Announcement

Collapse
No announcement yet.

MiniDSP and Amp help - static/whine

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

  • #31
    Excellent news!
    Expensive caps need time to break in. More expensive caps take a long time. Cheap caps sound great right out of the box.

    Why I don't spray in first gear: http://s1138.photobucket.com/albums/...t=100_2585.mp4

    Comment


    • #32
      The black wire connected to the RCA socket grounds - what connection is that?

      Comment


      • #33
        Mike that is the ground from the amp input connector (it came with the amp) although looking at the diagram that comes with the amp (on the PE product page) it's not labelled as ground - it's not labelled as anything and I must have assumed it was the ground. Is it?!

        If you mean the larger gauge black it was a jumper I was using to connect the ground terminals from one pair of rcas to the other. I cut it short when it didn't do anything.

        Comment


        • #34
          Originally posted by scottvalentin View Post
          Mike that is the ground from the amp input connector (it came with the amp) although looking at the diagram that comes with the amp (on the PE product page) it's not labelled as ground - it's not labelled as anything and I must have assumed it was the ground. Is it?! That is the amp's input signal ground common to the L & R inputs ... same as with a 3.5mm plug - L, R and gnd.

          If you mean the larger gauge black it was a jumper I was using to connect the ground terminals from one pair of rcas to the other. I cut it short when it didn't do anything.
          I was asking about the heavy black wire. I would think that the mini-DSP's 4 outputs all have a common signal ground from the board, why the black wire didn't do anything.

          Glad your up and running. But disconcerted that removing a signal ground on the amp's inputs solved the problem. That should be connected and working properly. Seems to be a ground loop somewhere else. I was going to suggest to take the PC out of the equation. Either a laptop on battery power or a phone/mp3 player as a source (with that signal ground reconnected).

          Comment


          • #35
            The whine occurred even with no source plugged in, and also with my phone as a source. So I don't think it was the laptop or is related to the source side on the MiniDSP board. Thanks for clarifying about my jumper.

            The supplied wiring diagram doesn't list the black wire of the little harness as a ground, but I agree, not sure why that made the whine disappear. I will leave it for now so I can actually play with some crossover alignments! Again, really appreciate the help along the way with this issue.

            Cheers

            Comment


            • #36
              Originally posted by scottvalentin View Post
              The supplied wiring diagram doesn't list the black wire of the little harness as a ground, but I agree, not sure why that made the whine disappear.
              It's implied. Disconnecting it likely eliminated a ground loop that, ideally, should be addressed in the power scheme.

              Enjoy.

              Comment


              • #37
                Originally posted by Millstonemike View Post

                Disconnecting it likely eliminated a ground loop ...
                ​Yes. You can't have heavy output currents flowing in the shield wire of an input, and having two channels of heavy current, each at a different frequency, is the kiss of death. Unfortunately, it is sometimes hard to visualize current flowing through "grounds", and the only "rule" for avoiding ground loops that I know of is the "star ground". Someday I'll try to draw a series of pictures that shows why this happens. Each picture would show how an extra ground created a path for current to flow through the input shield . Unfortunately, there are a lot of combinations.
                Free Passive Speaker Designer Lite (PSD-Lite) -- http://www.audiodevelopers.com/Softw...Lite/setup.exe

                Comment


                • #38
                  Originally posted by neildavis View Post
                  ... and the only "rule" for avoiding ground loops that I know of is the "star ground".....
                  from:
                  http://www.analog.com/en/analog-dial...-grounded.html
                  Star Ground The “star” ground philosophy builds on the theory that all voltages in a circuit are referred to a single ground point, known as the star ground point. This can be better understood by a visual analogy—the multiple conductors extending radially from the common schematic ground to resemble a star. The star point need not look like a star—it may be a point on a ground plane—

                  http://basicelectronicsengineering.b...e-created.html
                  http://basicelectronicsengineering.b...und-plane.html
                  "Not a Speaker Designer - Not even on the Internet"
                  "If the freedom of speech is taken away, then dumb and silent we may be led, like sheep to the slaughter."

                  Comment


                  • #39
                    Originally posted by neildavis View Post

                    ​Yes. You can't have heavy output currents flowing in the shield wire of an input, and having two channels of heavy current, each at a different frequency, is the kiss of death. Unfortunately, it is sometimes hard to visualize current flowing through "grounds", and the only "rule" for avoiding ground loops that I know of is the "star ground". Someday I'll try to draw a series of pictures that shows why this happens. Each picture would show how an extra ground created a path for current to flow through the input shield . Unfortunately, there are a lot of combinations.

                    A star ground is desirable and technically sound. But to achieve it, consider modern systems; switch mode power supply(s) feeding power amps, the PC and the mini-DSP (either directly via wall wart, or indirectly via USB from the PC's PS). While switch-mode PSs are isolated from the "mains" via their rectification bridges, IMO, many with 3-prong plugs tie their negative DC output back to the AC ground pin. And single DC supply digital amp boards tie their negative DC supply to their signal ground by default.

