Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Tube amp power supply question

  1. Jun 14, 2012 #1
    I have a US Navy "Plate and Filament Transformer" Manufactured by Hudson American Corporation.

    This monster weighs 14.5lbs, and is in a sealed steel casing. My output trans weighs 15.5lb as well.

    There are 16 unlabeled terminals, and I'm working to identify those. Suggestions on how to go about this efficiently? I've already taken resistance measurements between all terminals using a 16x16 table to record results

    It does however have the current and voltage labelings for the primary and secondary

    Primary : 110/115/120v AC
    Secondary 1:360-0-360 200MA DC
    Secondary 2: 6.6v 10A CT
    Secondary 3: 12.6V 3A
    Secondary 4: 5.25V 3A

    I'm set on using a solid state rectifier circuit so i don't need the 5.25V Line, or the 12.6V line.

    Could I somehow do something to get extra current on the HV line (S1 windings) with these unused secondaries?

    I want 4x 6550/kt88, so the extra plate current would be great to have around.
    260ma plate idle to 520ma at 100% load (at 450V I think with screen and preamp tubes current draw)
    6550 will handle plate voltage of 800v max
    Bias of 60-70% would be fantastic

    In order to squeeze more current out of the HT (HV) windings, I was told to wire the unused secondaries (12.6v and 5.25V) in series with the Primary, and use the 110V winding with my 120V mains. This increases the heat, but I've got fans (or an oil cooler, b/c I'm crazy) ready to combat this.

    How to I wire the unused lines with the primary?

    This is based around a choke input supply with a 12lb Stancor Swinging Choke.
    As far as the remaining PSU stages go, I was told to find a suitable "Swinging Choke" and couple it with a full wave bridge rectifier (rated atleast 1kv, .4a) and a pair of 100uf 450+VDC caps after the choke. I was also told to use 220k ohm 2+W resistors to divide the voltage between the series-wired filter caps as evenly as possibly

    I was told this would give me 650V and 300ma on the HT line with solid ripple reduction.
    Is this true?

    Now, I found a 12Lb 40's Stancor swinging choke, 3-17Henry, 120ohm DCR, 5000VDC and rated for 400ma for just $50, so I snagged that. I'm going to try and grab a few more electrolytic caps, as the ones i had were motor run caps, 40uf and 450VAC (600vDC), so i'll need more.

    My diodes are 12 1kv 1A 1n4007 diodes, and I'm wiring them three to a series for the bridge rectifier to be extra durable.

    Will all of this nonsense work?
  2. jcsd
  3. Jun 14, 2012 #2
    I don't think wiring the unused low voltage in series with the primary is a good idea.

    1) The power is still drawn from the secondary winding, so power is still provide by the secondary. Power limit mostly has to do with the core size, the thickness of the wires etc.

    2) You don't know how the insulation inside. It was designed to run the low voltage separately. Unless it was designed so the low voltage winding can be floated at high voltage, I would not play with it. I am not saying no, you just have to make sure that the LV wires are designed to float up. If the LV wiring was designed for rectify tube that floats up, you should be ok.

    Just by not drawing current from the LV wires already give you some room to increase the current drawn from the secondary. If it spec 200mA, You might be able to pull a little more.

    What are you trying to do? A guitar amp? If so, why do you want to have 4 6550 to have so much power? All my work is to lower the power!!!! To me, even 2 of the 6550, EL34, 6L6GC are too much power in usual case. I have a Marshall JCM900 and I pulled out two of the 6550, re-biased to Class A to lower the power.
  4. Jun 14, 2012 #3
    Yeah, the piece of equipment that the trafo was from was equipped with tube rectifiers.

