Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com
Asher Guitars WD Music Products Amplified Parts Mod Kits DIY Nordstarnd Pickups Warmoth.com

Help with a power transformer

Discussion in 'Shock Brother's DIY Amps' started by sds1, Nov 8, 2017.

  1. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Hey, so I'm getting squared away for my next build and once again I am using a donor amp as foundation. And once again all major complications will be related to the conversion part of the project.

    For my build the original schematic calls for a 320-0-320 HT secondary with a 430V B+. Just FYI The build calls for a solid state (bridge) rectifier, and a 3H choke is in series between B+1 and B+2. The power circuit can be modified as necessary, of course.

    In this case the donor (an Ampeg VT-40 that I picked up for $100) has the following power transformer:

    [​IMG]

    I annotated the voltages measured as indicated and according to the secondary wiring on the Ampeg schematic.

    - How am I going to get the 430VDC B+ (and current) I need out of this PT?

    - Can I use other combinations of secondary windings (red to CT, for example) to get closer to 320VAC?

    - Why do you think they separated the red HT and screen windings, does this indicate something about the PT current ratings?

    - Not having PT specs, is it possible to approximate the current ratings for each of the windings?

    - Can the power tubes plates really handle that much voltage (586VDC)??!?

    - Seems to be a lot of power circuit before the standby switch is tied back to the CT like a ground. Is that common? Looks odd to my inexperienced eye.

    - I was also wondering how that Screen power rail (E) got dropped so far, as it's after a full wave rectifier and I don't see any resistors in that circuit at all. I would expect the voltage to be much, much higher. Where all those volts go to?

    I have so many questions as usual. Thanks in advance.
     

  2. FenderLover

    FenderLover Friend of Leo's

    Jun 11, 2009
    Minnesota
    1 There is plenty voltage and current to work with: Node A shows 545V, Node C is 337V using dropping resistors
    2 Basically, no...nor would it be necessary
    3 This is sometimes done to have a separate supply for the screen grid circuit for better regulation at a lower design voltage. It is likely not a high current winding.
    4 No rule of thumb that I know of, but comparing the output tube compliment should give a hint what it is capable of
    5 It was the era of Rule Breaking... under load it was less. The capacitors will take 900V, the 7027A's will run on 500V.
    6 They are leaving C301/C302 charged whether in STDBY or not. Prevents suffering surge current when coming off STDBY
    7 The screen winding is a separate supply, and is not 'after' the rectifier, it has its own rectifiers.
     
    sds1 likes this.

  3. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Thank you. I understand most of what you're saying there, but I have some follow-up questions:

    1) I guess I should ask what your approach to a power circuit would be here? Node A is too much voltage, on most amps I've seen this is rectified B+ running to the OT, would you suggest for example for my purposes that this node would not be connected to anything? Then step down the voltage from there and send power around the amp as necessary? I guess that is the part that I'm hung up on, that no matter what DC coming right off the rectifier is too high for any usage in the new amp.

    7) Thanks, what I meant here was that the screen power supply circuit is using a full wave rectifier which would be 1.41 * 507V = 714VDC but they show E at 336VDC, and I'm just wondering how it got stepped down so much without any resistors. Or what am I missing?
     

  4. FenderLover

    FenderLover Friend of Leo's

    Jun 11, 2009
    Minnesota
    You could ignore the first node and create another one to step down the voltage. There is already an unused node after R306. If you change R306 to 1K, you'll be pretty close. Depending on your output tubes, you can increase/decrease that as needed.

    Look between the BLU-YEL and VIOLET wires and you'll see D206 and D207, with a center tap ORG wire. It's a standard 2-diode full wave rectifier like Fender uses. The connection from the bridge rectifier above it is the ground reference (ORG) center tap. The two supplies are separate. The screen voltage depends entirely on the voltage of the separate winding.
     
    sds1 likes this.

  5. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Gotcha OK thanks! Looks like maybe R306 and R308 are in series to handle the big voltage drop there?

    I would like to draw my new power supply using real transformer voltage and real expected current draw. I'll go hunting for some info on this but does anyone have any quick tips on adding up the amperage in order to get all the dropping resistors correct?

    I suppose I will want to have a variety of power series resistors on hand in case I need to tweak the voltages later?

