Announcement

Collapse
No announcement yet.

Vacuum tube triode and power mosfet substituting?

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

  • Vacuum tube triode and power mosfet substituting?

    OK... Goofing around today on my bench with some power mosfets... found a bunch of "high voltage" capable IRF820s and IRF830s. Sorry about the rough sketching...

    I have a variable bench supply that can deliver 100ma, from 25vdc to 500vdc.

    100K plate and cathode resistors, same with drain and source... just swapped out the triode tube for a power mosfet.
    B+ voltage divider at the grid of the vacuum tube triode and the mosfet gate are both around 3M and a grounded 1M.

    # 1 and # 3 work "absolutely fine" with a few hundred milivolts of drive ,
    #2 does not work at all... why?


    Click image for larger version

Name:	1.jpg
Views:	1
Size:	24.4 KB
ID:	867184Click image for larger version

Name:	2.jpg
Views:	1
Size:	27.5 KB
ID:	867185Click image for larger version

Name:	3.jpg
Views:	1
Size:	27.9 KB
ID:	867186
    Bruce

    Mission Amps
    Denver, CO. 80022
    www.missionamps.com
    303-955-2412

  • #2
    I'd expect #2 to work fine. What is it doing wrong? What DC voltages do you read at the drain and source?
    "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

    Comment


    • #3
      I am a complete novice when it comes to these things.
      I only tried three power mosfets before stopping and thinking I have something screwed up... 'cause this ought to just "work" right off the top!

      I hear this all the time from my vacuum tube kit builders.... "I checked the wiring layout a bunch of times and I'm sure the divider, DC blocking caps, the drain and source resistors are correct". ha ha ha... of course it never is.

      After these voltage tests and messing with it for at least 30-45 mins:
      with the FET out, the 1/2w resistors all check within 3-5% of spec, so no damage form excessive current.
      The B+ divider seemed pretty good at about 250vdc on the B+ feeding node with around 75vdc to 80vdc at the junction of the divider, where I'm, "biasing" the IRF820-830.
      I forgot to measure the source voltage, darn it!

      With the FET in... I think the drain side was passing signal as I expected, but the drop across the drain resistor was not as much as I expected and the AC out was pretty much the same as going in...
      but there was zero AC from the source side.
      And, after a few moments it seemed like the junction of the divider and the gate would go very low in DCV...
      But it did that with all three power mosfets.
      I can't see what would cause that and the FETs seem to be fine after... no shorts across elements.
      I can turn the FET's internal diode on and off with my 1.5v Ohm meter.... not sure if I am supposed to be able to do that...?

      Ha ha... first time in a while I'm scratching my head, saying what the H**L I feel like noob!!
      But entertaining...
      I'll pile the parts back in a little later and try it over again... this time I'll get voltage for all three elements.
      Attached Files
      Bruce

      Mission Amps
      Denver, CO. 80022
      www.missionamps.com
      303-955-2412

      Comment


      • #4
        Gate: 53vdc
        Source: 54.6vdc
        Drain: 247vdc
        Hmmm what the heck....? Maybe not optimal but shouldn't this should just work as is?!!
        Bruce

        Mission Amps
        Denver, CO. 80022
        www.missionamps.com
        303-955-2412

        Comment


        • #5
          The MOSFETs you're playing with, as I'm guessing you know based on your schematics, are enhancement mode devices. The gate needs to be more positive than the source, which it's not, so your MOSFET is turned off. The Vgs plots in the datasheet for the IRF820 don't start until the gate is 4.5v above the source. I'd try decreasing the source resistor, or change the biasing resistors at the gate. I've not played with these fets in this configuration, so no warranties expressed or implied. Void where prohibited. See store for details.
          -Mike

          Comment


          • #6
            Originally posted by defaced View Post
            The MOSFETs you're playing with, as I'm guessing you know based on your schematics, are enhancement mode devices. The gate needs to be more positive than the source, which it's not, so your MOSFET is turned off. The Vgs plots in the datasheet for the IRF820 don't start until the gate is 4.5v above the source. I'd try decreasing the source resistor, or change the biasing resistors at the gate. I've not played with these fets in this configuration, so no warranties expressed or implied. Void where prohibited. See store for details.
            I told you I was noobie.... oops... reverse that.
            I put an IRF830 in and it seems to work.
            I let it run for a few minutes like that with a 15vac drive on the gate... .
            Remeasured everything and I actually have 55vdc on the gate and almost 52.8vdc on the source.
            Still not optimal I assume.
            So should I still change that divider or source to get it closer to 4vdc?
            Amazingly, I can swing about 60vac across the source to drain now, as viewed on my scope, and almost zero distortion.
            Really appreciate all the help...
            Bruce

            Mission Amps
            Denver, CO. 80022
            www.missionamps.com
            303-955-2412

            Comment


            • #7
              Yep, a MOSFET should work much better than a tube in the cathodyne circuit. Maybe you had a bad one that was leaking excessive current from drain to gate.

              As long as the gate bias network gives one-third B+ at the gate, and the drain and source resistors are equal and the MOSFET is good, all the other voltages should find their correct levels. The source will sit at 1/3 of B+ and the drain 2/3 of B+.

              Actually since the MOSFET can conduct with a much lower voltage drop than a tube, you get more output by setting the gate to 1/4 B+. The source side can swing between 0 and 1/2, and the drain side between 1/2 and 1, and clipping will be symmetrical both sides.

              If you like playing with MOSFETs you ought to check out the LND150. This is a tiny 500V depletion mode MOSFET that behaves almost, but not entirely, unlike a 12AX7. (it biases up a bit colder when used with the usual 12AX7 cathode resistor)

              My tube emulator circuit using a LND150 is the second Google search result for "LND150".
              "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

              Comment


              • #8
                Hi Bruce and all,

                I've got a couple of simulations of cathodyne PIs using IRF840s if you're interested, though I'm afraid they use the free version of Icaps rather than the more popular LTspice.

                Helped me understand it. Though you probably have more understanding than me!

                Comment

                Working...
                X