Announcement

Collapse
No announcement yet.

Tube Rectification, GZ34 and 5U4GB, AC output and other limits

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

  • Tube Rectification, GZ34 and 5U4GB, AC output and other limits

    Two of my power txf I had made for a moderately high gain circuit also have 5V/6A taps. I had these included because for whatever reason I love the sound I get when using tube rectifiers. I assume it is nothing more than the lower voltage because of the tube's vdrop. I suppose I could have had a lower secondary txf made, but here we are nonetheless.

    One txf offers 360-0-360 on the secondaries and the other has 345-0-345.

    My target VDC is 480 to 500. With ss rectifiers I get 500 to 515 VDC (using the 360-0-360 - have not used the other yet) depending on time of day. Previously I have used 2x 5U4GBs and arrived to a DC voltage of 486.

    I was looking at old Marshall amps on Amp Archives... I have only seen (in photo of course) one or two other instances, in my various usage of the internets, a Marshall 1987 with a tube rectifier. I suppose some JTM50/JMP 1987 transition thing. But today I saw a 100 watt Superlead (likely also a JTM45/100 transition amp) with a single GZ34, no ss rectification whatsoever.

    So I started to poke around various datasheets and I am hoping to gain some clarification.

    Starting with a 5U4GB, Sovtek's datasheet (linked HERE for convenience) states:
    Full Wave Rectifier - Capacitor Input Filter
    AC Plate Supply Voltage (Each Plate RMS) - 450V
    DC Output Current - 275mA
    DC Output Voltage at Filter Input - 460V

    And Sovtek's GZ34 (linked HERE for convenience) :
    Full Wave Rectifier - Capacitor Input Filter
    AC Plate Supply Voltage (Each Plate RMS) - 550V
    DC Output Current - 160mA
    DC Output Voltage at Filter Input - 640V

    Here is where the clarification is needed... Reading these datasheets makes me wonder if I could run a single GZ34 and get the needed 485+ DV volts to power the amp rather than 2x 5U4GB. And perhaps I am too close to max anyway... so, yea, I am looking for a little knowledge from those of you with the experience to back this up, as I have NONE.

    I do realize though that the 5U4GB vdrop is much greater than the GZ34.

    Thank you, as usual, for your wizzzzdom!
    "'He who first proclaims to have golden ears is the only one in the argument who can truly have golden ears.' The opponent, therefore, must, by the rules, have tin ears, since there can only be one golden-eared person per argument." - Randall Aiken

  • #2
    BY THE WAY, the title should read "AC input" not AC output
    "'He who first proclaims to have golden ears is the only one in the argument who can truly have golden ears.' The opponent, therefore, must, by the rules, have tin ears, since there can only be one golden-eared person per argument." - Randall Aiken

    Comment


    • #3
      I don't see the numbers you quoted for the GZ34 on the data sheet you linked, and it would be a better idea to quote the numbers from the 450 VRMS column so you are comparing apples to apples. In addition, it looks like the GZ34 data sheet wants to have 150 Ohms in series with each plate for current limiting. Something you won't find on any guitar amp although it could just be the resistance of the transformer winding.

      You didn't specify what current will be when the target Voltage is expected. A pair of 5U4 might have a slightly higher Voltage drop at low currents, but they'll still be rockin when the GZ34 is red plating.
      WARNING! Musical Instrument amplifiers contain lethal voltages and can retain them even when unplugged. Refer service to qualified personnel.
      REMEMBER: Everybody knows that smokin' ain't allowed in school !

      Comment


      • #4
        Designing power supplies with rectifier tubes is a rather demanding task.

        The first thing to know/specify is the required DC current at full power. A 100W tube amp draws between 300mA and 400mA. Max. DC current of the GZ34 is 250mA, so a single tube won't suffice.

        Using 2xGZ34 will require the PT to provide a minimum DCR of 75 Ohm per side. Otherwise additional power resistors should be used. These series resistances on the AC side are necessary to limit peak charging currents.
        - Own Opinions Only -

        Comment


        • #5
          Originally posted by loudthud View Post
          You didn't specify what current will be when the target Voltage is expected. A pair of 5U4 might have a slightly higher Voltage drop at low currents, but they'll still be rockin when the GZ34 is red plating.
          Originally posted by Helmholtz View Post
          Designing power supplies with rectifier tubes is a rather demanding task.

          The first thing to know/specify is the required DC current at full power. A 100W tube amp draws between 300mA and 400mA. Max. DC current of the GZ34 is 250mA, so a single tube won't suffice.

          Using 2xGZ34 will require the PT to provide a minimum DCR of 75 Ohm per side. Otherwise additional power resistors should be used. These series resistances on the AC side are necessary to limit peak charging currents.

          Yes, I suspected that a single GZ34 would not be enough, but I need some education here and I, of course, was not confident in what I thought unless I asked.

          According to this tube amp power transformer current draw calculator, I will use up around 220mA... but I assume more than that. We're talking a 100 watt amp with 4x 5881 and 5x 12ax7 tubes. The calculation I factored in was for a 5U4GB rectifier, which also seems to be able to provide the most current.

          Alright, so 2x 5Y4GB.
          "'He who first proclaims to have golden ears is the only one in the argument who can truly have golden ears.' The opponent, therefore, must, by the rules, have tin ears, since there can only be one golden-eared person per argument." - Randall Aiken

          Comment


          • #6
            According to this tube amp power transformer current draw calculator, I will use up around 220mA..
            A DC current of 0.22A @100W is much too low. I checked the formulas used in the calculator and think the DC current result is based solely on output power, ignoring power losses (dissipation) in the tubes. Including these about doubles the DC current.