    I think I see all of that, it's just that I measured 507v between BLU-YEL and VIOLET so I'm not sure how it steps down to 338VDC at node E. After full wave rectifier at D206 and D207 wouldn't we have over 700VDC? Still not sure what I'm missing but that's OK as that winding will get taped off anyhow.
     

  6. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    OK, so I took a first crack at this.

    I created a spreadsheet to calculate plate and grid current for each tube, so that I could figure the total amount of current that would be coming across this dropping resistor I need. The current calculations include 10% padding on the power and 20% padding on the voltage. The numbers in red were guesses due to missing info.

    I came up with 168.36mA total. I don't know how close this is but call it a starting point.

    I then calculated that to go from 594V to 430V with that current draw I'd require approximately 1K of resistance, min. 28 watts.

    So to the transformer side of the standby switch I've copied the B+ filter caps and resistors to my diagram followed by a 1K 50W resistor. Everything past the standby switch is unchanged from the schematic I am looking to build from.

    What do you guys think? I had some questions like on which side of standby to put the new dropping resistor, if I'm using good numbers to calculate current draw per valve, etc. I'll see what kind of feedback I get and take it from there.

    [​IMG]
     

  7. Nickfl

    Nickfl Tele-Holic

    680
    May 24, 2016
    Florida

    It looks like you are assuming way too much current draw from the preamp tubes. Half of a 12ax7 is usually around 1 ma of current draw, and unless I'm misunderstanding your spreadsheet, you are assuming 3-12ma each? I think you are probably in the ballpark on the 5881s, look at the tube datasheet to check. My guess is that you would be somewhere in the range of 120-130ma total for 2x5881 plus a handful of preamp tubes.

    Also, be aware that using a resistor to drop the amount of voltage you are looking at here is going to create a huge amount of sag under load. A zener is usually a better choice for dropping B+ because it won't create sag, but this is quite a lot of voltage to drop that way as well, you would need one huge zener, bolted to the chassis and maybe a heat sink, or maybe several smaller ones.

    You are looking at 160v drop, lets say 130ma, so around 22watts to dissipate, at idle, so you'll want a 40watt resistor, or maybe something like 8 x 10 watt, 20v, zener diodes in series.
     

  8. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Hi @Nickfl thanks, I actually used your TOS Micro schematic as a reference here and noticed your considerably lower current ratings, and figured I was on the high side, but was unsure how you calculated current draw.

    Yeah that's right, what I did here for example the 12AX7 datasheet says max plate dissipation per triode is 1W. So V1A plate is at 161V according to the schematic, so around 6mA. Where did I go wrong? Is the 1W an unrealistic number to use here?

    Thank you, the zeners are cheap on mouser so I can deal. Any downsides at all to going this route? At any point in my endeavor here does it become more practical to just get a more appropriate PT?

    The original schematic here called for a tube rectifier but given donor specs I changed the build to solid state rectifier. Maybe a combination of zener diodes and a resistor would be a good idea, to get some of that sag back?

    Finally, with the zeners and/or zener/resistor combo do I really need those 2x100uF filter caps that I copied over from the Ampeg schematic? Or can I just go rectifier->zener series->standby switch->B+ node ?

    BTW this is the Tweed ODS. You probably recognized it but just in case you had doubts. :)
     

  9. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Another way of lowering the B+ voltage on that higher voltage HT winding would be using a shunt voltage regulator that is capable of supplying a high enough forward current at the desired lowered voltage. For this you need to know your expected peak current draw on that supply.

    http://diyaudioprojects.com/mirror/members.aol.com/sbench/reg1.html
     

  10. Nickfl

    Nickfl Tele-Holic

    680
    May 24, 2016
    Florida

    In the datasheet, you want to look under the "typical operation" section rather than the "maximum ratings" when you are calculating your power supply. You'll see it says 0.5-1.2 ma per triode for the 12ax7.

    Actually I didn't pay enough attention to recognize that this was the tweed ODS. I'm not sure I would have figured it out anyway, since mine was a micro version and had much lower B+.

    Personally, I would just stick with the zeners. To me, the stiffer SS rectification seems more appropriate to the high gain nature of the amp. Same thing with the filter caps, I don't see any reason not to use big filters if you are using SS rectification, though I don't know if you need those first two, maybe just up the size of the reservoir cap at node A. Those two caps in series on the Ampeg schematic are used because of the monster voltage on that transformer and because they are the reservoir cap on that schematic. In your case, I don't know that they are doing much (except extra filtering) and I think you would be ok just using that dropping resistor (or zeners) without any filter cap (like a choke input) since you are just using it for the big voltage drop (maybe someone else with more experience should chime in on this to confirm).

