Announcement

Collapse
No announcement yet.

Signal Collapse Under Load

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

  • Signal Collapse Under Load

    So I know that when a signal collapses under load in a SS output section it means that the output transistors cannot source the current that the load requires, usually meaning the driver transistors must be replaced. Or at least I think I know that from reading this forum.

    My question is, what is the best way to find the problem. Would a scope probe tracing backwards from the speaker jack eventually show a signal that is not collapsing. SHould I just do DMM voltage checks? Does the transistor reach a point where it breaks down or would it just test open/shorted under idle conditions also?

    I think this is what's happening on this Yamaha G100
    https://www.youtube.com/watch?v=UUoCFM-KI1U

    Signal appears fine via direct out with no load attached

  • #2
    Since you mentioned line out, I'll assume this is a G100-II and attach that schematic. If this is wrong, please post appropriate schematic.
    SS can be difficult to trace because of direct coupling and feedback paths. If all the DC voltages are not correct, it will throw everything off and you will not be able to trace signal. So proper DC voltages at all transistors is first priority.
    Unfortunately, this schematic doesn't tell you much for DC voltages.
    Transistors can breakdown under load or at idle. So a transistor that checks ok with diode check can still breakdown under voltage in circuit.
    Signal collapse is due to output transistor not supplying current as you mentioned, but it can be other things than the drivers. The output transistors themselves or their emitter resistors, or even bad solder could be problems.
    It sounds like you have good signal without load, but check at the speaker out rather than line out, with no load.
    If it is good unloaded, then yes the problem is in the final stages.
    Measure DC volts idling at TR309,310,311,312 first with no load, then with load connected.
    If you are using a limiter lamp, note whether bright or dim.
    Attached Files
    Originally posted by Enzo
    I have a sign in my shop that says, "Never think up reasons not to check something."


    Comment


    • #3
      Thanks for your insight, G. The amp is a G100-112 combo, so that schematic does help me. Did you see the video I posted? I only ask because you ask if I am using a lamp limiter, and it is going off about every 2 seconds in the video link. If it can be said, is this what it looks like for a signal to collapse under load, or would it typically remain collapsed without jumping back up and down?

      I will check these things you suggest and report back shortly.

      Comment


      • #4
        That pulsing could be caused by the lightbulb limiter. I can't remember what amp I had that did that but there was one a while back. Off the limiter it was fine.

        What was the original problem with the amp and what have you done?

        Remember that the power amp is just a bunch of parts that are set up to balance the two sides of the power supplies at idle. When the audio signal comes in it throws that balance off to one side or the other. If the bias is off or part of the circuit can't respond to the incoming audio signal, then problems arise.

        In a lot of cases, the power transistors don't really get turned on until there is a load on the output. The really low level signals that make it through the amp on the pre-drivers and driver alone.

        Comment


        • #5
          If you have no DC at the output and amplifier works well at low power (less than 1 W), plug it straight in the wall.

          Do you still have signal collapse every 2 seconds?

          I guess not.

          EDIT: if it's a Yamaha G100 (you didn't mention that in your original post) , it will not have DC at the output because it has an output capacitor.

          In that case check that speaker out rail sits at roughly 1/2V+ , and that it's not pulling excessive current (say, less than 100mA total from +V ),recheck that
          amplifier works well at low power (less than 1 W),
          and then
          plug it straight in the wall.
          Juan Manuel Fahey

          Comment


          • #6
            I think the amp is working. Try running it without any input signal, with loaded output. Does it stay on? I suspect it will. The bulb is dropping AC input voltage every time the amp starts to pulled current to drive the load.
            The lamp is only used to verify that the amp is not drawing excess current from an unknown problem when first powering up. I have never seen much use for a lamp limiter and never used one. I simply use a metered variac to monitor current and voltage and verify that the amp can be run at normal voltage without fireworks. It is all normal, as yours is, I run it at design mains voltage. Anything less will mean any measurement taken will be worthless.
            Only one thing bothered me about the video. Why the distorted signal? Is it a clean sine wave you are putting into it? And what does it look like with no load? What level input signal and how much gain is it seeing? Everything else seemed to be normal, and expected for an amp starving for power by a lamp limiter

            Also, the mention that driver transistors usually are to be replaced with this symptoms is not correct. You proved the drivers are fine since they can swing rail to rail without a load. The drivers are operating just like a very low current (high load impedance) version of the output current booster stage. You know the drivers are ok if they can take the signal from the class A stage just before the drivers and amplifier both halves of the waveform ok. The output stage is just converting the driver voltage switch to match the low impedance of a 8 ohm load. When you see a nice clean no load signal waveform you know many things about the amp, if is DC stabilized, it is running in AB class ok, the feedback network is working as expected, it has not DC offset(or very little) and basically all the voltage amplification stages are working. Loads are put on a SS amp only after judging that all those points are OK. Putting loads on it before those items are confirmed, is just a waste of transistors, resistors and fuses.
            Last edited by km6xz; 12-06-2014, 08:50 AM. Reason: added comment...