            From the GE 6L6-GC datasheet example I get a total DC current of 232mA for 2 tubes (55W) at full power. So 464mA for 4 tubes. Add 6mA for the the preamp tubes and we have 470mA.

            I see that the values I specified in my earlier post (300mA/400mA) were actually too low.
            - Own Opinions Only -

            Comment


            • #7
              These amps are not used to produce continuous full power so for that reason an input power of 170W (350mA at 480V) is conservative, not that there is anything wrong with that.

              I find a pretty handy lower limit rule of thumb is to say the power input is about the same as the power output. The rationale is that a class AB push pull tube amp is about 50% efficient and that since it's not continuous you can down rate by half, getting you back to 100W. 100W @ 480V is 200mA. I would not go any lower than that. In this light the estimator's figure of 220mA is not so bad. I'd probably shoot for 250mA.
              Last edited by nickb; 09-21-2019, 09:10 PM.
              Experience is something you get, just after you really needed it.

              Comment


              • #8
                As there is no indication from the datasheet that the current limit for the GZ34 of 250mA is a time average value, I interpret it as maximum allowable current. As such the 1967 Valvo databook allows for a maximum exceedance of 10% with music program.

                I agree that the efficiency of class AB PP amps is typically around 50% or lower (not including heater power). Means that max. DC current can be roughly estimated by 2*Pout/HT + preamp consumption. In this particular case (100W/480V) this means max. Idc> 417mA. Lower plate supply increases current demand for same output power.

                Power stage clipping further increases current demand.

                PTs are often rated at up to four times the output power.
                Last edited by Helmholtz; 09-21-2019, 08:16 PM.
                - Own Opinions Only -

                Comment


                • #9
                  Originally posted by Helmholtz View Post
                  As there is no indication from the datasheet that the current limit for the GZ34 of 250mA is a time average value, I interpret it as maximum allowable current. As such the 1967 Valvo databook allows for a maximum exceedance of 10% with music program.

                  I agree that the efficiency of class AB PP amps is typically around 50% or lower (not including heater power). Means that max. DC current can be roughly estimated by 2*Pout/HT + preamp consumption. In this particular case (100W/480V) this means max. Idc> 417mA. Lower plate supply increases current demand for same output power.

                  Power stage clipping further increases current demand.

                  PTs are often rated at up to four times the output power.
                  I guess I was really trying to get a number for the time averaged power and on reflection that's not really what the OP looking for.

                  That's interesting about the 10% increase on limit for music. I had never noticed that. Thx. Yes PT's need to be a lot bigger.

                  Gtr0, it might be heresy, but you might consider that silicon rectifiers will save you a lot of power, cost a LOT less, use less space, are easier to design in, run cooler, allow more capacitance (== lower hum), be more robust and if you add some series resistance will behave much the same as tube rectifiers?
                  Experience is something you get, just after you really needed it.

                  Comment


                  • #10
                    If you're looking for sag but don't mind the use of solid state diodes, and don't want to use series resistance to get the sag, you can also consider the use of a voltage doubler. A benefit to that is the transformer is often cheaper, and if you use a full wave doubler with 4 diodes you get pretty good performance, and you will get some nice sag when the amp is up loud. Use bigger caps than you would think in the power supply and the circuit will perform better.

                    Greg

                    Comment


                    • #11
                      Originally posted by Helmholtz View Post
                      A DC current of 0.22A @100W is much too low. I checked the formulas used in the calculator and think the DC current result is based solely on output power, ignoring power losses (dissipation) in the tubes. Including these about doubles the DC current.

                      From the GE 6L6-GC datasheet example I get a total DC current of 232mA for 2 tubes (55W) at full power. So 464mA for 4 tubes. Add 6mA for the the preamp tubes and we have 470mA.

                      I see that the values I specified in my earlier post (300mA/400mA) were actually too low.
                      Thanks for that nugget of info! In fact, I wasn't quite sure how to guesstimate required current overall outside of creating/calculating ac load line sheets, which I can do, but I have only done pre-amp tubes, never looked into it for power. But using the maximum allowable seems better.

                      But the issue I come across, and a reason I never considered this, is because some datasheets seem not to list this maximum (rather sometimes just at "normal operation") all the time. Sovtek's 5881 only states a "nominal" plate current of 100mA, when 250V is applied... not sure how to interpret that.

                      As far as rect. tubes are concerned, I realize that this is all more trouble than it's worth. In lieu of having 5V/6A taps added to my transformers, I could simply get a lower secondary tap made... switchable even. I seem to be going through lots of trouble to add two tubes expense and heat to my rig
                      "'He who first proclaims to have golden ears is the only one in the argument who can truly have golden ears.' The opponent, therefore, must, by the rules, have tin ears, since there can only be one golden-eared person per argument." - Randall Aiken

                      Comment


                      • #12
                        But the issue I come across, and a reason I never considered this, is because some datasheets seem not to list this maximum (rather sometimes just at "normal operation") all the time. Sovtek's 5881 only states a "nominal" plate current of 100mA, when 250V is applied... not sure how to interpret that.
                        There is little influence of the tube type on max DC current. Using the formula from post #8 and adding 20% will get you in the ballpark with different tubes.
                        Last edited by Helmholtz; 09-26-2019, 02:48 PM.
                        - Own Opinions Only -

                        Comment


                        • #13
                          Originally posted by Helmholtz View Post
                          There is little influence of the tube type on max DC current. Using the formula from post #8 and adding 20% will get you in the ballpark with different tubes.
                          Thank you, greatly appreciated for this info!
                          "'He who first proclaims to have golden ears is the only one in the argument who can truly have golden ears.' The opponent, therefore, must, by the rules, have tin ears, since there can only be one golden-eared person per argument." - Randall Aiken

                          Comment

                          Working...
                          X