    I'm not building it for me, but yeah, it's a guitar amp. The guy wants 4x6550 and unfortunately the customer is always right, even though 40watts is beyond loud enough. he's one of the many people that doesn't understand how DB and watts relate to each other

    I love my weird ancient 7591XYZ amp to death and its "only" 35watts. It has the misfortune of being my 'do-everything' amp, because it has 3 channels, multiple inputs, and a ridiculous amount of outputs, and I can do testing, hi-fi, guitar, and bass with it.
  5. Jun 14, 2012 #4
    This is just my opinion, I tend not be too worry about drawing a little more current in the secondary as any reasonable size wire should have no problem running 300mA through. It is my opinion that drawing extra current just produce more heat over time, not just burn the wire. As I said, if the 6.25V Secondary 2: 6.6v 10A CT, Secondary 3: 12.6V 3A, Secondary 4: 5.25V 3A are all left open, you already gave a lot of margin on the power dissipation. So you should be ok to pull more current from the HV secondary.

    Besides, if the secondary wires is small, you might give the customer some sagging that he might thank you for it. To me, using the rectify tube instead of diodes gives you more the sagging that give the "brown" sound that people desired.

    Again, that's just me. If it catch on fire.............I don't know you!!!!
    Last edited: Jun 14, 2012
  6. Jun 14, 2012 #5


    User Avatar

    that's two terminals, right?

    and that's 3 terminals, right? what does the "DC" mean? there's no rectifier in there, is there?

    probably won't need that unless you're thinking of putting some digital or solid-state analog hardware in there for some signal processing reason.

    well, S1 appears to be center-tapped to me. two rectifiers and you could have 400V+ DC or with a bridge circuit and you could have ±400V or even +400V and +800V (ground one side of the bridge). at 1/5 amp, that's a lotta juice.
  7. Jun 15, 2012 #6

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    Here's a link with some hints. Probably you already are beyond this stage.

    Highest resistance is doubtless the HV winding. Energizing a 360 volt winding with say 120V would produce 1/3 voltage at all other windings which you can measure safely and figure out which is which.
  8. Jun 15, 2012 #7
    I thought about pulling more than 200mA from the HV secondary. The more I think, I am not comfortable to suggest this as this is for the customer.
  9. Jun 15, 2012 #8
    It just might be possible to "open" the center tap and put the 360V windings in parallel, which doubles the current.
    The experts will tell you and me if there are side effects to this.
  10. Jun 15, 2012 #9
    Sorry, I didn´t put my brain in tube mode. The doubling of the current is almost certainly already accounted for in the secondary current rating, so parallel windings won´t help. (except for lowering resistance and ohmic losses).
    And, as rbj pointed out, you don´t need a bridge rectifier.
  11. Jun 16, 2012 #10
    That D.C. has to mean something other than direct current. there is just no way that there are any rectifiers in that box, because there's no room for tubes powerful enough, and transistors were not invented at the time.
  12. Jun 16, 2012 #11


    User Avatar
    Science Advisor

    No, it means direct current.

    This is a typical way of giving information about the transformer. Used with a full wave rectifier, (ie with the center tap grounded) this transformer and the rectifier would give 200 mA out.

    However the transformer winding is only rated for 100 mA continuously.

    With a full wave rectifier, each half of the 360 V - 0V - 360 V winding is only conducting for half the time, so the average current for each half of the winding is still only 100 mA.
  13. Jun 16, 2012 #12
    I got that quote from a guy who worked on the sputtering process for Westinghouse in the 1970s haha.

    He said i should just buy a different xfmr too (individual plate and filament xfmrs). I was really looking forward to having this one in there, but ohwell I suppose.
  14. Jun 16, 2012 #13

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    That transformer sounds like a treasure. I wouldn't let it get away.

    A reflex enclosure would do that customer as much good as his two extra output tubes.

    old jim
  15. Jun 16, 2012 #14
    Hey if you want it, I'm reselling it.

    I love reflex enclosures, it's just he's got 4x12 cabinets instead of acoustically optimized enclosures.
  16. Jun 16, 2012 #15
  17. Jun 16, 2012 #16

    jim hardy

    User Avatar
    Science Advisor
    Gold Member

    alpha check your pm's..
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook