Transformers, current, power, etc... group discussion

Discussion in 'Shock Brother's DIY Amps' started by jsnwhite619, Sep 12, 2019.

  1. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,782
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    Been away from electronic stuff for a little while, been building cabinets lately. But, I'm trying to get my mind back in electrical gear again. So, transformers, and I know I've had one or two posts before, but it's the subject that I keep coming back to....

    I would like confirmation or correction on a few bullet points. I'm not quoting anyone, and I'll try to use italics for my own conjecture points

    • Increasing your PT with a higher current rating PT should allow for more headroom and cleaner output at higher levels where the old PT was stressed or exceeding limitations, but not in general.
    • Increasing your OT size - especially in smaller amps - usually involves changing to a reduced primary load than your original. Example: "upgrading" your 15 watt Tweed Deluxe OT to a 22 watt Deluxe Reverb OT increases the headroom and firms up the bass. Is it because it's a "higher rated" OT or is it because the 6.6k allows more current to pass through the PT and power section and that is why you gain headroom? Is the end result because the OT staying cleaner, or is the PT supplying more current? Personal experience: I've used the Hammond 1750E 15-watt 5E3 OT for it, 5f10, 5F11, and Princeton Reverb, the PR being the last build. Until that one, I knew that the OT saturation must be part of the sound I was getting. The PR had amazing cleans - crisp, punchy, and clean pretty well up the dial. I used a 100mA Hammond 290AX PT -- the smallest PT I've used with that OT... the others being 115mA and 138mA. So, that tells me the circuit may have been a bigger factor than either transformer. Also, I've put a Hammond 1750H Deluxe Reverb 6.6k OT on a Harvard build. It is the absolute cleanest 6v6 Tweed amp I've built, and this is my 3rd Harvard, but first one with the 6.6k OT. Everything else is the same. So, is it cleaner because the OT is rated for 20-22 watts and "sounds" different, or is it cleaner because it's 6.6k vs 8k and it's drawing more current through the power section to keep things cleaner & tighter?
    • I know Fender used undersized OT's back in the day to limit bass response and help the low powered speaker situation at the time. But, am I correct that using higher OT primary impedance would allow the use of lower rated PT's as well? Not looking for optimum performance, but if you had a PT that was borderline for a 4k 6L6 amp - optimum performance - would shifting it to a 6.6k OT reduce the current draw, thereby making it a lower output amp suited for the borderline PT? In that line of thought I'm thinking of the Tweed Super/Bandmaster/Pro that I think used a 6-6.6k OT primary.
    Please don't limit responses to my questions. Once you get away from brand preferences, I feel like transformer specs are the real black magic for a lot of people. Anyone can spend $10-20 on caps and tinker with stuff, but I think the expense and potential for catastrophic & expensive failure keeps most average people from going crazy with transformer games.
     
    Last edited: Sep 13, 2019
    peteb and Paul-T like this.
  2. Wyatt

    Wyatt Tele-Afflicted

    Posts:
    1,129
    Joined:
    Nov 3, 2004
    The answers to most of your questions are, "yes, but it depends."

    It's all in the data sheets for each tube. Lets look at the 6V6GT for the Deluxe Reverb and Tweed Reverb.

    [​IMG]
    Maximum output is a 6k6 ohms. If we wanted minimal THD, that is at 6k ohms at a minimum of output loss. By the time we go to 8k ohms, output has started to drop off and THD has started to rise, if we go below 6k, output drops off substantially. There are of course other factors involved,

    The Deluxe Reverb puts out more output than Tweed Deluxe for four main reasons ....
    1. Higher plate voltages...410vdc vs 375vdc...plate voltage is the energy converted for output power, you need more to make more
    2. Optimal plate impedance 6k6 vs 8k
    3. Larger mass, interleaved output transformer that can handle more bandwidth
    4. More efficient long-tail PI using a 12AT7 which can deliver more current without compromising operation
     
    Last edited: Sep 13, 2019
  3. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    Great post Wyatt, and interesting questions jsn.




    What makes a DR louder and more powerful than a tweed deluxe?




    I was unsure of how to answer that question but I would have said,


    Bigger OT, more energy transfer. (I like the larger band width comment)
    Bigger PT
    Higher voltages
    (Forgot about the PI, but yes, the LTPI d finitely plays a role in boosting the signal)
    And then comparing your list to mine, I would say more signal from the pre amp hitting the power stage, along with the bigger OT to handle it, has to be a big part of it.






