Announcement

Collapse
No announcement yet.

Screaming Bright Switch Cap??? - 1974 Fender SF Twin Reverb Master Volume Push Pull Switch

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

  • in conclusion... everything I've measured is OK, every wire I followed was OK, components changed for new ones, tube socket as well, new tube, etc

    nevertheless:

    - i do not hear a change if I lift the 820R
    - measuring with the scope there's nothing strange going on when poking the board and components with the wooden stick
    - however, there is a big effect when i place the multimeter (video and explanation on post #174)

    I am wondering if there's anything I could measure with the scope, and by that I mean that if the circuit is supposed to invert the signal by 180 deg then I should be able to place the scope's probe somewhere in which I see a SINE wave and then disconnecting something (say the 820R) I should see that SINE wave flipping 180 degrees. Is that right?

    Comment


    • Originally posted by TelRay View Post
      in conclusion... everything I've measured is OK, every wire I followed was OK, components changed for new ones, tube socket as well, new tube, etc

      nevertheless:

      - i do not hear a change if I lift the 820R
      - measuring with the scope there's nothing strange going on when poking the board and components with the wooden stick
      - however, there is a big effect when i place the multimeter (video and explanation on post #174)

      I am wondering if there's anything I could measure with the scope, and by that I mean that if the circuit is supposed to invert the signal by 180 deg then I should be able to place the scope's probe somewhere in which I see a SINE wave and then disconnecting something (say the 820R) I should see that SINE wave flipping 180 degrees. Is that right?
      I don't think the signal will appear flipped, but the out-of-phase signal through the NFB loop makes the original, in-phase, signal smaller by subtracting a little of that signal.
      You should be able to see, with the scope, whether the signal changes at all from the point where the 820R resistor connects to the 100R resistor when you switch in/out the 820R resistor.
      Do you have a dual-channel scope? You should be able to see signals at the lifted 820R resistor and at the junction point that are out of phase. I would expect the signals to be not equal, but roughly of the same magnitude.
      If it still won't get loud enough, it's probably broken. - Steve Conner
      If the thing works, stop fixing it. - Enzo
      We need more chaos in music, in art... I'm here to make it. - Justin Thomas
      MANY things in human experience can be easily differentiated, yet *impossible* to express as a measurement. - Juan Fahey

      Comment


      • - however, there is a big effect when i place the multimeter (video and explanation on post #174)
        It is normal to get some speaker noise if you intermittently touch sensitive circuit points with your meter tip.

        The reasons are:
        - Your meter resistance changes circuit voltages and produces small transient signals (especially in high impedance preamp grid circuits)
        - If inductors are involved (chokes, transformers) the intermittent additional meter current may cause tiny sparks producing EMI
        - The meter leads act as antennas injecting interference signals

        Circuit measurements are typically done with a dummy load to protect the speaker(s).

        It is not normal that lifting the 820R does not produce a significant change in speaker volume. I wouldn't expect much change of sound at low MV settings otherwise, though.

        I suggest to build a 4 Ohm/100W dummy load and do signal voltage measurements at different circuit points. This may require to read signal amplitudes from the scope's display. What is the bandwidth of your DMM?

        Could you confirm the hum modulation on the signal envelope with the 10ms/div horizontal scope setting?
        Last edited by Helmholtz; 04-27-2019, 03:05 PM.
        - Own Opinions Only -

        Comment


        • by subtracting a little of that signal.
          Not only a little. Typically the NFB signal is not much smaller than the direct one.
          Last edited by Helmholtz; 04-26-2019, 08:01 PM.
          - Own Opinions Only -

          Comment


          • Originally posted by Helmholtz View Post
            Not only a little. Typically the NFB signal is not much smaller than the direct one.
            Yes, you are right. While the loop is open.
            If it still won't get loud enough, it's probably broken. - Steve Conner
            If the thing works, stop fixing it. - Enzo
            We need more chaos in music, in art... I'm here to make it. - Justin Thomas
            MANY things in human experience can be easily differentiated, yet *impossible* to express as a measurement. - Juan Fahey

            Comment


            • Originally posted by eschertron View Post
              Yes, you are right. While the loop is open.
              I meant the closed loop situation. In order to reduce the gain from e.g. 40 (open loop) to 10 in closed loop, the NFB signal at the 100 Ohm node (= inverting input) needs to be 75% of the signal at the direct PI input. The higher the ratio of open loop to closed loop gain, the closer NFB and direct PI input signals.
              Last edited by Helmholtz; 04-27-2019, 01:06 PM.
              - Own Opinions Only -

              Comment


              • Originally posted by Helmholtz View Post
                I meant the closed loop situation. In order to reduce the gain from e.g. 40 (open loop) to 10 in closed loop, the NFB signal at the 100 Ohm node (= inverting input) needs to be 75% of the signal at the direct (non-inverting) PI input. The higher the ratio of open loop to closed loop gain, the closer NFB and direct PI input signals.
                OK. I was instructing TelRay to measure the voltage at the 100R node, Not the grid.
                If it still won't get loud enough, it's probably broken. - Steve Conner
                If the thing works, stop fixing it. - Enzo
                We need more chaos in music, in art... I'm here to make it. - Justin Thomas
                MANY things in human experience can be easily differentiated, yet *impossible* to express as a measurement. - Juan Fahey

