Announcement

Collapse
No announcement yet.

Lower Heater Voltage With Lower Center Tap Resistors?

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

  • Lower Heater Voltage With Lower Center Tap Resistors?

    I haven't tried this yet but was wondering if anyone has. Many amps run high heater voltage due to higher wall voltage or possibly due to over rated power transformers so I'm thinking about using lower resistance value center tap resistors to sink some extra current to ground.

    I have always used two .1 ohm 2 watt resistors on a little perf board to lower heater voltage. I run each heater line through one of the resistors to keep the voltage balanced but I'm curious if anyone has tried lowering the center tap resistance. I realize this will push the power transformer harder.

    Here's the math:

    With 6.9v wire-to-wire each center tap resistor sees approximately 3.45v

    220 ohm resistors sink 15.6ma at .05 watts

    100 = 34.5ma .12 watts

    47 = 73ma .25 watts

    10 = 345ma 1.2w
    https://RobRobinette.com/Amp_Stuff.htm

  • #2
    Your traditional method of putting each heater line through one resistor to drop excess voltage is the best way to go. That method actually reduces the current load on the transformer while it reduces the voltage reaching the tube heaters.

    The other proposed method of reducing the virtual center tap resistance to load the transformer winding and thus reduce voltage is very bad practice for several reasons.

    1) Increases stress on the transformer
    2) Increases heat dissipated inside the chassis
    3) Requires higher wattage / more expensive resistors
    4) Just plain inefficient. It is analogous to reducing the speed of your car by riding the brake rather than letting up on the accelerator.

    Comment


    • #3
      Back to back diodes as a series dropper are an alternative to a resistor. Given the high current nature of the circuit, it may not reduce heat much though.
      My band:- http://www.youtube.com/user/RedwingBand

      Comment


      • #4
        Originally posted by Tom Phillips View Post
        The other proposed method of reducing the virtual center tap resistance to load the transformer winding and thus reduce voltage is very bad practice for several reasons.

        1) Increases stress on the transformer
        2) Increases heat dissipated inside the chassis
        3) Requires higher wattage / more expensive resistors
        4) Just plain inefficient. It is analogous to reducing the speed of your car by riding the brake rather than letting up on the accelerator.
        #1 I agree and this is my primary concern.
        #2 The extra .1 ohm voltage dropping resistors generate heat inside the chassis.
        #3 The extra .1 ohm 2 watt resistors cost pretty much the same as higher watt center taps.
        #4 The extra .1 ohm resistors are inefficient too, they are also turning electric power into heat.

        The pro is the amp looks cleaner without having the voltage dropping perf board inside the chassis. This is why I'm investigating this mod.
        Last edited by robrob; 03-22-2017, 02:50 PM.
        https://RobRobinette.com/Amp_Stuff.htm

        Comment


        • #5
          I used to have a separate transformer on my breadboard devoted to heaters (so that heaters could run at full voltage when I ran the transformer supplying the HV at lower voltages via the variac). It was an old Thordarson and its heater voltage was too high (either because line voltage differences or because the HV was unloaed, or both) and I used the same technique. Resistors on both sides. Never had an issue with it.

          Comment


          • #6
            pdf64, what diode circuit would you use to drop .5v from each heater line?
            https://RobRobinette.com/Amp_Stuff.htm

            Comment


            • #7
              #4 The extra .1 ohm resistors are inefficient too, they are also turning electric power into heat.
              But the wasted power is way higher, and to boot *inside* the transformer, because you are using the wire own internal resistance to do so.
              A horrible "solution".
              Juan Manuel Fahey

              Comment


              • #8
                Originally posted by pdf64 View Post
                Back to back diodes as a series dropper are an alternative to a resistor. Given the high current nature of the circuit, it may not reduce heat much though.
                To drop as little a 0.5V is a tough call. You'll need to use a beefy low voltage Schottky rectifier to get close. For example the SR520/30/40 in the attached datasheet will drop 550mV at 5A:

                SR520.pdf
                Experience is something you get, just after you really needed it.

                Comment


                • #9
                  Thanks Juan, I'll scrub the idea.
                  https://RobRobinette.com/Amp_Stuff.htm

                  Comment


                  • #10
                    If you use .1 or .22 ohms on each leg to drop the filament voltage, you are not putting that much heat under the chassis. I take out the #47 bulb and put a led bulb in there saving 150ma.

                    Nice dropping calculator here. http://www.mh-audio.nl/DroppingResistor.asp

                    Comment


                    • #11
                      After the feedback in this thread I have decided to scrap the idea. Thanks everyone for the help.
                      https://RobRobinette.com/Amp_Stuff.htm

                      Comment


                      • #12
                        Excessive heater voltage may be a good indication that the PT is being fed a higher primary voltage than it was intended for; hence a mains bucker before the primary may be the best (and lowest loss) option.
                        My band:- http://www.youtube.com/user/RedwingBand

                        Comment


                        • #13
                          We talked about heater voltage on this thread - some reliable technical info:
                          http://music-electronics-forum.com/t18776/

                          Cheers,
                          Ian

                          Comment

                          Working...
                          X