Announcement

Collapse
No announcement yet.

el84 bias...How hot?

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

  • el84 bias...How hot?

    First... Merry Christmas, or Happy Holiday.

    With 350 Vp and the screens at 335 using a cathode bias, how hot can I safely bias my el84s? I know the rules (75% yada, yada) and max diss and all. But how hot can I REALLY bias them?

    I ask because I'm trying to kill crossover distortion. I am driving the piss out of the tubes and I am using the zener trick already. I don't want to run fixed bias if I can help it.

    I don't need to squash the crossover distortion completely. In fact I think a really miniscule amount is OK. But I do want to try reducing it a little more.

    Any ideas?

    TIA

    Chuck
    "Take two placebos, works twice as well." Enzo

    "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

    "If you're not interested in opinions and the experience of others, why even start a thread?
    You can't just expect consent." Helmholtz

  • #2
    Fixed bias - 85%? Cathode bias - 100%?
    Building a better world (one tube amp at a time)

    "I have never had to invoke a formula to fight oscillation in a guitar amp."- Enzo

    Comment


    • #3
      Most fixed biased amps: 60-70% from Pa max (using the shunt-methode) over here in my shop.

      With the posted Voltages I would say:
      - 47-68 ohms shared cathode-R for 4 EL84s
      - use decent EL84s (I prefer mil.spec. rus. ones)
      - if too hot: lower the G2-voltage somewhat

      Hope this helps/merry xmas in advance.
      Love, peace & loudness,
      Chris
      http://www.CMWamps.com

      Comment


      • #4
        Hi Chuck, Hi tw,

        first of all I'd like to wish you Merry Xmas and a truly Happy New Year!

        As to Chuck's question, I agree with tw, cathode bias can be placed a little higher because it's a self-adjusting design to some degree.

        EL84s are rated at 12W, but they're biased even hotter than that in AC30s ( 14,5W ) so I guess even 100% would be easier than that on the tubes, especially talkin' 'bout modern ones, which I think to be veeeeeeery far from the old tubes' manufacturing standards.

        If your only concern is to kill cross-over distortion, I doubt you will have to go that far, usually it's not necessary to bias at 75 or 100% to get rid of it.
        If you have a scope you will be able to track down the waveforms and bias just a little bit over the point at which cross-over distortion disappears;
        this way your tubes will last longer.

        If, OTOH, you want to bias your tubes higher than that for "sonic" reasons, nothing keeps you from trying higher settings.

        Hope this helps

        Best regards

        Bob
        Hoc unum scio: me nihil scire.

        Comment


        • #5
          Im told part of the cathose bias EL84 "mojo" is punishing the tubes at past 100% . My Crate V32 dissapates 13.5 watts at idle using a 60 ohm resistor with 305 on the plates. Id say your looking at using a 100 -150 ohm resistor to keep them at 100% idle. Bob
          "Reality is an illusion albeit a very persistant one " Albert Einstein

          Comment


          • #6
            Thanks for all the replies. I'm going to try going a bit hotter.

            There are a couple of misconceptions around el84 bias that never seem to be addressed. It seems alot of people think their el84s are biased too hot. Because of my 350Vp (actually 355) I've been a little conservative.

            With el84s it's important to recognize that as a low wattage tube the screens account for a good deal of the current compared to a bigger bottle tube. So generic statements like "85% fixed, 100% cathode biased" are confusing. Example:

            I'm using a 160 ohm shared cathode R now. It has 10.5 volts across it at idle. So thats 11.6 watts per tube. I'm at 100% dissapation now. But if you account for the screens you can subtract 6 or 7ma from that. At idle probably more like 5ma. So now I'm at 9.8 watts per tube of actual "plate" dissapation, or 81% dissapation.

            The other issue is the voltage. Most amps for which the 100% dissapation rule has been used are running around 300 to 330VP and 300V on the screens. But when you get above a certain voltage, pushing the current in an el84 can get dangerous. I've run el84s at 398 Vp and 360 screens. When I tried to push them to 100% "plate" dissapation I got red plates even though I still had crossover distortion.

            Chris (CMW), Always happy to get your experienced advice. I'm using two tubes. And I'm going to try a 130R shared. 100R maybe for class A. But I'm not going to try that at 355 volts.

            Thanks again everyone

            Chuck
            "Take two placebos, works twice as well." Enzo

            "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

            "If you're not interested in opinions and the experience of others, why even start a thread?
            You can't just expect consent." Helmholtz

            Comment


            • #7
              When you overdrive your power tubes, the grid current results in an additional bias voltage being developed across the coupling capacitors, that biases them colder. You can start off idling as hot as you want, and if you drive them hard enough, you'll still get into crossover distortion at some point.

