Announcement

Collapse
No announcement yet.

Peavey CS-800 (C) unexpected behavior has me stumped

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

  • Peavey CS-800 (C) unexpected behavior has me stumped

    Actually, I'm sure if I had the patience I could figure it out, or I could just replace *everything*, but I'm trying to avoid that.

    Amp was originally delivered to a friend of mine with a couple of shorted output transistors, RCA #68271, in channel B. He replaced the entire set (all ten on that channel) with MJ15024, but the dim bulb test indicated the unit was still shorted so he sent it on to me for further diagnosis.

    Eventually discovered that one of the pins of one of the op amp ICs on the driver board essentially wasn't connected at all, as well as one of the predriver transistors blown. Replaced both ICs and their sockets, and both predriver transistors -- Q5 with MJE5730 and Q6 with TIP48. This had the amp working but with intermittent "scratchy" noises (sounded exactly like a dirty pot, but even with the level all the way down). I recognized the sound as a leaky transistor (had a similar problem with a Lafayette receiver decades ago). Followed around with an oscilloscope, determined it to be the "Transient Supressor" (Peavey's spelling) transistor Q1, a P-channel JFET, which I replaced with 2N5461. At this point the amp seemed to be completely working, although still connected with a 150W floodlight in series with the AC line. Both channels produced clean sound.

    That said, when the amp is first turned on there is a considerable DC pulse -- not long enough to trigger the triac board, but enough that, if a speaker is connected, it pulls enough current to light the floodlight all the way and prevent the power supply caps from charging up high enough to stabilize the amp and end that pulse. This alarmed me at first, but I happen to have another identical amplifier here that I personally own, so I know they all produce a considerable "pop" noise when the power is turned on. I plugged MY amp (which works fine) into the series bulb and verified that it behaved the same way, so I passed this off as normal behavior. For testing, I power on with no load and then connect the output after the bulb dims down.

    So I run it for a couple HOURS, with the bulb in series, everything is fine. Sounds great, heatsinks not even noticeably above room temperature, I'm thinking this is great and I'm done.

    Change the speaker wiring around a little bit to try bridge mode. Amp continues to work very well.

    Remove the bulb from the circuit - full current now. Unhook the speaker out of habit, power it on, connect the speaker (still in bridge mode). It continues to work fine. Turn it off. There's the pop which I expected. Turn it back on. There's also the pop I expected, but then another pop. I give it a second thinking the channels just didn't come up at the same instant and channel A's pop had just been passed through by channel B which was already up. Wrong answer. Two seconds later there's another pop, then a hiss, then a sizzle and smoke.

    Power off (quickly), pull it apart to assess the damage. The first thing I notice is R30, which is two parallel resistors between CR11 and CR12 (the thermal tracking diodes in the bias circuit) is charcoaled. Then I notice the emitter leg is blown off of Q7, which is the positive current limit transistor. So far I've got CR13, CR19, R24, R25, R30, and Q7 gone on the driver board, and R1 (the 100ohm resistor from the positive-side MJ6387 base to the output) on the power board.


    I imagine PART of the cascade being caused by R30 going open, which would more or less bias the outputs straight into a short, which is probably what toasted the 100-ohm R1 which happened to be sitting between Q9 trying to drive it to +V and the output that was being pulled to -V by the negative outputs... I suppose a similar path of destruction occurred through R25 and Q7 I can't imagine what took out R24, since CR15 seems to be okay...

    ... but above all of that... beyond how the cascade failure happened... the reason I'm here... I'd like some input as to what caused the first failure. I haven't seen this one before.

  • #2
    Please post your schematic or link to it.

    It is all about current. Your 100 ohm resistor R24, for example, burns because too much current came from SOMEWHERE. My first impression? Open R8 in the power board.

    My procedure:
    1. Measure resistance between red and black speaker posts. SHorted together? Bad crowbar triac SAC187. Not short, good.
    2. Open unit from rear, eight screws ( I think) and the two power assemblies lift up and out to the sides on the slack in their wire harnesses. The four pin connector on each disconnects it from the circuit. By running one at a time, your bulb will tell you which one (not usually both) has a short. Leave the good one unplugged.
    3. Next to the four pin connector is your inductor. It is a coil of wire, next to it is R19 5.6 ohm. Measure that resistor right at its leads. If it checks as a SHORT, zero ohms, that is GOOD, means the coil is intact. if the resistor measures 5.6 ohms, then your coil is open, likely cracked loose at one end.
    4. I brought the inductor in because it is a handy test point. Look at the two huge filter caps in the power supply. Ground in the center, then each has a screw that is one of the main rails. Now on the module measure resistance between the inductor and each power rail. This finds shorted output transistors.
    5. Once any shorted outputs have been found or replaced, ALWAYS check all those 0.33 ohm 10W resistors for opens. R11,13,15,etc. R8 goes open, then R24 becomes a current path. And check small resistors R1,2,3,22.
    6. Pull the driver card, does the thing settle down?


    A thing about bulbs, they are fine for saving fuses and not blowing things up, but amps are not made to operate on the bulb. It runs the amp on spongy low voltage.

    7. Make one channel 100% before even thinking about bridge mode.
    8. Does the channel pass signal with NO LOAD? is it clean? Does the waveform change in ANY way if you apply a load?


    Emitter Q7 blows? Then it was a current path for a failure upstream, whaddaya know, same one that took R24. Once again I am tied to R8.

    R30 simply should never burn up, make sure CR12 is connected to the right place on the driver card.
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      Hello. Thank you for your reply. I am at work right now but I will provide a link to the schematic: http://users.glytch.net/emolatur/cs800c.pdf

      R8 is fine. That was my first theory as well. It measures the same as the other emitter resistors. Since the "power board" doesn't actually have very many components, I went ahead and tested each and every one of them -- the only failure (on the power board) is R1, which is actually splattered across several other components.