    Do you mean more current or more signal voltage? What the PI passes to the 6V6 is signal voltage, not signal current.




    I never considered that in the effort to offer a full range of products, that they would purposefully hamstring an amp by mismatching the OT. That brings up interesting possibilities including increased THD.




    That’s an interesting graph. What graph is it?
     
    Last edited: Sep 13, 2019
  4. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,782
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    This website calculator is fun to mess around with. https://thesubjectmatter.com/calcptcurrent.html

    I know I've brought up some of these points in the past, but it still seems to be an area where I can always learn something new. If you play with the OT primary impedance values, you get to see how things could go badly for your PT in a hurry. Setting values for a 325v-0-325v transformer, a GZ34, 2x 6L6 tubes, and 3x 12ax7, here is how the OT primary changes the PT current draw (it also states that it has a 10% safety factor figured in). I assume these would be the numbers if the amp were running maxed out(?)
    • 16k primary = 71.56mA
    • 10k primary = 100.24mA
    • 8k primary = 119.37mA
    • 6k primary = 151.23mA
    • 4k primary = 214.97mA
    • 2k primary = 406.18mA
    SO, does mean that running your 4k - 8ohm amp ext. jack paralleled at 4ohms, your OT would then see the load as a 2k primary, and you could potentially add 90% to your PT current demand?

    And on the other side, going back to the Tweed 6L6 amps, running a 6k vs a 4k allows for a PT with 30% less demand on it? So, a smaller, lighter, and cheaper PT would work.
     
  5. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia



    I agree with your conclusions, except the power tube isn't the only thing requiring current, so the % change at the PT is probably going to be a little less.



    is the current draw for the PT on the primary or the secondary? they may be right for the secondary but look too low for the primary.
     
  6. schmee

    schmee Poster Extraordinaire Silver Supporter

    Posts:
    9,480
    Joined:
    Jun 2, 2003
    Location:
    northwest
    Here are my comments just from actual experience, not theory:
    • Increasing your PT with a higher current rating PT should allow for more headroom and cleaner output YES. More headroom. My current example would a Princeton Reverb where the only thing non stock at this point is a huge Super Reverb PT. This damn thing holds up even outdoors. No change in filters, no change in phase invertor.
    • Increasing your OT size - especially in smaller amps - Same primary winding but more robust tranny = cleaner amp at more output.
     
    jsnwhite619 likes this.
  7. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,782
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    So, the stock Princeton PT is a 70mA. The calculator I listed above puts a PR at 120mA (assuming maxed out). I guess that a 180-200mA PT would allow a Princeton to do whatever it needed to do without ever missing a beat. No replacement for displacement! :D
     
  8. FenderLover

    FenderLover Poster Extraordinaire

    Posts:
    5,250
    Joined:
    Jun 11, 2009
    Location:
    Minnesota
    Bigger motor, bigger exhaust.
     
    Wally and jsnwhite619 like this.
  9. FenderLover

    FenderLover Poster Extraordinaire

    Posts:
    5,250
    Joined:
    Jun 11, 2009
    Location:
    Minnesota
    Notice if your Princeton output went from 70mA to 120mA, you would have a Deluxe output. More current, similar B+, less reflected load: 8K to 6.6K.
     
    Last edited: Sep 14, 2019
    jsnwhite619 likes this.
  10. FenderLover

    FenderLover Poster Extraordinaire

    Posts:
    5,250
    Joined:
    Jun 11, 2009
    Location:
    Minnesota
    Exactly. That an AT7 can pass more current is beside the point, it is not used that way, and it is capacitor coupled. Concertina or LTPI, either will pass enough signal amplitude for the output to overload first anyway.
     
  11. jsnwhite619

    jsnwhite619 Friend of Leo's

    Age:
    35
    Posts:
    2,782
    Joined:
    Sep 10, 2013
    Location:
    Georgia
    I like that. It's funny how it seems like so much transformer "upgrade" talk always centers around output transformers instead of power transformers. @schmee 's experiences with it seems to confirm the "bigger motor, bigger exhaust" analogy with his Princeton. I guess it doesn't matter how great your transmission is if the engine is still redlining?

    The PR I built earlier this year used a Hammond 290AX, 100mA and the stock 15 watt Hammond 1750E OT. It sounded incredible, cranked with a Les Paul was one of the best overdrive tones I've ever had. The whole time, I'm wondering how people can complain about a Princeton's headroom/staying together/etc...this is the only one I've ever played. I guess my 40% extra current rating on the PT did more than I realized. And maybe why my 115mA PT Harvard & 138mA PT Tweed Deluxe seem not to have the general tweed issues that I generally read about?:rolleyes:
     
    Last edited: Sep 14, 2019
  12. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    A bigger OT and PT are going to make their differences.