              I believe this is part of what accounts for the tube amp sound, but if you don't like it, Ray Ivers once published a trick here with a zener diode (a different one, not the zener across Rk) that claims to do away with it.
              "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

              Comment


              • #8
                Originally posted by Steve Conner View Post
                When you overdrive your power tubes, the grid current results in an additional bias voltage being developed across the coupling capacitors, that biases them colder. You can start off idling as hot as you want, and if you drive them hard enough, you'll still get into crossover distortion at some point.
                Right. Thats the problem. Because I'm driving the heck out of 'em. I am using the "Paul Ruby" zener trick. I'm not using the zener across Rk because then it's not really cathode biased anymore, is it? Well, it IS biased at the cathode I guess. But it the bias is still fixed. Or at least when it hits the point chosen by the zener voltage.

                So, following what you wrote, as long as I'm not getting red plates I could bias hotter and hotter until the crossover just dissapears and actually USE the grid conduction as a part of the bias? Why not I guess. Sounds good on paper.

                With respect to crossover distortion as part of the tone, I agree. My intent is not to squash it completely. But I want a little swirl, not mosquitoes. Some of that has to do with the OT's top end also. If you cant hear mosquito frequencies then what your left with is swirl.

                I wonder also if some cap construction might discharge faster than another helping to reduce the time constant. Just thinking out loud.

                Chuck
                "Take two placebos, works twice as well." Enzo

                "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                "If you're not interested in opinions and the experience of others, why even start a thread?
                You can't just expect consent." Helmholtz

                Comment


                • #9
                  Originally posted by Chuck H View Post
                  I'm using a 160 ohm shared cathode R now. It has 10.5 volts across it at idle. So thats 11.6 watts per tube. I'm at 100% dissapation now. But if you account for the screens you can subtract 6 or 7ma from that. At idle probably more like 5ma. So now I'm at 9.8 watts per tube of actual "plate" dissapation, or 81% dissapation.

                  Chuck
                  You don't measure the plate current and the screen current separately ?

                  -g
                  ______________________________________
                  Gary Moore
                  Moore Amplifiication
                  mooreamps@hotmail.com

                  Comment


                  • #10
                    I haven't yet, but I do account for it approximately. Didn't think it was that critical as long as it's understood that with an el84's lower wattage you do have to account for the screens drawing some current. If a rough guess of screen draw for a typically wired power amp is off by one or two ma it just doesn't seem that critical. Listening tests would seem more important. We're not building low distortion high fi systems here. But now you have me wondering. I'm going to measure the screen current.

                    I can understand how you would be more inclined to be critical of actual screen draw because you use the screens as part of a power reduction circuit.

                    Merry Christmas

                    Chuck
                    "Take two placebos, works twice as well." Enzo

                    "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                    "If you're not interested in opinions and the experience of others, why even start a thread?
                    You can't just expect consent." Helmholtz

                    Comment


                    • #11
                      Try lowering your screen voltage, that helps with crossover distortion albeit at a somewhat reduced power output. The problem with cathode biased AB amps is that the bias shifts (colder) as you get it cranking. Another trick is to reduce the B+ to the PI just enough so the PI clips right when the Power tubes are clipping or even a little before.

                      Comment


                      • #12
                        Yes I am aware of the bias shift. I wrote a rather long post about it that no one responded to...I can't blame anyone for that.

                        Reducing PI output isn't an option either. These amps are supposed to get max clean headroom but still overdrive when pushed such that there is little difference in volume between the clean and overdriven settings.

                        I have worked it out nicely. First I changed to a 130R Rk. I've used the zener on the grid load mod to limit power tube drive voltage. This is still an enormous amount of drive. So this mod gives the power tube grid/ PI decoupling cap circuit a chance to discharge. Then I used another zener across Rk to limit the voltage rise to a point just beyond clipping. That way I get that cool cathode biased compression on clean tones, but the heavily overdriven tones tighten up. Then I added a very, very small value conjunctive filter to even out impedance shifts on the very high freqs and squash the leftover hash. You almost don't notice it's there except that the last tiny bit of "fwizzeraa" is gone and the scope trace is almost a perfect square wave.

                        The result is an EL84 output circuit that is putting out 15 watts pure clean ( 18 watts cleanish) and a full 23 or 24 watts overdriven. With only a tiny trace of crossover distortion audible. Almost none on the scope. I'm very happy with how it's worked out. I've never heard performance like this from EL84s before. I'm stoked actually.

