Announcement

Collapse
No announcement yet.

Quick question regard biasing

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

  • Quick question regard biasing

    Hello,
    I wonder what is happen from a biasing point of view if one triode , say 12ax7, is cathode biased through fixed resistor say 2,2 k, for a certain potential point, say 1.6 V, but I put in the grid a large 10M resistor as grid leak, to serve as convenient load for previous stage ? As know such of big grid resistor create by "leaking" current a voltage potential I wonder if the initial bias fitted by cathode resistor will change, please? Thanks

    Or...if is changed it will be disturbed by operational input voltage condition swing ?
    Last edited by catalin gramada; 12-02-2019, 01:35 PM.
    "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

  • #2
    Originally posted by catalin gramada View Post
    Hello,
    I wonder what is happen from a biasing point of view if one triode , say 12ax7, is cathode biased through fixed resistor say 2,2 k, for a certain potential point, say 1.6 V, but I put in the grid a large 10M resistor as grid leak, to serve as convenient load for previous stage ? As know such of big grid resistor create by "leaking" current a voltage potential I wonder if the initial bias fitted by cathode resistor will change, please? Thanks
    A tube needs a negative bias voltage. Bias is grid voltage minus cathode voltage. So if grid voltage is zero and cathode voltage is +1V, bias is -1V.
    A large grid leak resistor develops a negative grid voltage caused by the negative (electrons) grid leak current. This tends to cause more negative total bias, decreasing tube idle current. But as lower tube current means less cathode voltage, actual bias changes much less. In other words, the cathode resistor provides negative DC feedback, which tends to stabilize the idle current.

    Just measure cathode voltage with a 10M grid leak and compare with grounded grid.
    - Own Opinions Only -

    Comment


    • #3
      Thanks, I can measure , but just edited for an additional question,please:

      ...if is changed it will be disturbed by operational input voltage condition swing ?
      Last edited by catalin gramada; 12-02-2019, 01:48 PM.
      "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

      Comment


      • #4
        Originally posted by catalin gramada View Post
        Thanks, I can measure , but just edited for an additional question:

        ...if is changed it will be disturbed by operational input voltage condition swing ?
        If " operational input voltage condition swing " means an AC signal, average DC cathode voltage (determing the bias point) will not change. To be precise, this only holds for linear operation. Asymmetric plate/output signals generally cause a bias shift independent of biasing method.
        - Own Opinions Only -

        Comment


        • #5
          Thank You.
          I.m trying to figure a "better" volume control with better frequency characteristic over a more consistent impedance. A second order in this position also hope to get some tonal benefits. (It is a bass aplication) This is my plan:

          Click image for larger version

Name:	20191202_160534.jpg
Views:	1
Size:	2.41 MB
ID:	856213
          "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

          Comment


          • #6
            Originally posted by catalin gramada View Post
            Thank You.
            I.m trying to figure a "better" volume control with better frequency characteristic over a more consistent impedance. A second order in this position also hope to get some tonal benefits. (It is a bass aplication) This is my plan:

            [ATTACH=CONFIG]56191[/ATTACH]
            What benefits do expect over connecting the grid directly to the vol pot wiper?
            - Own Opinions Only -

            Comment


            • #7
              Think freq reponse will be more consistent preserved over volume control range. Does it works ? Thanks
              "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

              Comment


              • #8
                Originally posted by catalin gramada View Post
                Think freq reponse will be more consistent preserved over volume control range. Does it works ? Thanks
                I can't think of a reason why the frequency response should improve.
                Connecting the grid to the pot wiper eliminates the second coupling cap and actually gives slightly better bass response.

                What's wrong with the frequency response?
                - Own Opinions Only -

                Comment


                • #9
                  I did not expect to improve into tonal way, just to have more consistency over volume pot position. I cannot manage a bypass cap over pot in a way to keep consistent tone from low to loud volume. Trying everything from 100pF to 1nF
                  "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                  Comment


                  • #10
                    This is the first pole, the second one will be determined by input capacitance of the valve itself in conjunction with series resistance and input impedance whatever the signal is shunted to the ground...I thought...
                    What I tried to do is to preserve the big input impedance of second stage and to present a series capacitor, as second pole in a way to not load suplimentarry the first stage as well.The second grid will "see" a more consistent condition in this way, no matter by volume pot position...if it have any sense...
                    It is 10 times diference, should be better 100x, but practical consideration...
                    Last edited by catalin gramada; 12-02-2019, 03:35 PM.
                    "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                    Comment


                    • #11
                      Originally posted by catalin gramada View Post
                      This is the first pole, the second one will be determined by input capacitance of the valve itself in conjunction with series resistance and input impedance whatever the signal is shunted to the ground...I thought...
                      What I tried to do is to preserve the big input impedance of second stage and to present a series capacitor frequency compensation in a way to not load suplimentarry the first stage as well...if it have any sense...
                      Anyway the second coupling cap and grid leak resistor don't improve anything. You can't compensate a low pass filter with a high pass filter having a much lower corner frequency. A bright cap of 100pF to 200pF should be a better solution. I don't understand why it didn't work for you.

                      I deleted my earlier post because it was based on a calculation error. What tube do you use?
                      Last edited by Helmholtz; 12-02-2019, 03:38 PM.
                      - Own Opinions Only -

                      Comment


                      • #12
                        Understood. 12AX7 btw.
                        "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                        Comment


                        • #13
                          OK, from tube data the gain of your 12AX7 is G=72. From this the input capacitance calculates as 120pF. Real value will probably be closer to 150pF, because of wiring influences.
                          Worst case treble attenuation happens at max. source impedance as seen by the grid. This is around the 50% setting of the vol pot, amounting to about 280k.
                          Means that at this setting the upper corner frequency will be about 4kHz.
                          A treble cap of 150pF should be sufficient to compensate the treble loss at this worst case setting.

                          Edit: The gain of 72 above is with bypassed cathode cap. Without cap, gain drops to 40 and input capacitance will be around 100pF. Treble attenuation can be expected to start above 5kHz.
                          Last edited by Helmholtz; 12-03-2019, 12:15 PM.
                          - Own Opinions Only -

                          Comment


                          • #14
                            Thank You Helm.
                            Have another newbie question then, as follow:
                            From a triode perspective the grid leak it count just as load for previous source ? Otherwise, the valve didn't care if the signal swing at its input into 1k or 1M grid leak resistor as input impedance supposed the source can drive it in any circumstances? (don't talk about charging/discharging timing caused by possible capacitive coupling, as can be fitted for any time constant),please ?
                            Last edited by catalin gramada; 12-03-2019, 12:14 PM.
                            "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                            Comment


                            • #15
                              Originally posted by catalin gramada View Post
                              Thank You Helm.
                              Have another newbie question then, as follow:
                              From a triode perspective the grid leak it count just as load for previous source ? Otherwise, the valve didn't care if the signal swing at its input into 1k or 1M grid leak resistor as input impedance supposed the source can drive it in any circumstances? (don't talk about charging/discharging timing caused by possible capacitive coupling, as can be fitted for any time constant),please ?
                              Sorry, not clear what you mean (double language barrier). Maybe you give an example.
                              - Own Opinions Only -

                              Comment

                              Working...
                              X