      I don't know why R30 opened -- but I do know, if it does, you'll get all of the outputs going full-on and fighting each other (why didn't it blow the fuse?!) which will definitely toast R1... Q7 turning on (attempting to do its current-limiting job...) will then kill itself and CR13... I'm unsure of CR19... so I think it's fair to disregard those components and focus on what might have opened R30...

      R24 is a big mystery though. You used it as an example, suggesting that the current came from Q3 due to R8 being open. In addition to R8 NOT being open: even if it was, the other side of R24 hits C15 (which is not shorted -- I checked) or CR15 (.6V forward, open-circuit reversed) -- after CR15 it hits CR16 biased the other direction (also tests as a valid diode) or R29 (470 ohms) or R33 (4.7K) -- enough current to toast a 100-ohm resistor would more than definitely also toast a 470-ohm or 4700-ohm resistor (current is the same, resistance is higher, so the 470/4700 resistors would have higher voltage across them, thus more wattage, thus more spectacular toasty moments) . . . I am at a loss for how R24 died at all.


      Knowing (I hope) more about this amplifier than I do, can you explain the 'POP' at turnon (which is normal across all of the CS800s I've ever owned) and how it is managed? I'm hoping this will lead me toward the right area of the circuit. The amp was tested a LOT before putting it in bridge mode (I didn't bother explicitly mentioning ALL of the testing in the original post, in attempt to keep it a readable length) -- all output waveforms were clean (I gave it 1 kHz sinewave at various levels both with and without load, I use an electric water heater element as a dummy load) and saw no distortion or sign of trouble, before using an actual speaker to 'hear' things. I omitted a bunch of testing in my post BUT I can attest that it was clean and survived being cycled on/off with and without loads in stereo mode, and also played fine in bridge mode -- the only combination that killed it was being powered on with a load already connected in bridge.



      Originally posted by Enzo View Post
      Please post your schematic or link to it.

      It is all about current. Your 100 ohm resistor R24, for example, burns because too much current came from SOMEWHERE. My first impression? Open R8 in the power board.

      My procedure:
      1. Measure resistance between red and black speaker posts. SHorted together? Bad crowbar triac SAC187. Not short, good.
      2. Open unit from rear, eight screws ( I think) and the two power assemblies lift up and out to the sides on the slack in their wire harnesses. The four pin connector on each disconnects it from the circuit. By running one at a time, your bulb will tell you which one (not usually both) has a short. Leave the good one unplugged.
      3. Next to the four pin connector is your inductor. It is a coil of wire, next to it is R19 5.6 ohm. Measure that resistor right at its leads. If it checks as a SHORT, zero ohms, that is GOOD, means the coil is intact. if the resistor measures 5.6 ohms, then your coil is open, likely cracked loose at one end.
      4. I brought the inductor in because it is a handy test point. Look at the two huge filter caps in the power supply. Ground in the center, then each has a screw that is one of the main rails. Now on the module measure resistance between the inductor and each power rail. This finds shorted output transistors.
      5. Once any shorted outputs have been found or replaced, ALWAYS check all those 0.33 ohm 10W resistors for opens. R11,13,15,etc. R8 goes open, then R24 becomes a current path. And check small resistors R1,2,3,22.
      6. Pull the driver card, does the thing settle down?


      A thing about bulbs, they are fine for saving fuses and not blowing things up, but amps are not made to operate on the bulb. It runs the amp on spongy low voltage.

      7. Make one channel 100% before even thinking about bridge mode.
      8. Does the channel pass signal with NO LOAD? is it clean? Does the waveform change in ANY way if you apply a load?


      Emitter Q7 blows? Then it was a current path for a failure upstream, whaddaya know, same one that took R24. Once again I am tied to R8.

      R30 simply should never burn up, make sure CR12 is connected to the right place on the driver card.

      Comment


      • #4
        Originally posted by emolatur View Post
        can you explain the 'POP' at turnon (which is normal across all of the CS800s I've ever owned) and how it is managed? I'm hoping this will lead me toward the right area of the circuit. The amp was tested a LOT before putting it in bridge mode (I didn't bother explicitly mentioning ALL of the testing in the original post, in attempt to keep it a readable length) -- all output waveforms were clean (I gave it 1 kHz sinewave at various levels both with and without load, I use an electric water heater element as a dummy load) and saw no distortion or sign of trouble, before using an actual speaker to 'hear' things. I omitted a bunch of testing in my post BUT I can attest that it was clean and survived being cycled on/off with and without loads in stereo mode, and also played fine in bridge mode -- the only combination that killed it was being powered on with a load already connected in bridge.
        The Turn-On 'Pop', I suspect is from the power amp's input IC's U2 & U1 being powered up off of the main supply caps and the zeners.....roll of the dice when you watch that initial charging of the bipolar supplies.....one side will come up slightly faster than the other, then you have the LF feedback cap C6 (I think)....100uF/25V in series with R8 charging up. Since there's no delay in energizing the amplifier circuits....it all comes up at once, you get the turn-on pop. At BGW Systems, we used output relays that had delayed turn-on and fast-off circuits to open and close the relay, yielding quiet turn-on/off. Without them, we got the same turn-on pops. Similar Quais-comp circuit (750A) without the refinements Peavey added.
        Logic is an organized way of going wrong with confidence

        Comment


        • #5
          Can someone offer any comment on the suitability of MJE5730 for Q5 and TIP48 for Q6?

          Comment


          • #6
            I suppose they should function, as long as you keep in mind the original parts were EBC across the front and your new subs are BCE across the front.
            Education is what you're left with after you have forgotten what you have learned.

            Comment

            Working...
            X