1. Win a Broadcaster or one of 3 Teles! The annual Supporting Member Giveaway is on. To enter Click Here. To see all the prizes and full details Click Here. To view the thread about the giveaway Click Here.

A question for all loadline experts (6L6 amp content)!

Discussion in 'Shock Brother's DIY Amps' started by tweedy_woodpecker, Sep 20, 2020.

  1. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,268
    Joined:
    Dec 29, 2012
    Location:
    United States
    That's extremely tidy considering how dense the circuit is. Nice work.
     
    tweedy_woodpecker likes this.
  2. DjimiWrey

    DjimiWrey Tele-Meister

    Age:
    66
    Posts:
    447
    Joined:
    Aug 15, 2020
    Location:
    austin
    wish i had the tech skills y'all share.. wadda kool site
     
  3. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    Thank you, @robrob! I visit your site often, its has invaluable information that is extremely well structured!
     
    robrob likes this.
  4. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    Thanks again @andrewRneumann!

    I prefer the sound of the lower load, although the difference is not dramatic, I did not think about the relation of plate load and screen current yet, I thought that going for a higher impedance would be easier on the tubes since the load line would be less steep. So the screen current will go up with the higher plate load?
     
  5. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    You’re most welcome! It helps me to clarify my own understanding by attempting to put it clearly in words.

    Yes, but only when you are driving the tubes hard up at the knee or below it. If you keep operations at Va > 100V or so, screen current will stay fairly minimal. Once Va drops below about 100V (for the 6L6), the screen becomes much more attractive to electrons vs the plate. The screen starts stealing the current from the plate. The screen has its own maximum power dissipation limit just like the plate does.
     
    tweedy_woodpecker likes this.
  6. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    @robrob: I have one more question that I can not solve on my own concerning drawing the right loadline for my powertubes:

    1) You write on your website regarding drawing the DC load line for cathode biased triode preamp stage:

    "We start our exercise with the DC plate load line. When you see just "load line" that's what is meant, the "DC plate load line". The bottom point of the line is the plate supply voltage (B+3) of 250 volts. We use the supply voltage because when current flow is cutoff by the tube there well be no voltage drop across the plate load resistor so plate voltage will equal supply voltage. The bottom of the chart is the 0 current (cutoff) line and at cutoff plate voltage = supply voltage.

    We calculate the left end of our line by using Ohm's law and dividing the supply voltage by the plate and cathode resistors (100k plate load resistor + 1.5k cathode resistor)

    250V / 101,500Ω = .00246A or 2.46ma

    We add the resistance of the cathode and plate resistors because all the current that flows through the tube has to flow through those two resistors."

    So shouldn't the value of the cathode resistor (in this since doubled since it is used for both two tubes) be added to the load of the OT here? All current of both power tubes is going through the cathode resistor.
    In your example for a preamp tube it does not really make a difference with the large 100k plate resistor but with a small load like half the primary winding it really does...

    So for example in my case:
    8 Ohm speaker load on secondary with a OT winding ratio 23:1 > 23*23*8=4,23k primary load
    270*2=540R Cathode resistor value for only one tube

    Class B :
    4,23k/2+0,54= 2,66k instead of 2,1k

    Class A:
    4,23k/4+0,54= 1,6k instead of 1,06k

    Does this make sense?
     
  7. robrob

    robrob Poster Extraordinaire Ad Free Member

    Posts:
    8,268
    Joined:
    Dec 29, 2012
    Location:
    United States
    For drawing load lines I do believe you should include the cathode resistor.
     
    tweedy_woodpecker likes this.
  8. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    True for DC current. Not true for AC signal current if cathode resistor is fully bypassed by an appropriately large capacitor.
     
    D'tar and tweedy_woodpecker like this.
  9. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    Of course! I completely forgot about the bypass cap!

    I took measurements again yesterday with different 6L6s, they pass a bit more current and the plate voltage is a little lower. At idle I got about 55mA of plate current, 365V plate/cathode, screens are at 360V, cathode voltage is 31V.

    So these measurements now match the transfer characteristics chart (-31V, 55mA @ 360Vg2) and I tried to "accordionate" the grid lines for the plate chart:

    Bildschirmfoto 2020-09-24 um 08.46.36.png
     
    Last edited: Sep 24, 2020
    andrewRneumann likes this.
  10. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    That looks good. I submit “accordionate” to the Vocabulation Hall of Fame!

    FWIW, something seems weird with those transfer characteristics to me. They should all be for a constant plate voltage, but it’s not labeled. So if you take any point in Vg2=400V curve, it should match up directly with the current at the same plate voltage in the plate characteristics. It is the exact same data in the transfer characteristics as in the plate characteristics, except in the transfer characteristics plate voltage is held constant while in the plate characteristics screen voltage is held constant. Take a look at the GE datasheet page 5 and you will see transfer characteristics with plate voltage (Eb) constant at 250V. They seem to match up with Vg2=250V curve in the chart you are using. But if I look at the Vg2=400V curve on your transfer characteristics it looks like it corresponds more to a plate voltage of 500V. I know that is confusing, but I’m confused by it myself. Where are you getting those graphs from?

    Don’t worry about it though. I think your estimate of screens at 360V (vice 400V) corresponding to a +5V increase in labeled Vg on the chart is spot on regardless.
     
    tweedy_woodpecker likes this.
  11. tubegeek

    tubegeek Friend of Leo's

    Age:
    60
    Posts:
    3,882
    Joined:
    Jan 31, 2020
    Location:
    Brooklyn, NY
    "Accordionate" is a perfectly cromulent word.
     
    tweedy_woodpecker likes this.
  12. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    I took the graphs from the Telefunken sheet. The TUNGSOL 6L6GC datasheet has a very similar transfer characteristics sheet though, it also shows about 60mA for -30V and Vg2= 350V
    Bildschirmfoto 2020-09-24 um 18.23.41.png
     

    Attached Files:

  13. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    004036EB-DA68-461A-8C79-6F25EAF045F6.jpeg

    88BFB4D7-C871-4D3D-AF60-6211FAF9AB08.jpeg
    One chart gives me 360mA and one gives me 340mA. I’m not too concerned about the difference, but since these charts show the same data in a different way, you’d think they’d be exactly the same.
     
  14. andrewRneumann

    andrewRneumann Tele-Holic

    Posts:
    593
    Joined:
    Mar 22, 2020
    Location:
    Cincinnati, OH, USA
    9ABEB501-AFB4-4373-905D-642ACF06CDBF.jpeg
    Same thing on the combined chart. The little circles are the exact same voltages, yet they yield different Ia current. I don’t know why.
     
  15. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    I think maybe when using tube graphs our desire for preciseness can not be satisfied. Or its just a bad graph. It is the same as published by JJ tubes.

    Also, put in two different sets of powertubes of the same batch and get different readings. After all there would be no need gor matching powertubes if all had the exact same values ;)
     
  16. tweedy_woodpecker

    tweedy_woodpecker Tele-Meister

    Age:
    49
    Posts:
    143
    Joined:
    Jun 25, 2019
    Location:
    Austria
    I put in a nice set of 6V6s in today along with a 5y3 and it also sounded fabulous and the readings where very similar to my 5E3 type amp. 360v PV, 20V Cathode voltage, about 35mA plate current...this is where the impedance switch comes in handy.
     
    Last edited: Sep 24, 2020
    andrewRneumann likes this.
IMPORTANT: Treat everyone here with respect, no matter how difficult!
No sex, drug, political, religion or hate discussion permitted here.