Announcement

Collapse
No announcement yet.

Ampeg SVT-4 Pro

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

  • Ampeg SVT-4 Pro

    Hello all,
    This is my first post on this forum. I have been a long time member at Diyaudio and was referred by a member.

    I have an SVT-4 Pro that is being really difficult to work on. I have seen a few of these, but this one is a head scratcher.

    Quick background- This amp was given to me after two people tried to fix it but hit a wall. I think I have restored it to where the wall was hit.

    I have replaced all outputs with UNMATCHED devices as per the owners instructions. He does not believe they are necessary. (Thats a whole other thing)

    I replaced all emitter resistors and any gate resistors that were burnt or out of spec. I have replaced the bias transistors and tested all diodes around them. I did replace one 10v zener. I have also replaced the other MPSA and MJE transistors and the 5532 behind them. I also checked out all diodes and resistors in front of the 5532 for shorts/open.

    Here's whats odd. When the amp is brought up via a variac the unit starts drawing excessive current early on. ~25vac at the primary and about +-15v dc on the output rails. The rails look symmetrical.

    I find one transistor conducting on its own and the opposite bank of 5 conducting evenly. Hence the heavy current draw.
    If I bleed the rails down and bring it up again the same transistor, or another single transistor will start conducting and it's opposite bank will start evenly.
    Which single transistor starts first seems to be random.

    My confusion is that the amp should come up until the bias starts with no problem as the gates should be held off by design, but the +-15v rails that the bias runs off is only at 3v or so and the bias is at ~2mV when a transistor latches, so bias isn't awake yet.

    I should add that the output section will behave the same way with the low voltage rails connected or not.

    Any ideas?

    Thanks!!

  • #2
    Well, you should have your head examined for putting money into it.
    Two other people worked on it, and we don't have a clue what they did...
    And, its a whatever amp, made in wherever. It really wasn't "meant" to be fixed.
    And so the combination of failed attempts to fix it,
    putting unmatched devices into it,
    combined with what it is in the first place, make for a difficult assignment, to say the least.

    Comment


    • #3
      Hi imix, welcome to the forum

      Guruman likes to hear himself make grand editorial comments. Ignore it.
      Education is what you're left with after you have forgotten what you have learned.

      Comment


      • #4
        Fake transistors that can't withstand the supply voltage? Wrong transistor pinout or installation? NPNs and PNPs mixed up? (15V is in the ballpark for base-emitter breakdown)

        Parasitic oscillations? Do you have a scope? If so, look for oscillations on that transistor that's latching and conducting prematurely.

        Also recheck the connections. If the transistors are MOSFETs and the gates aren't connected to anything, that would also explain the symptoms, as they would tend to turn themselves on randomly through leakage. Could be hairline cracks in PCB traces, etc.

        The opposite bank of transistors conducts because, if one side starts to leak excess current, the feedback loop will turn the other side on harder to keep the output at 0V. So that's a good sign, it suggests that the driver stages and one half of the output stage is working.
        Last edited by Steve Conner; 06-28-2012, 10:15 AM.
        "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

        Comment


        • #5
          You may be heading in circles.
          Some amp topologies do not like being brought up slowly.
          Plus, you have a double whammy with the current limiter throwing voltages off.
          If you are confident that you have found all of the failed components, turn on the amp at full line voltage with the lamp limiter in series.
          If there is still a gross fault that you cannot identify, it will be apparent from the lamp brightness.

          Comment


          • #6
            Hey guys! Thanks for the responses.

            This was brought to me because I have worked on a few and I have all available resources as far as service docs. If it was mine I'm not sure I would continue.

            I believe the irf's are genuine as they came straight from Mouser. In my two decades of weekly orders from them I have not had an issue with fakes yet. Doesn't mean it's not possible of course.

            I am going to start checking and rechecking components now. It's great to have some sets of virtual fresh eyes on thing.

            Thanks!

            Comment


            • #7
              Ok, well it didn't take long to find. The gate resistors on the negative side of the channel that was running away are open. Explains why it was taking off. I also found one open gate resistor on the negative side of the other channel.

              Comment


              • #8
                More good news. I found a pair of 47k resistors which supply Vd to the MJE345/350 blown open. I have a feeling they were part of the original fault and were not replaced in pervious repairs. The amp now comes up without drawing an unusual amount of current. Gonna check a few things and start biasing it really low.

                Comment


                • #9
                  Originally posted by imix500 View Post
                  Hey guys! Thanks for the responses.

                  This was brought to me because I have worked on a few and I have all available resources as far as service docs. If it was mine I'm not sure I would continue.

                  I believe the irf's are genuine as they came straight from Mouser. In my two decades of weekly orders from them I have not had an issue with fakes yet. Doesn't mean it's not possible of course.

