transformer gets grounded. If you don't have a center tap, you can't use this circuit.

... tom

[SteveH]

Tom,
Ahh, again Im not sure what I was doing there...In any event, I replaced the old one with the new (above). Im pretty sure this is what you were telling me.

Thanks !
Steve

[TomMcNally]

You got it now Steve !

... tom

[SteveH]

Gents,
Ok, I have a final design which I am much more happy with. See the image above, I have edited the post to reflect the changes. This will also be a much cheaper PSU to build, as 2H 50ohm chokes are much cheaper than the 20H ones I was planning on using before !

Regards,
Steve

[TomMcNally]

Hi Steve -

Five chokes ! Wow ! Most commercial power
supplies have one or none, maybe two in a high
end amplifier. That DC should be ruler flat.

... tom

[SteveH]

Tom,
Well, I guess the question is what is an acceptable flutter level? From reading elsewhere, .5mV is the dividing line for phonostage psu's. I could get that with 3 chokes of the 2H value, but my initial ramp-up looked like crap. This way is about as smooth as I can get for both ramp-up and steady power output. Mainly, it is due to using the Hg rectifier tubes I believe. If I used like a 5V4 or something like that, one choke would get the job done. So much for form following function on this one........... :D
I won't be building this for a bit yet, as I still have to complete my ST35 (The pebble that rolled away from Sisyphus's stone....). I am about 75% done with the enclosure (I just need to cut the holes in it for the jacks, and stain/finish it), and then I need to cut my top plates. I still need to acquire some iron as well....

[TomMcNally]

I kind of just mess with things, not too much computer design.
I was looking at Hagerman's Cornet phono preamp, which is
pretty highly regarded, and he has no chokes at all.

http://www.hagtech.com/cornet.html