                Comment


                • The signal at the 100R node is identical to the grid input signal of the right side PI triode. Without NFB this signal is close to zero. With NFB it should be a high percentage of the input signal at the grid of the left side PI triode (meaning the actual direct PI input).
                  Last edited by Helmholtz; 04-27-2019, 12:53 PM.
                  - Own Opinions Only -

                  Comment


                  • Originally posted by Helmholtz View Post
                    The signal at the 100R node is identical to the grid input signal of the right side PI triode. Without NFB this signal is close to zero. With NFB it should be a high percentage of the input signal at the grid of the left side PI triode (meaning the actual direct PI input).
                    Thanks for that. I had to refer to the drawing (post #13) so see where everything went.

                    I'm a little confused why the 'ground' lug of the MV goes to the 100R node and not directly to ground. Is this another of those cases of "manufacturing expedience"?
                    If it still won't get loud enough, it's probably broken. - Steve Conner
                    If the thing works, stop fixing it. - Enzo
                    We need more chaos in music, in art... I'm here to make it. - Justin Thomas
                    MANY things in human experience can be easily differentiated, yet *impossible* to express as a measurement. - Juan Fahey

                    Comment


                    • I'm a little confused why the 'ground' lug of the MV goes to the 100R node and not directly to ground. Is this another of those cases of "manufacturing expedience"?
                      Good question.
                      I actually think so. Don't see any technical benefit.
                      - Own Opinions Only -

                      Comment


                      • Originally posted by eschertron View Post
                        I'm a little confused why the 'ground' lug of the MV goes to the 100R node and not directly to ground.
                        I have actually made myself the same question and double checked with the schematic. It is supposed to be grounded through the 100R.
                        Having the oscilloscope I've tried both ways (direct to the chassis and through the 100R), have not sign any improvement or worsening, so I left it as the schematic dictates.

                        Comment


                        • Since you have a scope, check for any volume difference with 820R NFB resistor connected/disconnected (using fixed test tone at input).
                          Originally posted by Enzo
                          I have a sign in my shop that says, "Never think up reasons not to check something."


                          Comment


                          • Originally posted by g1 View Post
                            Since you have a scope, check for any volume difference with 820R NFB resistor connected/disconnected (using fixed test tone at input).
                            You are totally right! Being a fan of "scientific methodology" and knowing the level of trust in your ears decrease with the amount of time you spend in these kind of projects I should have thought about using the scope just to measure the volume difference I do not seem to hear

                            So... attached the scope to the output SPKR jack, which is connected to a 4 Ohm Dummy load. Put a 400 Hz SINE signal into the amp and you can see the result in the video below.



                            I've followed the reasoning below:
                            Master volume was set at 6.5 and Vibrato Channel Vol at 3.5
                            vertical resolution is set at 0.50 mV per division and what I see is a 3.75 division wave amplitude -> 0.50 mV/div * 3.75 div = 1,875 mV
                            now, when I connect and disconnect the 820R I do see a change of about 1/8 div (either on top and bottom of the wave) -> 0.50 mV/div * 2 * 1/8 div = 125 mV
                            I guess the right way to evaluate if this is an audible change would be to translate that in some kind of logarithmic / dB scale. But as I have not thought of any good way to do that I calculated that the volume is increased by 6.7% when disconnecting the 820R

                            In the second part of the video I have increased the VIBRATO CHAN VOL to 7 to see the parasitic oscillation, connecting and disconnecting 820R does not show an effect on this.

                            My question is, is the 6.7% volume increase what I should expect as an effect to the NFB being disconnected or is this too low?

                            Comment


                            • I couldn't say about how much increase in gain you should see when disconnecting the NFB. I've never considered it beyond "it's working and in phase because the gain decreases when it's connected." I have the math in a book somewhere, but it's never been a primary concern for me. And I don't think it's the primary concern here either. What we do know now is that the NFB loop is in proper phase and DOESN'T change the parasitic. So it's probably getting into the circuit earlier than the from the power amp. I would induce the problem and use the scope to check signal along it's path through the amp in an attempt to locate which stage it starts happening.

                              I've suspected interaction with the reverb for this all along. Helmholtz did ask on a couple of occasions that you test the amp with the reverb driver tube pulled and the reverb gain at zero. I haven't retraced the thread yet, but I think you did something like that and said it was better in a playing test. With the scope you could actually SEE if it's better. Finding out what's interacting is the first order of business for this problem. Once you have clues about the interacting circuits you look for possible failed circuit causes, try things like altering lead dress or move grounds. It's worked before. Many times.
                              "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


                              • thx Chuck!
                                the first indication of oscillation appears when connecting the scope to V2 pin 7
                                it’s very subtle and I would have missed it before, but not now that I know how it looks like.
                                It goes away either disconnecting the REVERB INPUT RCA plug or removing V3
                                (moved wires and tapped a lot of components around but nothing else seems to have an effect)

                                Comment

                                Working...
                                X