                  I am going to start checking and rechecking components now. It's great to have some sets of virtual fresh eyes on thing.

                  Thanks!
                  The best approach can be taking every part out testing it, replacing or reinstalling it,
                  but still, parts can test good, and fail when put into a loaded condition.
                  And therefore you have to ask if it's really worthwhile to put XXX time and money into it.
                  You can get an amp like this working, but being what it is, will usually fail for another design or materials quality reason later on.
                  If you were paying a tech by the hour, it would not be worthwhile. Face it, even at $35 an hour, that's a lot of hours.
                  But if doing it yourself, it's maybe a good educational pursuit.

                  But from a "value" point of view, you passed the "silly" benchmark long ago. There are far far better amps to put time and money into, from a "value" point of view.

                  Comment


                  • #10
                    Yeah I hear what you're saying on this one. As long as the owner keeps agreeing to another hour of work I'm making it a mission.

                    I think I've pretty much figured out whats going on and it seems to be a combination of issues that have been fighting each other as this amp has been worked on.
                    R11 and R12 failed pretty quickly after replacement. All the outputs and MJE's still check ok, but R135,R136,R235,R236 are now open and many of the 1.5K feeding the base of Q108,Q109 and Q208,Q209 are also open. Q108,Q109,Q208,Q209 are now reading 300 ohms from base to emitter. The transistors test ok out of circuit.
                    So, I think having Vd restored for a few seconds revealed whats going on by causing failures around Q108,Q109. Certainly not the best way to see it, but at least it's something I can follow.
                    I am going to replace all the bad resistors and finish checking the rest of the passive components.

                    Comment


                    • #11
                      Originally posted by soundguruman View Post
                      But from a "value" point of view, you passed the "silly" benchmark long ago. There are far far better amps to put time and money into, from a "value" point of view.
                      Some people pass the silly benchmark with every post!

                      The trouble with solid-state power amps is that a failure of an output device can take out a lot of parts in a kind of domino effect, and not all of the failures are obvious. If you don't catch them all, then the next time you apply power it blows a bunch more parts (including half of the ones you replaced) and so on.

                      I heard a joke about the guy who thought he could fix a switchmode power supply by just replacing the visibly burnt parts, but you could tell the same joke about many solid-state power amps, even expensive ones.
                      "Enzo, I see that you replied parasitic oscillations. Is that a hypothesis? Or is that your amazing metal band I should check out?"

                      Comment


                      • #12
                        Used to be an SMPS had a primary power source and a switching transistor or two driving a little transformer with a lot of secondaries. Nowdays they are in multiple sections one feeding the next. SO the high voltage supply won;t even start unless the low voltage supply is running to power it.

                        I just spent the evening figuring out a new Samson SMPS. I got it, but it took some figuring. BAd drive IC on the high voltage section. Nothing visibly damaged at all. Bunch of component changes in an ECO, so I broke my own rule. I did the ECO mods before fixing it. I'm bad, I should be punished.
                        Education is what you're left with after you have forgotten what you have learned.

                        Comment


                        • #13
                          I know what you mean about switchers. Prior to this amp I just repaired a stack of QSC PLX amps. It's a different ball game rebuilding those supplies.

                          Oh- I'm actually at the hospital hopefully having our first today, so it will be a few days until I can get back to the Ampeg.

                          Thanks!

                          Comment


                          • #14
                            Hello all, just wanted to follow up with what hopefully is the conclusion to this repair.
                            I have grown to loath this amp and have made a deal with the owner to replace the entire amp pcb and take a hit on labor to get it off my bench.
                            Now, we just have to wait for Loud to find a new manufacturer for the boards and receive delivery in 6-12 weeks.
                            Thanks!

                            Comment


                            • #15
                              Originally posted by imix500 View Post
                              Hello all, just wanted to follow up with what hopefully is the conclusion to this repair.
                              I have grown to loath this amp and have made a deal with the owner to replace the entire amp pcb and take a hit on labor to get it off my bench.
                              Now, we just have to wait for Loud to find a new manufacturer for the boards and receive delivery in 6-12 weeks.
                              Thanks!
                              It's not worth putting money into.
                              You might replace the board, but it will fail again soon.
                              remember my original answer? Two months ago?
                              What we have learned about Chinese amplifiers?

                              The thing is, if you put the board into it,
                              the customer will blame YOU when it fails again.

                              "Well, you should have your head examined for putting money into it.
                              Two other people worked on it, and we don't have a clue what they did...
                              And, its a whatever amp, made in wherever. It really wasn't "meant" to be fixed.
                              And so the combination of failed attempts to fix it,
                              putting unmatched devices into it,
                              combined with what it is in the first place, make for a difficult assignment, to say the least."

                              Comment

                              Working...
                              X