Ad Widget

Collapse

Announcement

Collapse
No announcement yet.

DC elevation of heaters: what size cap, and how many volts?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • DC elevation of heaters: what size cap, and how many volts?

    I completed my AC30 inspired amp yesterday, and I decided to go with DC elevation for the heaters instead of just referencing them to the top of the cathode bias resistor (because of the cathode-lift standby scheme, it would cause problems).

    I used what I had on hand and ended up with a 220k/68k divider (both 2W) which gives me about 75V of elevation (jumps to 90V on standby).

    For the cap across the 68k resistor, I didn't have any small electrolytics left, so I used a .1uF poly cap I had left.

    Should the cap be bigger than that? There is almost no current going through there and it's a constant load, I'm not even sure I understand why the cap is necessary.

    As for the elevation. Is the 90V on Standby too much? The preamp tubes are still working on Standby.
    I see the cathode follower stage has the cathode floating at 134V on idle, so DC elevation is really not a bad thing here!

    What happens if the filaments are floating too high? Will they start acting like the anode and start drawing electrons?

  • #2
    I should mention that I connected the top of the voltage divider to the screen node, not the first node. So this is on the other side of a 15H choke and linked to a 16uF cap, should be pretty steady voltage.

    Comment


    • #3
      Just did my homework

      EL84 maximum heater to cathode voltage: 100V (positive or negative)
      12AX7 maximum heater to cathode voltage: 180V (positive or negative)

      So all is good I guess!
      Just wondering if I should increase the cap value now...

      Comment


      • #4
        people over at sloclone make super high gain amps and many love damafx's elevated (DC!) heater design:Sloclone Forums • Login

        he uses 47uf on the HV input (of course 300x that on the DC)

        hmm these guys (ac30 types) seem to favor 100-220uf...

        http://forum.metroamp.com/viewtopic.php?f=4&t=16228
        http://forum.metroamp.com/download/file.php?id=5062

        Comment


        • #5
          Originally posted by Hardtailed View Post
          For the cap across the 68k resistor, I didn't have any small electrolytics left, so I used a .1uF poly cap I had left.

          Should the cap be bigger than that?
          Yeah something between 8uF and 200uF or so is 'better' - more ability to flatten out any wriggle

          Originally posted by Hardtailed View Post
          There is almost no current going through there and it's a constant load, I'm not even sure I understand why the cap is necessary.
          Its desirable because without it, you are more likely to get B+ 'micro-wriggle' which flows back through the voltage divider and plays havoc with the heater ground refence point, introducing noise into each stage.


          Originally posted by Hardtailed View Post
          As for the elevation. Is the 90V on Standby too much?
          90V is optimal. Wouldn't go higher than that tho'

          Originally posted by Hardtailed View Post
          What happens if the filaments are floating too high? Will they start acting like the anode and start drawing electrons?
          The heater-to-cathode insulation coating breaks down, and there is arcing between the filament and the cathode, and then the cathode and heater short and ...
          Building a better world (one tube amp at a time)

          "I have never had to invoke a formula to fight oscillation in a guitar amp."- Enzo

          Comment


          • #6
            I use a 470K/47K divider off the screen node and a 10uF filter cap to decouple the 47K. On a 450V supply this gives a 40 volt reference.

            10uF is all you need. I tried increasing it and heard no difference in noise. The heater's not actually drawing from that filter cap so since the cap isn't really having to source anymore current than what the 47K is pulling from it (about 1mA roughly), it doesn't have to be all that high.

            The 470K will need to be 1 watt or greater. You can use a 1/2 watt for the 47K.
            Jon Wilder
            Wilder Amplification

            Originally posted by m-fine
            I don't know about you, but I find it a LOT easier to change a capacitor than to actually learn how to play well
            Originally posted by JoeM
            I doubt if any of my favorite players even own a soldering iron.

            Comment


            • #7
              I typically elevate between 60-80V mostly to bring the heater/cathode differential closer for CFs and LTP's, but it doesn't take much honestly 12V or so to be just as effective regarding noise. as for the cap, I usually just stick with a 22uf, and call it good.

              Comment


              • #8
                I have a question re: dc elevation...

                Does it matter where on the B+ line you tap for the DC reference? I notice a bunch of design that do it at a B+ node AFTER the standby switch. Wouldn't you want to use a spot BEFORE the standby switch, so the heaters are always DC referenced, wouldn't doing it after the standby switch float the Heater CT when in standby?

                Comment


                • #9
                  The CT is always connected to ground thru the voltage divider. Connecting it further down the supply - typically at the screen supply node -allows for cleaner DC.

                  Comment


                  • #10
                    Originally posted by fyl View Post
                    The CT is always connected to ground thru the voltage divider. Connecting it further down the supply - typically at the screen supply node -allows for cleaner DC.
                    Yep, that's what I did, screen supply.
                    I figured I'd have less volts to drop, but turns the screen supply is merely 10V lower. That's what you get when you use a choke

                    Comment

                    Working...
                    X