Announcement

Collapse
No announcement yet.

IRS 2092 Class D Amp Kit - Oscillation or Normal?

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

  • #16
    Ok... will end the night with a close up of the protection circuitry.

    Again, the amp is working. But the voltages in this section are bizaro!

    There is 48 volts across the relay so the output is connected 22uH inductor! My iTunes sounds fine!!

    One thing about the transistor pairs, they are installed back-to-back. You can see from the images below, that the bases of T1 (right side) and T2 (left side) are connected (center traces). You can also see that the Emitter of T1 is connected to the Collector of T2 (bottom view image, bottom traces). There is nothing else connected there.

    There is 2.0v on the bases of T1 and T2. The Emitter of T1 has .4v. So that transistor is not turned on, or is it? The Emitter of T2 is at 0V. In fact, that point is tied to Ground. I updated the schematic.

    There is .6v on the bases of T3 and T4. The Emitter of T4 is at 0v. However, the Collector is at 0v (because the connected points go to Ground). The same scenario with T3. We have the proper Base to Emitter voltage but the Collector is at 0v - odd indeed.

    I tried to be very careful tracing this out while I inserted the parts. I checked the continuity from point to point during assembly.

    So this is a mystery!
    Attached Files
    Last edited by TomCarlos; 03-20-2020, 05:09 AM.

    Comment


    • #17
      That cannot be right. The relay probably has a diode across it coil. The resistor cannot be 390k, more like 390R if it is in series with the relay. The relay must connect to a transistor collector (most likely) or emitter. There will also be a connection from the main output to the transistors so they can detect any DC offset. Back to bank transistors are often used to detect bi-polar voltages, one doe the positive and the other the negative.

      I'd expect something like main out->resistor->back to back transistors-> relay driver transistors -> relay with resistor and back emf diode.
      Experience is something you get, just after you really needed it.

      Comment


      • #18
        Thank you Nick....

        I have been looking for equivalent circuits or to see if anyone has posted this somewhere. That resistor is indeed 390K. I posted a larger image of the PC board. You can see that resistor on the left side of the PC board, near the Speaker Ground.

        Before installing any components, especially the resistors, I measured them with my DMM. The color bands are hard to make out and who knows, someone might have made an error in grabbing a resistor from a parts bin.

        As we know, this circuitry is NOT needed for the amp to work. But my curiosity got the better of me, so I decided to install it and see what would happen.

        One thing we do know.... if the relay no longer had 48 volts across it, the contacts would open. So for the left side of the relay's coil... if that were to go "high", you would not have the 48v across it. So the contacts would open. And that is the question. What would make this circuit change it's state.

        Tom
        Attached Files

        Comment


        • #19
          Can you provide a clear image of both sides of the unpopulated board and one of the populated board? Maybe the relay is not a mechanical type. Still if the supply is 48V and the series resistance is 390K that's only 125uA and not enough even for a solid state relay.

          According to your diagram, the relay contacts are in series with the speaker and therefore the relay has to be populated and the contacts closed for the amp to work. This is in direct contradiction of the statement "this circuitry is not needed for the amp to work".

          Some very large pieces of this little puzzle are missing...
          Experience is something you get, just after you really needed it.

          Comment


          • #20
            Originally posted by nickb View Post
            Still if the supply is 48V and the series resistance is 390K that's only 125uA and not enough even for a solid state relay.
            Earlier he had mentioned there was a B+ connection to that side of coil, so 390K not in series.
            Updated version of 'output.jpg':

            Originally posted by Enzo
            I have a sign in my shop that says, "Never think up reasons not to check something."


            Comment


            • #21
              Nick:

              My comment of "this circuitry is not needed for the amp to work" means that the PC board would work if I was to remove the protection circuitry, remove the relay, and hard wire the to the 22uH inductor. After the inductor, the only thing that is needed would be C12, the .47uf cap (250v) to ground, the 10 ohm in series with C14 (the .1uf) to ground, and the 2.2K resistor to ground. The typical application schematic (generic schematic) has those items. (See schematic in #10). I have seen other applications where the only thing after the 22uH inductor is the .47uf cap.

              Pic 3 is the front end to this. With the exception of the components that I noted in the post above, there is nothing different compared to the typical application hookup.

              Pic 2 is from the heat sink to the back end. You can see the blue relay, the 22uH inductor next to that (black component), and the big grey component is the .47uf cap I just mentioned.

