Announcement

Collapse
No announcement yet.

Behringer DSP-8024 Help Needed

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

  • #16
    Don't replace the crystal until you prove it is faulty. Every time you or anyone, changes parts randomly or semi-randomly, the clear diagnosis becomes more ambiguous and less likely to be repaired successfully.
    Try injecting 11.2896 into the processor port. If a crystal fails to oscillate it is not high odds that it is the crystal. The crystal is an extremely high Q reactive resonate network and requires three things to begin and sustain oscillation, gain, noise and phase shift. The chip is connected to supplies the gain and therefore the noise and the two chip caps mounted next to the crystal provide the phase shift. Any leakage or value change in either of those caps can prevent sustained operation.

    Comment


    • #17
      And just for old school sake, any crystal that doesn;t want to sing, sometimes will start up if you rap on it. Use a plastic screwdriver handle or something. Oh. doesn;t always work of course, but it takes seconds.

      Oh...

      The 11.xxxMHz and 12.xxxMHz rocks are not oscillators, they are bare crystals. Yes they are in oscillator circuits, but they are not the little oscillator circuit cans. Q2 and Q3. Each is driven by a gate in IC45, a 74HC00. Further, each of those gates is controlled by another gate as an inverter, so ONLY ONE of them will oscillate at a time as selected by control signal MSEL at IC45-1 . Then the two outputs are ORed together by another gate in the IC, so whichever one is selected ultimately comes out IC45-8. If you have clock at IC45-8, then they are all working, at least enough. As far as I can tell, both are free running when selected. One or the other should always be going, but never both.

      Crystal Q1 is 12.0MHz and is also a crystal, not an oscillator, it is directly driven by IC36.
      Education is what you're left with after you have forgotten what you have learned.

      Comment


      • #18
        Is there a schematic available?

        I did resolder the crystals, no luck. The amplitude on the crystal located right by the PS was lower in amplitude than the middle one.

        Comment


        • #19
          Here's a pic of the MHS S-80C31-16 microprocessor.
          http://www.datasheetcatalog.org/data...c/mXyzuzvq.pdf

          The index corner is on the bottom left of the chip in the pic.
          The RESET pad is tied to Pin 4:

          RST
          A high level on this for two machine cycles while the
          oscillator is running resets the device. An internal
          pull-down resistor permits Power-On reset using only a
          capacitor connected to VCC. As soon as the Reset is
          applied (Vin), PORT 1, 2 and 3 are tied to one. This
          operation is achieved asynchronously even if the
          oscillator does not start-up.

          The left pad of the two smaller pads to the right of the cap is tied to Pin 10. It has 5.0Vdc on it. The other is ground.

          TD (Timer 0 external input)

          The 10uf/50v cap is OK. I put a meter on the little 22pf caps and the seem OK. After doing that it booted up and ran for awhile. I did short a resistor fro the RESET pad to ground. Nothing happened. Should I short the reset to the 5.0vdc?

          Click image for larger version

Name:	IMG_0634.jpg
Views:	1
Size:	2.07 MB
ID:	826775

          Comment


          • #20
            Seems to be acting in intermittent mode,so it actually boots !,flexing the board may help.......
            I would reflow the processor as a starting point.
            Then if that fails repace the xtal(s) attached to it...or the other way round ,if you don't fancy a reflow,?
            Then take it from there.

            Comment


            • #21
              Re-flowed the processor. It started up and I thought that fixed it. But then it went back to flashing the LED's. One thing it does when it does boot up all the way is display a message of Bad Check Sum Error, Memory Cleared. If I turn it off and and it boots back up fully it doesn't show that message again however.

              I've tried flexing the board doesn't seem to make a difference. I'm thinking it's a weak output from the xtal. On the scope it's amplitude is 0.34v compared to the middle xtal's 0.90v.

              Update: Found some 12.000Mhz HC-49U on ebay.
              Last edited by Tejaus; 09-14-2012, 03:50 PM.

              Comment


              • #22
                The DSP-8024 seems to be acting just like the mixers in this thread. http://music-electronics-forum.com/t24165/

                Comment


                • #23
                  I hate to say it. But this is begining to sound like a double sided board with an intermitant feed through. You may never find it. If you have a circuit viewer type microscope it would help. Did you reflow the processor using flux?

                  Comment


                  • #24
                    I have some flux that came with a Chip Quik kit. I could try that I suppose. It's pretty hard to trace out the board even with some 2x readers and a magnifying glass LOL.
                    A schemo would be so helpful.

                    I've tried flexing the board and freeze spraying it. It doesn't seem to make it stop or start. I'm tending to think the xtal is flaky. It's signal is about 1/3 the amplitude relative to the middle one.