                        I'd be happy to cover more detail if anyone is interested.

                        Chuck
                        "Take two placebos, works twice as well." Enzo

                        "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                        "If you're not interested in opinions and the experience of others, why even start a thread?
                        You can't just expect consent." Helmholtz

                        Comment


                        • #13
                          Originally posted by Chuck H View Post
                          Yes I am aware of the bias shift. I wrote a rather long post about it that no one responded to...I can't blame anyone for that.

                          Reducing PI output isn't an option either. These amps are supposed to get max clean headroom but still overdrive when pushed such that there is little difference in volume between the clean and overdriven settings.

                          I have worked it out nicely. First I changed to a 130R Rk. I've used the zener on the grid load mod to limit power tube drive voltage. This is still an enormous amount of drive. So this mod gives the power tube grid/ PI decoupling cap circuit a chance to discharge. Then I used another zener across Rk to limit the voltage rise to a point just beyond clipping. That way I get that cool cathode biased compression on clean tones, but the heavily overdriven tones tighten up. Then I added a very, very small value conjunctive filter to even out impedance shifts on the very high freqs and squash the leftover hash. You almost don't notice it's there except that the last tiny bit of "fwizzeraa" is gone and the scope trace is almost a perfect square wave.

                          The result is an EL84 output circuit that is putting out 15 watts pure clean ( 18 watts cleanish) and a full 23 or 24 watts overdriven. With only a tiny trace of crossover distortion audible. Almost none on the scope. I'm very happy with how it's worked out. I've never heard performance like this from EL84s before. I'm stoked actually.

                          I'd be happy to cover more detail if anyone is interested.

                          Chuck
                          Not to bust your bubble but I had a Traynor guitarmate (2 EL84's) that put out 25 watts pure sine wave. I added a conjuctive filter on it too. Great sounding amp.

                          Comment


                          • #14
                            The Guitarmate whomps the tubes with 400 to 420 volts. That would increase clean headroom a bit. Provided the tubes your using can take it. Still, I'd be interested in knowing how you measured the wattage.

                            Chuck
                            "Take two placebos, works twice as well." Enzo

                            "Now get off my lawn with your silicooties and boom-chucka speakers and computers masquerading as amplifiers" Justin Thomas

                            "If you're not interested in opinions and the experience of others, why even start a thread?
                            You can't just expect consent." Helmholtz

                            Comment


                            • #15
                              Originally posted by ETR View Post
                              Not to bust your bubble but I had a Traynor guitarmate (2 EL84's) that put out 25 watts pure sine wave. I added a conjuctive filter on it too. Great sounding amp.
                              And I don't want to bust yours, but I think 25W RMS out of a couple of "normal" EL84s is simply impossible ( I' m talking about RMS, not peak to peak power ). Pushing two EL84s to their SOA limit yields about 17 W ( RMS ), and remember an EL84's maximum plate dissipation is 12 W. Even an AC30, which is known to be pretty rough on the output tubes ( they're biased at 14,5 W ) yields "only" some 33 to 36 W RMS ( with four EL84s ) when fully cranked.

                              Shifting the working point higher than this would give an EL84 a glorious ( but extremely brief ) life, much like James Dean.

                              I think your amp's operating conditions are simply too much even for the old, good quality EL84s , let alone the current production ones....

                              With a bias voltage around -17VDC and a + B of 398 V your bias current is more or less 45 mAmps, so the resulting static plate dissipation would be 17 W.

                              Are you sure your amp didn't have 7189s in the first place?

                              I'm asking this because, IMHO, the only way to go that high without shortening the tubes' life is to use 7189s, which are a mil-spec extremely rugged version of the 6BQ5/EL84 and can withstand your amp's operating conditions. Given the voltages present inside your Traynor amp and its mode of operation, I consider it more likely to have been designed with 7189s in mind. ( 440 VDC max. Plate voltage, 400 V max. Screen voltage, 72 mAmps max. Cathode current , 13,2 W max. static Plate diss. ).

                              BTW, the Philips Databook reads 24 W RMS for two 7189s in AB1 at 400 VDC ( Plate ) and a bias voltage of -15 VDC, which seems to be pretty close to your amp.

                              Best regards

                              Bob
                              Last edited by Robert M. Martinelli; 01-03-2009, 06:39 AM.
                              Hoc unum scio: me nihil scire.

                              Comment

                              Working...
                              X