Connect with us

Variable power supply for supplying valve amp builds

Discussion in 'Power Electronics' started by Solidus, Apr 26, 2013.

Scroll to continue with content
  1. Solidus

    Solidus

    349
    4
    Jun 19, 2011
    As the title suggests, I want to build a variable power supply that allows me to supply valve amplifier prototypes with a B+ supply on the fly.

    It beats having to have a hefty (and damn expensive) stock of different power transformers around.

    Here's what I've considered to this point:

    -Putting a high-wattage potentiometer/variable resistor in series with the primary of say, a 600VAC transformer. Rectification would give me up to 800V easily.
    -Making a multitapped transformer, that would allow me to change primary windings to modify the output voltage.
    -Variac supply, which I don't want to do, knowing the exorbitant cost of Variacs and that I'm working on a student budget.

    I don't want to jimmy down the output using resistors, as that would require I have a large stock of varying resistances around and would also assume I have a pretty well-factored load. With building tube amps, only so much can be garnered from the datasheets and at least to me, nothing beats mocking them up on a breadboard to test that everything operates as it should.

    Ideally I'd like to build a constant voltage source, that is, one that would hold the output voltage within a certain degree assuming the load changes a reasonable amount, but how exactly I'd go about this dictates whether indeed that is possible. It doesn't need to go to an extreme degree, say 600V max. The larger RF tube projects I'd still break down and buy a transformer that would give me dedicated output voltage for my needs.

    Does anyone have any ideas? I've tried googling this, but information on builds of this type seems pretty scarce.
     
  2. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    It wouldn't be regulated but a Variac feeding the input of your transformer would give you the adjustable output voltage that you want. HV regulation was not a common feature in the day.

    Chris
     
  3. Solidus

    Solidus

    349
    4
    Jun 19, 2011
    That makes sense. It also wouldn't be very hard to house the HV transformer and rectification/filtering components together so no matter the input voltage, the output is filtered DC ready to use.
     
  4. duke37

    duke37

    5,364
    771
    Jan 9, 2011
    You could use a switched potential divider across the HT and then feed a cathode follower. If I remember rightly, a 12E1 was used for this but any power valve would do. The heater would need a separate supply because of heater/cathode insulation limits.
     
  5. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    Duke, I like that. good thinking! ;)

    Chris
     
  6. Solidus

    Solidus

    349
    4
    Jun 19, 2011
    How would I wire that - would I use a potentiometer to feed the grid?

    Heater supply would not be an issue as many HV trafos designed for valve use carry 1-2 6.3V windings.
     
  7. duke37

    duke37

    5,364
    771
    Jan 9, 2011
    I suggested a switched potential divider instead of a sliding potentiometer since the voltage is so high. You could use a pot if you could find one fat enough.

    If you use a tapped resistor chain with a switch, make sure the resistors and switch are capable of the duty.

    If you use 10k resistors with 100V across each then each resistor will dissipate 1W (use 2W resistors) then special high voltage resistors will not be needed.
     
  8. Miguel Lopez

    Miguel Lopez

    252
    63
    Jan 25, 2012
    A valve regulator would do the job. What duke37 says is very important cause the heater-cathode insulation has limits. See the attachment #1.

    I once designed a circuit for this purpose for a friend, using transistors. The schematic includes a current protection circuit and a rapid discharge circuit for the caps. See attachment #2.
     

    Attached Files:

  9. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    The voltage requirement that Solidus is desirous of is quite a bit higher than your circuit provides. Interesting though but could use some component values. ;)

    Chris
     
    Last edited: Apr 30, 2013
  10. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,496
    2,837
    Jan 21, 2010
    Excellent!

    The valve equivalent of a three terminal regulator.
     
  11. Miguel Lopez

    Miguel Lopez

    252
    63
    Jan 25, 2012
    I know Chris. It was just an example. By adjusting components values, he can get the desired voltage.

    Come on Steve.:)
    I was talking about the first image in my post. Althought, thinking better on it....with proper wiring and ECL82 could be considered like that. :eek:

    If there were valve operational amplifiers, why there wouldn't be valve monolithic voltage regulators :D. Something like a 78-250 or a 78-300. :D
     
  12. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    What compontent values? :confused: There aren't any. :D

    Chris
     
  13. Miguel Lopez

    Miguel Lopez

    252
    63
    Jan 25, 2012
    Here is the schematic with the values that I used.

    To increase the output voltage, you must increase the value of Vnr, the 100k potentiometrer should be increased too (depending on the output needed), the transistors must be selected properly, etc.

    Several calculations have to be made. I can do it if needed.
     

    Attached Files:

  14. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    Actually, the vacuum tube model interested me. ;)

    Chris
     
  15. Miguel Lopez

    Miguel Lopez

    252
    63
    Jan 25, 2012
    Sorry Chris. I apologize for the misunderstanding. Please, be more especific next time.

    I would have to calculate (or simulate) those values. Sorry. It will take some time. Try these examples:
     

    Attached Files:

  16. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    No, don't go through that trouble just for me. I'm not the OP. I thought it was a circuit that you designed.

    Chris
     
  17. Weezykid

    Weezykid

    1
    0
    Jul 24, 2013
    If this should be in a new topic, please let me know, but it is along the same subject.

    What about using a couple resistors and switch to drop the b+ volt before it powers the ouput tubes? You could cut the power by simple 1/2 and a 1/3.. (Possibly more with more resistors to have multiple dropping steps). Just not sure how to decide what resistor values to use and the type of switch..
     
  18. CDRIVE

    CDRIVE Hauling 10' pipe on a Trek Shift3

    4,960
    652
    May 8, 2012
    Resistive voltage dividers have limited applications. They're fine for circuits that draw minimal current but that's about it.

    Chris
     
  19. Miguel Lopez

    Miguel Lopez

    252
    63
    Jan 25, 2012
    You can improve that idea, by adding a shunt regulator to the resistor divider circuit.
     
  20. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,496
    2,837
    Jan 21, 2010
    All linear regulators can be viewed as potential dividers.

    A shunt regulator has an active "bottom" resistor.

    A series regulator has an active "top" resistor.

    There's a similar arrangement with buck and boost smps designs, but it's stretching the analogy. It breaks down totally with flyback and other topologies.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day

-