                    I bought some 12MHZ xtals off ebay, should be here next week.

                    Comment


                    • #25
                      Spend $3 and get a bottle of flux and a small brush. I used to repair a lot of boards using a Hako hot air extractor machine with a single tube tip. Have you tried tickling the crystal with freeze mist. My guess would be that the other crystals are on 5v supplies and that at one is on 3v. Just a guess but fairly common.

                      Comment


                      • #26
                        The symptoms do not sound like a bad plate through hole, few of the chips are through hole so the vias and header connectors are about the only parts of that portion of the circuit that is impacted. I never really saw a problem with board plating on Behringer gear, in fact other than physical damage to a board, I do not remember ever having a defect that was related to the board itself and that means many thousands of repairs since for some time we were the only warrant service center west of NYC.

                        The rest test with the external pull down resistor tells you nothing other than the internal pull down is working as intended.

                        The amplitude of the oscillator needs to be around 0.5s for reliable operation and there is a good chance that what is being seen is loading on the high z circuit by the probe. If the clock is starting reliably, kicking in as soon as power is applied, it is not likely the xtal. The output of the clock will be at normal CMOS levels and should be 5v peak, regardless of what is seen across the xtal.

                        A clue as to the condition of the crystal would be measuring the frequency by comparing with a stable generator. You know it is oscillating and a counter would average jitter if it is unstable, so use a stable RF generator, preferably one that does not depend on a PLL which would have its own jitter, eiher DDS or a LC oscillator and compare the clock out in an XY mode on the scope to see if the cock is jumping all over. A xtal is so easy to verify if it is OK that is seems like a waste of time to keep replacing parts that are probably good. It also introduces ambiguity to the diagnosis. Rather than replacing things, diagnosis it. Repair and replacement of proven defective parts comes later.

                        Comment


                        • #27
                          The problems with the mixers were caused by a bad 32.768kHz crystal. Are you sure you don't have one of those that you've overlooked? It looks different to the other crystals, a tiny, long, thin, cylindrical silver can with two leads coming out of one end.
                          "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

                          Comment


                          • #28
                            Spend $3 and get a bottle of flux and a small brush. I used to repair a lot of boards using a Hako hot air extractor machine with a single tube tip. Have you tried tickling the crystal with freeze mist. My guess would be that the other crystals are on 5v supplies and that at one is on 3v. Just a guess but fairly common.

                            I tried the Chip Quik flux on the MP pins, worked great. Shot the MP and xtal with freeze spray, no effect.

                            A clue as to the condition of the crystal would be measuring the frequency by comparing with a stable generator. You know it is oscillating and a counter would average jitter if it is unstable, so use a stable RF generator, preferably one that does not depend on a PLL which would have its own jitter, eiher DDS or a LC oscillator and compare the clock out in an XY mode on the scope to see if the cock is jumping all over. A xtal is so easy to verify if it is OK that is seems like a waste of time to keep replacing parts that are probably good. It also introduces ambiguity to the diagnosis. Rather than replacing things, diagnosis it. Repair and replacement of proven defective parts comes later.


                            I have a Leader LAG-125 generator that will probably work. Low Distortion Audio Generator LAG-125 Equipment Leader Elec
                            But please forgive me as I'm a first class noob with oscopes. With this being intermittent it's hard to find what is and is not defective. Do you know how the fault protection circuit trips? That's what seems to be happening.

                            The problems with the mixers were caused by a bad 32.768kHz crystal. Are you sure you don't have one of those that you've overlooked? It looks different to the other crystals, a tiny, long, thin, cylindrical silver can with two leads coming out of one end.


                            Nah, it just has those HC-49U case xtals. I have 5 on the way from ebay for 3.00 bucks.

                            Comment


                            • #29
                              Since this post didn't get resolved it appears I thought I might add a bit of what I experienced with this same unit.
                              After checking the PS voltages and ripple, I found that if I leave it turned on for a while, 30 minutes or so with the lights flashing,
                              then turn it off and back on again, the symptom seems to disappear and I am able to control the unit as normal.
                              It may help to hold the IN/OUT button down when turning it back on again
                              Hope this helps.
                              My guess is that this indicates the problem may be a leaky cap.

                              Comment


                              • #30
                                I had a similar problem recently and when I checked the voltages on LM317, I noted that the output voltage was around 4.3 V. The output voltage should be 5V. The input ripple voltage was also high - close to 3.87 volts. So I decided to check the Capacitor (C77) and noted that it was faulty. Replaced it with 2200uf 16V electrolytic capacitor and the 8024 started working properly. Thought i should share this will all.

                                Comment

                                Working...
                                X