Announcement

Collapse
No announcement yet.

Help with SE 6V6 - excessive dissipation

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Help with SE 6V6 - excessive dissipation

    I copied this from another thread http://music-electronics-forum.com/t18715/ because I wasn't getting any responses to it --it is kind of an old thread. I have been reading and re-reading Merlin's SE power amp chapter but I am still puzzled.

    My OT is an 8watt SE OT with 5000k:8ohm ratio and I am plugging into an 8ohm speaker.

    Originally posted by Groover View Post
    I still think I may be getting too much plate voltage, quiescent current, and dissipation in the 6V6 in this amp.

    measurements at idle: plate to gnd - 385 V, screen to gnd - 383 V, cathode to gnd -23 V

    according to the Weber Bias calculator tool I am running at ~ 40 mA plate current and ~ 14 watts dissapation.

    What is the best way to bring this tube into safe operating range? How about the easiest way? How about the cheapest way?

    Note: This amp has a Mercury Magnetics Tone Clone VC PT. It may be called tone clone, but I think it puts out more volt-amps than a stock Fender. I think that is why I have this excess dissipation problem.
    OK, I just increased Rk to 583R (measured), but my Va is now = 389, Vk = 24V, both voltages measured to ground. Screen voltage is just a couple volts lower than the plate voltage. According to the Weber bias calculator I am still idling at 39mA for 14W dissipation.

    I seem to remember reading in some threads somewhere that 6V6s are frequently used "beyond spec" in guitar amps without many problems. Am I OK with these figures? or

    Should I just keep increasing Rk? Is it possible that this OT primary impedance is not compatible with a 6V6? I don't have any 16ohm speakers to use to increase the reflected impedance, but I do have a 6L6 I could put in the amp. I just wanted to see if I could get it working with a 6V6 first.

  • #2
    Is your tube redplating?Does it sound good?Are you using a cathode bypass cap?If it is not redplating and it sounds good,just leave it.It is not unusual to see a SE 6V6 idling this high.If it really concerns you,use a bigger cathode resistor.

    Comment


    • #3
      Originally posted by stokes View Post
      Is your tube redplating?
      Not that I can see.
      Originally posted by stokes View Post
      Does it sound good?
      Yes
      Originally posted by stokes View Post
      Are you using a cathode bypass cap?
      Yes 25mFd, but per Merlin's article I may go bigger!
      Originally posted by stokes View Post
      If it is not redplating and it sounds good,just leave it.It is not unusual to see a SE 6V6 idling this high.
      That is good to know, and I guess I will leave it as is for now.

      Comment


      • #4
        Another thing to bear in mind: In a SE Class-A amp the maximum plate dissipation is at idle. As you crank it, the plate gets cooler!

        (Proof: The supply current in a SE Class-A amp is constant from idle to full power, and actually decreases when overdriven. Therefore the power input is constant, and the power to drive the speaker must have come from somewhere, so it could only have come out of the plate dissipation.)

        Therefore the 70% rule probably doesn't apply, as it's presumably meant to leave some spare dissipation for Class-AB amps.
        "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

        Comment


        • #5
          Originally posted by Steve Conner View Post
          Another thing to bear in mind: In a SE Class-A amp the maximum plate dissipation is at idle. As you crank it, the plate gets cooler! Therefore the 70% rule probably doesn't apply, as it's presumably meant to leave some spare dissipation for Class-AB amps.
          More good news!

          Originally posted by Steve Conner View Post
          (Proof: The supply current in a SE Class-A amp is constant from idle to full power, and actually decreases when overdriven. Therefore the power input is constant, and the power to drive the speaker must have come from somewhere, so it could only have come out of the plate dissipation.)
          That makes it sound to me as if using some more resistance at the top of the B+ string could be a successful (if wasteful) way to knock down my available plate voltage, although the rest of the info above makes me think I don't really need to.

          I still want to try an EL34 in the amp (as soon as I get ahold of an extra one) but I wanted to make sure I could get it running successfully with a 6V6 first, as a learning exercise at least.

          Comment


          • #6
            If you use an EL34 you will have to adjust the bias resistor again."Therefore the 70% rule probably doesn't apply, as it's presumably meant to leave some spare dissipation for Class-AB amps".That goes without saying Steve.If you were to lower the current down to 70% you wouldnt be in class A territory anymore.I dont know what you would call it then,as far as "class" goes,but,at the risk of starting the whole"is it true class A" I dont think you could call it class A.

            Comment


            • #7
              Groover, what primary impedance are you running in this thing? 5k?

              Comment


              • #8
                Yes 5k

                Comment


                • #9
                  I've been fooling around with a power amp load line plotter spreadsheet for a day or two, and I think it is working properly now. I don't know what the voltages are expected to be for this circuit (and they're not given on the old Fender schematic), but it looks to me like a 6V6 with a 5k load wants about 250V B+ and 48mA idle current. The way you have it (362V) it won't have much headroom because it is not center biased; my calc indicates only about 2.7W. With 250V B+, and 48mA you can get 5W.

                  MPM
                  Attached Files

                  Comment


                  • #10
                    Looking at those plots, it looks as if you have drawn them for higher voltages than what I am getting in my amp - the upper one appears to meet the abscissa at > 500v, and the lower meets the abscissa at ~ 480v. So unless there is something different about drawing the load line w.r.t. voltage because it is a power stage I don't understand what you are inferring from those plots.

                    Comment


                    • #11
                      The B+ is at the bias point.

                      Comment


                      • #12
                        I think I get it now. The load lines shown are not the actual load line based on the B+ voltage, but the load line shifted higher while maintaining the same gradient, as per this article:

                        The Valve Wizard -Single Ended

                        You are showing the line drawn in purple on the graph on the above referenced page, but not showing the line drawn in blue on the same graph, correct?

                        I also am not sure how I can apply the information in your graphs to my amp. Seems like unless I get my Va down I am stuck with what I have, although acccording to some of the posts in this thread the 6V6 is OK with the current operating conditions.

                        Comment


                        • #13
                          Correct, but the higher load line is the actual load line. The blue one is just the first step along the way in plotting it.

                          You can run it the way it is if it doesn't red-plate, but you can see that it will start to clip earlier (on the cut-off side) than it would if you lowered the B+ and got it biased in the center of the load line. This may be why you are seeing an early break-up.

                          MPM
                          Last edited by martin manning; 05-13-2010, 02:10 AM.

                          Comment


                          • #14
                            The only way I would be comfortable lowering the B+ would be by brute force. I got a good deal on some 150ohm 10w resistors and I could put a couple of those in series with the rectifier before the filter. I''m not opposed to the idea (I could use the waste heat to generate electricity to power the amp!), but is there a better way? I have read about selenium diodes or zener diodes, but they seem complicated to work with.

                            Comment


                            • #15
                              You could use an extra cap at the rectifier and then a 10 watt series resistor between that and the OTCT cap,I think bigger than the 150ohm,tho.A zener on the PT center tap works well,too.

                              Comment

                              Working...
                              X