24vdc to 300vdc adjustable regulated converter

Discussion in 'The Projects Forum' started by tubeguy, Nov 6, 2012.

  1. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    Hello...

    I'd like to build a regulated boost convertor for use in vacuum tube projects.
    Yes... I said vacuum tubes..:)

    Specs:
    Input = 24VDC
    Output = adjustable 50-300VDC @50ma, regulated within 5% or so

    If possible I would like to stick with common components, not special IC's or transformers.

    Attached is a circuit idea with calculations.
    This circuit appears to be workable, but I need help with regulating it.
    Is it possible to use a 555 or op amps for PWM along with some sort of feedback??
     
  2. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    How is this done normally. Is the voltage rectified or just taken straight from a transformer
     
  3. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    It is typically done from an AC line transformer with multiple secondaries. One secondary supplied 6.3 or 12.6 VAC directly to the filaments. Another winding for the plate, aka B+. In some cases there were high voltage batteries for this purpose. The relatively low current requirements mean that smaller diameter wire can be used for those windings. Typically a half wave rectifier and a big filter cap are all that is required. Since there is not much difference between no load and full load further regulation may not be required.
     
  4. crutschow

    Expert

    Mar 14, 2008
    13,016
    3,235
    You can use a 555 to generate a PWM signal and regulate the voltage with feedback, but the feedback circuit needs compensation to prevent oscillations of the feedback loop. Much easier to buy a boost switching regulator IC which includes the compensation network design.
     
  5. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    Hello,

    I want the capability of regulated and adjustable voltage from 50 to 300 vdc so I can both test and match audiio tubes at various operating voltages in real world circuits.

    Standard 60hz transformers and 100uf @450v capacitors are becoming costly. With those as a starting point an adjustable high voltage regulator circuit would still be required.

    I already have some high voltage mosfets and shottky diodes, so this circuit is intriguing because it requires only a small inductor, and high voltage caps of less than 0.1uf
     
  6. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    I hadn't considered oscillations in the feedback loop.
    I did research some boost regulator IC's but they all seemed to require a special HF transformer, not just an inductor. Any suggestions on a particular and easily available IC would be greatly appreciated.

    Thanks.
     
  7. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    Designing you own inductors for a DC-DC converter will be more challenging than buying an HV (HF(?)) transformer to work at 60 Hz.. You can stack capacitors in series to up the capacitance and the voltage rating of the string. Think 10, 10 μF @ 50VDC in series to get 100 μF at 500 V.
     
  8. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    If you somehow can handle the power dissipation. You could make a linear power supply using LM723. Since you live in USA this will require a 1:2 transformer from the mains. Anyway using the mains directly is a no-no
     
  9. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    Any tube project requires both filament and plate supplies. Solving one without the other won't get you very far and as has already been pointed out, choosing a set of tubes whose filaments add up to 120VAC for direct connection in series to the mains is a Bozo no-no.

    Remember the 50C5?
     
  10. t06afre

    AAC Fanatic!

    May 11, 2009
    5,939
    1,222
    Are you really sure about this:confused:
     
  11. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    Hmmm. I was thinking of capacitors in series. The capacitance does not add, but I think the working voltage does, since each capacitor in a series circuit "sees" half the applied voltage.

    Capacitors in parallel add their capacitance and have the same working voltage as the lowest working voltage of any unit in the parallel combination.
     
  12. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    Hello,

    Interesting discussion... Thanks for the input.

    I'm looking outside the box here. making an adjustable regulated hi-frequency PWM supply using simple components I already have.

    I am well aware of normal linear power supply design, filament supplies etc.
    I'm looking for small, and efficient. Just need help with regulation. Need the regulation for close matching of power tubes.

    The circuit attached above page 2 is running at 50khz and needs only very small value cap and inductor.

    Any help with regulation ideas would be greatly appreciated. Thanks in advance
     
  13. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    Yes, the capacitor voltage does add in a series string so 10 50-volt caps would equal a 500 volt cap. Its also a good idea to include a string of balancing resistors across the capacitors.
    If each capacitor in a series string is equal value then the total capacitance is equal to CapValue/CapQty.
     
  14. ScottWang

    Moderator

    Aug 23, 2012
    4,855
    767
  15. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    The OP did ask for a DC-DC converter with 24 VDC input and 50-300V output. I'm not sure if a design meeting that criteria will be forthcoming, but it is hard to argue with the simplicity of the the linear design. Actual performance will require testing of course.
     
  16. tubeguy

    Thread Starter Well-Known Member

    Nov 3, 2012
    1,157
    197
    I agree, it is hard to argue with the simplicity of linear supplies.
    I do need the adjustability. Scott, the circuit you provided is perfect for that with the added benefit of current limiting, and so simple!

    I probably don't need the regulation, because a meter will handle the voltage setting and even when matching tubes, they just need to be stabilized at a particular current setting, so no real need for extra regulation.

    I'll have to break down and buy a transformer and some caps :):).

    First forum I've participated in - lots of fun !!!

    (I started in electronics before the Internet, the computer of choice was a
    Commodore 64.)

    Thanks for the help !
     
  17. Papabravo

    Expert

    Feb 24, 2006
    10,144
    1,791
    Glad you enjoyed it.

    Mine was Big Iron. An IBM 7090 with Model 729 II 9-Track Magnetic Tape Drives and a 1410 with punch card input.
     
  18. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    You could try the LM2577 simple switcher as a boost and see if the free software will kick you out a design. Check TI's website.
     
    Last edited: Nov 9, 2012
  19. bountyhunter

    Well-Known Member

    Sep 7, 2009
    2,498
    507
    Nope, that's wrong. Caps stacked in seriews REDUCES the total capacitance, parallel increases. Two 10uF caps in series give you 5 uF.
     
  20. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    For such a low current level, it's not going to be so bad. Our OP could make either a boost or a flyback supply without working too hard.
    Ahh, I'm afraid you got that backwards. Capacitors in series are calculated like resistance or inductance in parallel; 10 10uF, 50v caps would get you 1uF @ 500v, and you'd need to have high-value resistors across them to balance the network.

    Tubeguy,
    Have a look at Ronald Dekker's great "Flyback Converters for Dummies" page:
    http://www.dos4ever.com/flyback/flyback.html
    There are both a boost converter and a flyback converter presented. One thing you'd have to do for operation on 24v is to add a regulated 10v to 16v circuit to power the 555 timer, as most of them max out at 16v. The R4/R5/R6 network provide feedback to control the output voltage. You'd need to decrease R4/increase R5 to allow a lower output range. Make sure to increase R6 correspondingly.
     
Loading...