              Pic 1 has the .47uf cap and just in front of that is the 2.2K that goes to ground (mentioned above). Next to that is one of the 4004 diodes and next to that is 100K that connects between the 4004 diodes. In this picture, you can also see T3 (next to the electrolytic caps) and T4 which is next to the 2.2K. The larger of the electrolytic caps (Blue) is the 200uf cap. The smaller cap next to that is the 10uf.

              Pic 1 also shows T1 and T2 tucked in the corner next to hte relay. T2 is next to the relay. T1 is near the pc board mounting hole.
              Attached Files

              Comment


              • #22
                Here are some suggestions that might help you in the redraw.

                I see that two IN4004's and the 100K go to the amp output, not 0V as drawn.
                D33 is in parallel with the relay coil which needs about 7.5mA to operate. That's quite small.
                I think the transistor pinout is ECB not EBC which I think will make them into darlington pairs.
                Experience is something you get, just after you really needed it.

                Comment


                • #23
                  I'll desolder some parts to double check.

                  In the mean time, here is the 2N5551 (see image). I will desolder one of the transistors to verify the pin out.
                  Attached Files

                  Comment


                  • #24
                    Originally posted by TomCarlos View Post
                    I'll desolder some parts to double check.

                    In the mean time, here is the 2N5551 (see image). I will desolder one of the transistors to verify the pin out.
                    Read https://music-electronics-forum.com/...ghlight=2n5551
                    Experience is something you get, just after you really needed it.

                    Comment


                    • #25
                      Holy Pinout !!

                      The Collector is in the middle !! I need to find another version of the spec sheet and save it for reference. When I used my meter previously, I just checked to see if it showed a working NPN. I did not look carefully at the pin out mapping.

                      Nick, you da man. That changes everything. I will go back and redraw that schematic and look at the other items you mention above.

                      ::: Bowing head in shame :::
                      Attached Files

                      Comment


                      • #26
                        With some help, I think I am getting closer. And THANK YOU !!

                        Nick was right, these transistors are hooked up as some type of Darlington pair. I am still suspicious when it comes to the base of T4 and Emitter of T3 both going to Ground. But I am looking at the traces and measuring the voltage and there it is.

                        With respect to the 100K that connects to the 4004 diodes. My previous "0" meant I was measuring 0 volts. I did not mean to imply that it was connected to Ground. But I do not see that point connected to the Output.

                        I don't know if there is connectivity happening between the two sides of the board. All I can do is draw what I see and take voltage measurements.
                        Attached Files

                        Comment


                        • #27
                          The holes will almost certainly be plated through so there should be connectivity between the two sides at each one.
                          Experience is something you get, just after you really needed it.

                          Comment


                          • #28
                            Yes, that is true. The big holes are easy to trace. It is the smaller holes that are the challenge - like the one next to D34. It is the hidden traces that are challenging.

                            The BIG diode (D3) hanging off pin 15, with the help of the mother of all magnifying glasses, that one is marked as ON 4UD R532. I looked that one up. It is called a Damper Diode, rated at Reverse Voltage of 1500v and Forward Current of 3 amps.

                            One thing that I need to do - there are some zener diodes on the board that I have identified - ie Z1, Z102, Z103, Z104. You can see these on the bottom side of the PC board. I also assume the other diodes like D32, D33, D34, those must be something equivalent to 1N4007s.

                            My friend is ordering one of these boards. We will continue to trace this out with no extra components. So I am hoping the next version of the schematic will include the zeners I mentioned above.

                            And at some point in time, I will send a link to this thread to the manufacturer of the board. I know he is on the DIY Audio site. But in reading some of his replies to questions, they are very cryptic.

                            Comment


                            • #29
                              As Nick pointed me to another thread, note that NCC 5551 is a different pinout compared to the 2N5551. This kit uses the NCC 5551.
                              Attached Files

                              Comment


                              • #30
                                Originally posted by TomCarlos View Post
                                With respect to the 100K that connects to the 4004 diodes. My previous "0" meant I was measuring 0 volts. I did not mean to imply that it was connected to Ground. But I do not see that point connected to the Output.
                                You can follow it right to one of the relay 'switch' pins. So it has to be the output.
                                Originally posted by Enzo
                                I have a sign in my shop that says, "Never think up reasons not to check something."


                                Comment

                                Working...
                                X