    What are they? Can these effects be verified? Measured?



    The larger OT must pass more signal. This would show up as more signal voltage at the speaker.




    What I am reading above must mean that when an undersized PT is stressed out with a heavy load, the output voltage will drop. Does the current rise? I’m not sure, but the voltage will drop.



    Voltage sag or a lack of sag in the power transformer would be easy to verify.




    Jason, how much voltage sag are you getting on the B+ on your three over powered 2x6V6 amps? Is it less than would be expected for stock circuits of the same amps? I would think it must be, but maybe by not that much. In the extreme case of Schmee’s 2x6V6 amp with a PT meant for a 2x6L6 amp, it must have significantly reduced B+ sag.




    I have measured power supply sag on different amps. Champs sag not. A SS rectified Bassman head sags about 30 volts. I assume this is power transformer sag. I have measured a 100 volt sag in the B+ of my 6G2 Princeton. I used to attribute that to the sag of the 5Y3 rectifier, but thinking about it, it must be a considerable amount of PT sag as well.
     
    Last edited: Sep 15, 2019
    jsnwhite619 likes this.
  13. Wally

    Wally Telefied Ad Free Member

    Posts:
    34,679
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    Here is an extreme but opposite example of Fender Lover’s ‘bigger motor, bigger exhaust’ observation. The fellow who did all of my tech work for four years before he moved was a great tech...well-versed, lots of experience. He had told me a story about a Rickenbacker M16 combo...2 x 6L6, 4X10...that was on his bench years before. The PT chose that time to fry. The owner was upset and took the amp unrepaired. Maybe ten years after he told me that story, just such an amp came to me. It had very little output, and the owner told me that the PT had burned down years ago while being serviced.....it was the same amp. The owner had found someone to ‘work’ on the amp. I still have the tiny little PT that some rank amateur put in. That PT would have been lucky to have powered an early tweed Deluxe. It could in no way handle the demands of that Ric’s circuit, and the results proved it. Little motor...not much out of the exhaust. I rebuilt that amp with an appropriate PT. Last I heard of it, it lived in Montana. Odd note....the fellow who bought it did not want to pay enough to get the original P10Q Jensen’s!!!!! I got to keep those.
    Indeed, thePT has a lot to do with what comes out of the exhaust.
     
    Last edited: Sep 16, 2019
    jsnwhite619 likes this.
  14. Snfoilhat

    Snfoilhat Tele-Afflicted

    Age:
    38
    Posts:
    1,324
    Joined:
    Apr 8, 2016
    Location:
    Oakland, CA
    I just noticed a difference between these calculations and the plate currents on the 6V6GT data sheet that Wyatt posted. Here's 6L6GC so that it's apples to apples comparison.

    One way of estimating current seems to be assuming the B+ voltage drops across the OT primary impedance and you get these very large increases as primary impedance drops. It's a very steep slope.

    But check out how not-steep the slope of the plate current line on the data sheets is with respect to the same large jumps in primary impedance.
    Screenshot from 2019-09-15 08-47-46.png
    You aren't seeing some worrisome jump in plate current between 4k and 2k, for example.

    When the operating point (not shown, that would be on the plate volts vs plate current data sheet) gets pushed toward wide open the plate voltages drop and the screen current increases, but it doesn't look like the tubes can ever pull those huge currents estimated just by taking the B+ and dividing by the OT primary.

    I'm just pointing out the difference -- I don't actually understand. So that's my contribution to the discussion:eek:
     
    Last edited: Sep 15, 2019
    jsnwhite619 likes this.
  15. King Fan

    King Fan Friend of Leo's

    Posts:
    4,743
    Joined:
    Jan 1, 2013
    Location:
    Salt Lake City
    Ok, I get that a smaller (lower mA) PT will sag and will run hot if it’s often maxed out, as in a stock PR (one of the more extreme stock examples, I’m guessing).

    But I don’t know how sag affects *clean* headroom. (Hmm, is that a redundant term? Is there ‘dirty headroom'?) Less volume, clearly. But is there more distortion? Less distortion? Good distortion or blatty bad distortion? Insert confused face here.