            Comment


            • #7
              Certain amps don't work at all well off a lamp limiter when there's a load connected. Fine just to check voltages off-load, but occasionally an amp on-load will stick to one rail, light up the limiter and give the impression of a serious fault. Sessionette 75 and some trace Elliott models spring to mind as recent examples.

              Comment


              • #8
                It is easy enough to guesstimate, if the amp is under load, versus no load, the a 100 watt light bulb in series will drop the mains voltage to 50-60 volts if the amp normally has a 100 watt power consumption and when it is unloaded its mains will pop up to 90 or so volts. Test it. Measure the voltage at the AC plug of the amp with no load and 1/2 power and then try full power. I assume at some lower power output, say 10 watts into the load, the amp works fine because it has only a 20% deficiency in mains input but increasing the signal into the load, keeps dropping the mains voltage until regulators stop, bias shifts to class b then C. What amps work as intended on half voltage or with swinging in direct inverse proportion with signal level? I can't think of one but more to the point, NO tech would torture an amp that way. It creates a low rep rate oscillation where the frequency depends on the reservoir cap size. It is a form of feedback that is usually called Motorboating.

                Comment


                • #9
                  I was referring to certain amps failing with the load connected, but not with a signal. If the amp checks out fine unloaded with a limiter but fails with a dummy load connected this points to either the amp not being fixed, or it's one that won't run off a limiter.

                  Comment


                  • #10
                    Yes this amp works. ,!!!

                    I don't know what happened but amp came in turning on but with no sound. Speaker read about 4M and there was no DC on it. I put on a dummy load and scoped the output which looked terrible and the preamp out which looked normal. Switched to the other channel to see if it was any different and whoops, master volume only works for one channel not both. Volume on the other channel was on 10. When I turned on the amp the 4 amp fuse blew in about 1 second. Nothing looked scorched so I repkaced the fuse and put it on lamp limiter just in case. After that it showed what looked like a weird repeatedly shorting phenomenon so I haphazardly checked some caps and transistors in the output section, found nothing wrong, realized I was searching around with no real direction and decided to come here and ask.

                    I don't know if I had an error of settings or my scope probe was making poor contact or something, but initially scoping the speaker jack led me in all kinds of wrong directions. this thing just needs a new speaker

                    I think it's extremely funny I came in saying hey this amp doesn't work and multiple people were like, nah it probably works
                    Thanks for everyone's input

                    Comment


                    • #11
                      A solid-state relay (SSR) is an electronic switching device that switches conduction states when a small external voltage is applied along its n-type and p-type junctions. SSRs consist of a sensor which responds to an appropriate input (control signal), a solid-state electronic switching device which switches power to the load circuitry, and a coupling mechanism to enable the control signal to activate this switch without mechanical parts. The relay may be designed to switch either AC or DC to the load. It serves the same function as an electromechanical relay, but has no moving parts. This custom essay writing service*** will help you to write your future writing works.

                      *** Spam link removed by moderator 11/17/2017.
                      Last edited by Steve A.; 11-17-2017, 08:38 PM.

                      Comment


                      • #12

                        Comment


                        • #13
                          Spambot, what did you expect?

                          The telltale sign, red flag, whatever, is the "payload/warhead" , the included link included hoping maybe you click it.

                          The rest is just stuffing, because Forum Software is intelligent enough (meaning not that much) to reject or quarantine posts without text, containing only a link.

                          Text is generic, and roughly mated to expected Forum content, in this case any one with "Electronics" in the name.

                          By the way, the clumsily written catchphrase
                          This custom essay writing service will help you to write your future writing works.
                          does not exactly bathe in a shining light their written writing lessons.

                          By the way, checked their site, looks like another of those " start your own internet business and earn money working at home" book examples.
                          Last edited by J M Fahey; 12-26-2014, 03:02 PM.
                          Juan Manuel Fahey

                          Comment


                          • #14
                            Agreed. I reported the post as a Spambot with a different type of narrative to tempt the reader to click the link.
                            Last edited by Tom Phillips; 12-26-2014, 03:54 PM.

                            Comment


                            • #15
                              Wow.
                              What a site.
                              Lazy SOB students can't even write there own papers.
                              It is a form of cheating and plagiarism.

                              Comment

                              Working...
                              X