                    It's different from older days when transformers were the predominant PS architecture. Earth ground was a safety ground and kept separate from the DC supply output.

                    Scott tried a separate wall wart for the mini-DSP and I presume it was two prong so no link back to ground. But was the PC and its switch-mode PS the source providing a link? He also tried a phone as input with no success. But was the mini-DSP powered off the PC and it's PS or the two prong wall wart for that test? It could be a bad amp board - perhaps a solder bridge or shorted cap bypassing the DC signal isolation between devices?

                    For this issue, you really need a controlled lab type investigation.

                    Comment


                    • #40
                      Originally posted by Millstonemike View Post

                      ...For this issue, you really need a controlled lab type investigation.
                      ​I agree that sometimes you can't implement a star ground scheme. That's why I said it was the only "rule" to follow, but there are many real-life applications where it is impossible to apply that rule. In fact, I just had a long struggle with the Dayton WiFi module, which was "grounded" through the antenna to the chassis, which was in turn "grounded" to the amplifiers via the mounting holes. It can be a difficult problem.

                      ​But that doesn't mean you have to resort to experimentation as the only solution. For many cases, you can draw out the system "circuit" and model the grounds as low impedance paths and predict where the current will be flowing. As I said, I'd be interested in trying to draw a series of pictures that explained the current flows through the grounds in a typical system of DSP, sources, amps and power supplies. The diagrams would be along the lines of the discussion of ground loops in the LM3886 data sheet (see page 21).
                      Free Passive Speaker Designer Lite (PSD-Lite) -- http://www.audiodevelopers.com/Softw...Lite/setup.exe

                      Comment


                      • #41
                        Hi Scott, It sounds like you've got it figured out, which is great. But I had a similar experience with a laptop power supply and a sure 4-channel amp, and in my case at least, the answer was a little different than what others have posted so I thought I'd mention it.. The 'whine' in my set up was audible from across a large room. Anyway, just when I was about to give up, I disconnected the laptop brick from the amp, kept it plugged into the wall, and listened to the brick itself. And there was the whine! Same irritating fluctuating rhythm and everything, just unamplified. So a cheap ebay power supply has solved my problem. (I've listened to a few other laptop bricks since then, and they've all had that whine.

                        Gordon
                        Gordon
                        --------------------
                        Speaker projects:
                        Microbe: http://i234.photobucket.com/albums/e...Picture005.jpg
                        Extremish: http://i234.photobucket.com/albums/e...h/IMG_0013.jpg
                        Seas27TBFCG/VifaXT18: http://i234.photobucket.com/albums/e...Picture155.jpg
                        in progress: http://s234.photobucket.com/user/gor...-way%20project

                        Comment


                        • #42
                          Thanks for that Gordon, I wouldn't have thought to check that!

                          Actually, I have an update to add. I since re-tested everything, and the whine was back. Turns out when I disconnect that ground wire (above) the other amp side gets the whine, and I only had one amp connected to speakers when I tested previously.

                          So after much trial and error, I started with reconnecting that ground wire (Mike your gut was not wrong, that didn't fix it). I then just used a jumper to go from the amp RCA input grounds to power ground, to amp power in ground, and so on.

                          I do have it fixed, but it took 3 more ground jumpers going back to the power ground from the RCAs and even just adding a second cable to go from the power ground to the amp power ground made a difference and I can no longer hear the whine. I will test and see if my wife can hear it later - her ears are WAAAAAY better than mine!

                          Comment


                          • #43
                            My wife does hear the high pitched whine faintly, and I can hear it if I put my ears closer to the speakers. Not sure what else to do!! I tried a jumper in with the current, almost acceptable setup, and ran it from a lot of different places, but could only manage to make the whine louder. I tried from the amp power grounds to the mini dsp board, to the inputs, and more between the amp RCA input grounds, and the only difference was that a few made it louder.

                            I think I will move on to getting these going since I can't really hear the noise. A new thread will probably make more sense.

                            Comment


                            • #44
                              Did you try to synchronize the clocks between the two boards, as Neil suggested? It's not that complicated on these boards. There's actually a test point that allows you to do it. I have instructions on how to perform the rework posted in this document:
                              https://docs.google.com/document/d/1...XTTrtSrVw/edit

                              Check out page 11, section 18.a.1.
                              DARPA Jr - 2015 InDIYana Winner - RS180-8 + RS100P-8 + ND25FA
                              The Aria's - RS150-4 (or RS150-8) + XT25SC90
                              The Mariposa's - TEBM65C20F-8 + ND16FA
                              The Canzonetta's - RS100P-8 and ND16FA
                              AudioSource AMP-100 Mods OR Pyle PAMP1000 Mods

                              Comment

                              Working...
                              X