    I have only used zeners to lower B+ once, as a modification on my 5e3. They usually seem to be implemented on the center tap of the HT winding, which obviously isn't going to work in your case. I am pretty sure you can still use them with a bridge rectifier, but you'll need to do some further investigation on that to figure out it is done, I don't have any experience with that.
     

  11. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Thanks guys. I looked into some more things today based on your suggestions here -- choke input filter, the tube-based voltage regulators. Seems like the zener diodes are going to be the most straight-forward (hopefully).

    I updated the schematic to reflect a few changes:

    - 8x20V 5W zener diodes connected to the ground side of the rectifier. I found some illustrations elsewhere that were actually posted by @tubeswell and indicate this being acceptable.

    - i put the B+ reservoir caps on the PT side of the standby switch, kept 2 to get the higher rating to handle the entire rectified voltage because I thought it might be safer in case the zener diodes were not doing their job. I forget where I gleaned this, it might have been more related to input choke filter where idle current affects the operation of the choke.

    - I updated my current spreadsheet to reflect more realistic operating values. The total current as @Nickfl predicted -- 125.34mA.

    Thoughts?

    A couple quick questions:

    1) What do you guys suggest with respect to wiring rectifier ground on the PT CT versus on the chassis ground? Ampeg connected it to the CT and I'm not even sure why I'm carrying it forward here.

    2) What about noise from the zener diodes, I think I read louder ones create noise? Will I need capacitors to tame these noises?

    Thank you.

    [​IMG]
     

  12. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    Yes you can lower the B+ with a string of sufficiently rated zeners, but they should be going to ground. Not to another winding. The rectification on Marshall and some Ampeg power supplies is different altogether. If switching noise concerns you, then use another method of regulation, or simply find a PT that delivers the requisite VAC without needing reduction/regulation
     
    Last edited: Nov 9, 2017

  13. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    Thanks! Do you think I should worry about switching noise? I don't have the experience to know whether it's a valid concern or not.
     

  14. tubeswell

    tubeswell Friend of Leo's

    Jul 1, 2008
    NZ
    You won't really know until you switch the amp on. Different circuits with different loads and power supply configurations will produce different permutations. Switching noise is caused by sudden current draw on SS diodes during the peak of the charging cycle. When the diodes stop conducting, the release of energy stored in the leakage inductance in the PT, which contains a lot of HF energy, is easily coupled into the signal via stray capacitance. This produces a hum/buzz that is more prevalent in power supplies that use bridge rectifiers.
     
    sds1 likes this.

  15. Old Tele man

    Old Tele man Tele-Afflicted

    May 10, 2017
    Tucson, AZ
    Select and use only CORRECTLY "sized/rated" components.
     
    tubeswell likes this.

  16. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    But what if you already have a PT and want to adapt it to a particular build? That is the spirit of this discussion and I appreciate those willing to hash it out with me. I may not even use this PT but I've learned quite a bit just exploring the idea.
     

  17. sds1

    sds1 Tele-Meister

    Age:
    41
    364
    May 4, 2017
    Orlando, FL, USA
    I am starting to feel like I'm trying to fit a square peg in a round hole to some degree. I'm willing to take on a bit of risk but maybe not this much. I'm now thinking I could get the Super transformer as specified and put the original Ampeg one on eBay -- maybe even make a few buck off the deal. Also with the "correct" transformer I get the 5V secondary so the tube rectifier gets back into the build.
     
    Old Tele man likes this.

  18. Commodore 64

    Commodore 64 Friend of Leo's Ad Free Member

    Age:
    42
    Mar 1, 2010
    Kent, OH
    Another option would be to get a filament transformer, 5V, and add it to the circuit. Then you could drop some voltage with a 5U4G. You'd have to make it a hybrid ala merlin, care of robrob below:
    [​IMG]

    I did this with a 360V (180-180) power transformer in a Bassman 10. It was just perfect for me, but you've got a bit higher voltage to start (230-230). But just wanted to throw that out there for consideration.

    ~20 bucks: https://www.tubesandmore.com/products/transformer-hammond-filament-5-vct-3-open-style
     
    Last edited: Nov 10, 2017

IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.