    Sent from my iPad using Tapatalk
     
  16. Wally

    Wally Telefied Ad Free Member

    Posts:
    34,679
    Joined:
    Mar 17, 2003
    Location:
    Lubbock, TX
    My thoughts......Sag is compression. A little tweed Champ is a very good amp with which to experience this. Compression puts a lid on amplitude. Set at a certain point, one can get an uncompressed clean. If you increases the output signal of the guitar slightly with the volume control and/or pick attack the result can be a clean signal that is compressed. Pushed farther, distortion can occur. That distortion will be at the same amplitude as the clean .....weather that clean was the clean that occurred just before the compression set in or the compressed clean or the compressed distortion. Some people can ‘hear’ distortion as louder simply because it is louder, but that is a misleading way to judge volume in such a situation.
     
  17. Wyatt

    Wyatt Tele-Afflicted

    Posts:
    1,129
    Joined:
    Nov 3, 2004
    Not me, that was @FenderLover.

    As to this...

    I am guessing that this online calculator is assuming something as constant or linear that isn't in real life. It's designed to give ballpark estimates for spec'ing PTs, not tube operation. There is no claim to accuracy. To an EE, there is no such thing as supplying too much current, so estimating on the high side or rounded up is just fine.
     
    FenderLover likes this.
  18. clintj

    clintj Friend of Leo's

    Posts:
    4,945
    Joined:
    Apr 4, 2015
    Location:
    Idaho
    This seems like a good thread to jump back in on. I've been taking a break to be outside and enjoy our brief summer, and get my head out of chassis for a while.

    OT primary impedance relates more to plate voltage and idle current, and is chosen for ideal conditions. As an aside, pentodes like it a little on the low side of ideal. As far as current goes, you need to delve into load lines for power tubes. You'll have some static value based on your chosen bias point, but the impedance determines how far from that point you go in each direction under signal conditions. It depends on where the Vg=0 line and load line cross, and then you usually end up grid current limited. The standard cathodyne or long tail pair PI has trouble driving the power tubes past that point since the power tube input impedance rapidly drops as you approach A2 or AB2 operation.

    OT size/power handling relate to each other. Transformers rely on the AC signal in creating a varying magnetic field in the coil and iron core, then transferring that varying field into the secondary coil by induction. The mass of the transformer largely determines how powerful a field can be handled before the iron core saturates, which limits how much power can be transferred through the OT into the speaker.

    Sent from my Pixel 2 using Tapatalk
     
    LudwigvonBirk, Wally and D'tar like this.
  19. peteb

    peteb Friend of Leo's

    Posts:
    4,200
    Joined:
    Apr 25, 2003
    Location:
    Cascadia
    I’m surmising that the current draw at the PT does not have much to do with the impedance of the OT. The impedance only affects AC and the PT only deals in DC production. Interestingly, when a amp uses more power, it is the AC signal in the tube making the tube draw more AC and DC current. Class A does not draw more DC current, class AB does.




    The DC current draw from a pair of 6l6s is mostly dependent on the tube type. At 425 plate volts a pair of 6L6 tubes will idle at 91 mA (70% of MPD) and peak out at 130 mA (100% of MPD). Even if the speaker is changed from an 8 ohm to a four ohm in an impedance mismatch, the tubes are still going to be run between 90 and 130 mA. I don’t think a rebias is necessary when the impedance goes from a match to a mismatch. I’ve changed out a mismatched 8 for a 4 before but I don’t know that I checked the bias first. That would be easy to check, does a mismatch change the bias?




    I’m not sure. Does this make sense?
     
    Last edited: Sep 17, 2019
  20. clintj

    clintj Friend of Leo's

    Posts:
    4,945
    Joined:
    Apr 4, 2015
    Location:
    Idaho
    It should generally be a direct swap, with little to no change in bias. What a transformer swap can change is the DC resistance in the primary coil, so there may be a few volts difference at the plates at each bias setting. But even if it's 5V different in an amp that supplies 500V, that's only a 1 percent difference. There may also be a different imbalance in the two halves of the winding as far as resistance, which can make a difference in how the tubes idle relative to each other. It's a good idea to check both in a pair when setting an amp, because if you set one tube to a max safe setting the other may be over that.

    What determines the idle current flow through the OT is the power tube bias. The power supply is demand based, so adjusting bias current upwards draws more current, and what determines that is grid voltage relative to cathode voltage.*

    *Screen voltage does also have an effect, but is not generally used to adjust bias.

    Sent from my Pixel 2 using Tapatalk
     
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.


  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.