Announcement

Collapse
No announcement yet.

Bias sense resistor question

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

  • Bias sense resistor question

    Hello does the bias sense resistor in a 1-10 ohm range into a 6l6 power tube cathode affect its output impdance please ? Thanks.
    "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

  • #2
    I guess you mean the output impedance at the plate?

    There shouldn't be a noticeable influence with that low values. Internal plate impedance is maybe 20k and the tiny feedback voltage developing across the sense resistor is insignificant compared to the grid drive.
    - Own Opinions Only -

    Comment


    • #3
      Thanks. One more please: there is a good ideea to place a parallel resistor over the speaker to cut a bit the impedance peak please ? I thinking more adding a parralel 20-40 ohm resistor over a 4 ohm woofer...I trying to get a little bit more control in bass region please.
      "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
        A 20R resistor will steal and dissipate (waste) one fifth of the output power.
        I don't understand what you mean with more control.
        - Own Opinions Only -

        Comment


        • #5
          The bass is flabby, some I consider related with speaker resonance. Hoped that taming the resonance peak will help, even with a price of some power loss. The nfb works till a point it alter the midrange which is not what I want.
          "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
            Can't you increase the NFB for a higher damping factor?

            The feedback attenuation of 1/150 is huge.

            What are your measured values of open loop and closed loop gain?

            For good speaker damping the amp's output impedance should be lower than the nominal speaker impedance.
            A 20R shunt alone won't get you there.
            Last edited by Helmholtz; 08-04-2023, 11:53 PM.
            - Own Opinions Only -

            Comment


            • #7
              There was 720 mVpk input for open loop gain onset clipping, 1.44Vpk with nfb applied . Output tubes biased at -49V. 31Vpk into 4 ohm load
              "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
                This is how the inverter/driver shows. The output is 66l6 tubes into 1.9k OT suplied at 45 0v around

                Click image for larger version  Name:	20230805_034055.jpg Views:	0 Size:	1.55 MB ID:	984874
                Last edited by catalin gramada; 08-05-2023, 12:44 AM.
                "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                Comment


                • #9
                  Ok, now I understand how you're getting an NFB ratio of 6dB: You added a bypass cap across the 2.2k cathode resistor above the feedback injection point.
                  That makes a difference as it significantly increases the NFB signal at the cathode.

                  If you always used the bypass cap with this configuration, it's no wonder it sounds different from the 100k/2.1k variant, which results in less NFB.
                  Last edited by Helmholtz; 08-05-2023, 04:32 PM.
                  - Own Opinions Only -

                  Comment


                  • #10
                    One more please: the plate voltage of the second triode is ramping excessive slow. It takes more than one minute till get the same as first triode with 1uf coupling cap and 11 sec around with 100 nF. That is normal please ?
                    I can make the coupling as small as 10 nF for frequency purpose but the phase at 50hz will shift from 0.2 deg /1uf to 20deg /10n What will be the best compromise to do please ? Thanks

                    Click image for larger version  Name:	20230809_120931.jpg Views:	0 Size:	1.59 MB ID:	984974
                    Last edited by catalin gramada; 08-09-2023, 10:05 AM.
                    "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
                      One more please: the plate voltage of the second triode is ramping excessive slow. It takes more than one minute till get the same as first triode with 1uf coupling cap and 11 sec around with 100 nF. That is normal please ?
                      I can make the coupling as small as 10 nF for frequency purpose but the phase at 50hz will shift from 0.2 deg /1uf to 20deg /10n What will be the best compromise to do please ? Thanks
                      After heat-up the cap must be charged through large series resistance.
                      Only after the cap has charged to 200 VDC, the grid voltage is zero and the plate voltage will achieve it's final value.

                      Why does 20 phase shift bother you?
                      Do you see a boost in bass response with NFB applied?

                      - Own Opinions Only -

                      Comment


                      • #12
                        Originally posted by catalin gramada
                        Thank you. Does exist a sonic benefit using individual leak resistors and coupling caps to each tube in parallel configuration vs shared grid leak / coupling cap - or is just a safety measure with no sonic implication please ? Thanks
                        Don't understand what this relates to.

                        Don't expect me to spend time with thinking about your circuit issues if you don't bother to answer MY questions.


                        - Own Opinions Only -

                        Comment


                        • #13
                          I.m sorry. No nfb used yet.
                          Last edited by catalin gramada; 08-11-2023, 09:30 AM.
                          "If it measures good and sounds bad, it is bad. If it measures bad and sounds good, you are measuring the wrong things."

                          Comment

                          Working...
                          X