Pioneer A-9: Rebirth...more like a heart/lung transplant w/- kidney failure

Discussion in 'Pioneer Audio' started by Tangerine, Jan 30, 2017.

  1. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Hello all.

    This thread will be documenting that path to getting my A-9 back up an running.
    I bought it a few years ago, it was not running and I assumed an easy repair was needed.
    It's remained in storage in a sub-tropical environment in Sth East Queensland, Australia (this fact has weight later in this post).
    The repair is being done by an engineer colleague of mine who is a hobbyist with vast experience.
    There are no pics as yet, they are coming and will be worth a thousand words.
    Current status (as yet power has not been applied to the amplifier):

    0. Corrosion of the chassis and shielding covers (remember sub-tropical...?).

    1. A prior repair has permanently shorted out the Protection Relay R3, which according to the circuit and manual description drives the main audio output (both channels) to the sub-harmonic Indication circuit, power meter circuit and also drives the Headphone outputs. Quite perplexing as to why this was done, but at this stage I assume the original relay (which is a specialised solid state device) has failed and repair by replacement was not an available option. The relay part number is ASR-027, and was recognised on eBay.

    Q1: What is your intent with this type of prior issue?

    Two choices:
    1. Leave it as is,
    2. Fix it to original specification. If you leave as is, there is a risk that if the audio outputs go DC then all the voltage is applied to the three circuits as described earlier. This could be disastrous to the internal circuits as the relay is supposed to remove the signal if a fault is detected. Currently this self protection cannot occur and the amp may have already been subject to this exact scenario.

    Ans: 1 is not an option!!

    2. The DC voltage regulator has burnt components. The regulator is a simple series regulator that produces +/-60 VDC regulated to the main amplifier board from +/-80 unregulated at about 100ma load. Hence the regulator is about 12 watts total. A prior repair has also been carried out on the regulator, resulting in many of the PCB tracks being built up with solder rather than just copper. I can't see that any of the original components have been replaced, but the failed components now need to be replaced. This would seem to be a common fault at this time. The prior repair has left flux all over the regulator board (a pet hate of mine) since flux can become conductive over time and heat. (reason: prior sloppy workmanship). The painful part is that prior tampering with this board (movement) has stressed the wire-wrap joints to the regulator, some have been soldered and others are unwrapping.

    3. There are several dry joints on the main amp board. This is something for your forum: The dry joints are caused by mechanical vibration of the relays, the relays produce a shock every time they open and close (not the solid state relay mentioned in 1.) but the high power relays protecting the driver and speakers. Over time this continuous shock produces dry joints to components with any mass, such as the inductors, relays etc....


    Way forward:

    1. Take some snaps of the original condition.

    2. Repair the regulator and load it off (out of) circuit

    3. Turn amp on and check ALL voltages for amplifier.

    4. Determine what else is wrong with the amp.


    Comment: there is no way this amp functioned in the current condition and as found, my assumption is the amp must have bought it on spec. The reason the amp did not smell on power up is that it had already burnt out and hence couldn't burn much more. Note: No fuses have blown, which means whatever happened thus far has not involved over-current conditions (which is a saving grace).
     
    jobrewer1983 likes this.
  2. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Naturally, if there's any intel on where to get A-9 spares I'd appreciate the help.
     
  3. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Update 1

    Attempted to get power onto the amp last night but failed miserably. :(
    Turns out the pre-amp board also has a series voltage regulator on board and obviously designed by the same engineer. It too has burnt components that needed replacing. While the pre-amp was out of the chassis, the main driver boards are easy to view and uncovered more corrosion on the heatsinks for the on board transistors (not the big black sinks :banana:, just the small ones).
    I couldn't consciously leave the sinks in that state and turn it on. So out came the main driver boards for de-contamination.
    The whole amp is now reassembled but I didn't have the heart to switch it on as it was getting late.
    I have some pictures of the surgery they are coming soon. :needpics:
     
  4. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Update 1a

    With all this, don't worry about anything till I do (and I'll let you know if I am, currently I'm not).

    Both the burnt (now repaired, but not tested) regulators can cause offsets in the DC balance of the amp. The amp is a DC coupled device (no AC coupling capacitors in the signal path, hence why it (will) sounds so good) hence if any regulator is producing an unbalanced output (like +45V and -60V or worse +0V and -60V, instead of +60V and -60V [all DC]) then the protection circuits sense that DC offset from the regulator, it will assume a signal path DC error and turn the speaker relays off and RL3.

    Unfortunately, from the original state of the amp the repair has to start from ground zero, the power supplies, and all of them. I will probably put mains power on it tonight and the next report should be much more enlightening as to the amps real repair prognosis.

    There are two predictable outcomes from here:
    1. The regulators are now ok and the amp just comes to life (best case), or
    2. The amp still has the original fault in the main amp or in the detection (protection) circuits (worst case).

    If you want to chase something for the cause, see if you can find the solid state relay part number ASR-027, but if you can't doesn't matter <-- I'm proposing a circuit redesign in this area for reliability, but only once the amp is actually up and going.

    Main heatsinks are fine, now cleaned and hence should function efficiently as designed.
     
  5. QSilver

    QSilver Super Member

    Messages:
    1,036
    If it was me, I wouldn't be switching it on until I was sure everything was working to factory spec. Sounds like that amp has failed monumentally at some stage and so your gonna have to rebuild each board starting with the most critical, the power supplies, DC regulators etc. All parts should be able to be subbed with new, readily available ones. I have an A-8 and an A-7 awaiting a rebuild but both of those do work fairly well.. they could just do with a refresh.

    If possible (I've not seen the schematic yet so I can't tell), disconnect the power supplies from the rest of the amp, and remove the output transistors from the circuit. This can either limit damage or prevent further damage. If you can disconnect the power supply from the rest of the amp, work on rebuilding it and getting it stable - even replacing the heatsinks if u need to, replacing the pass transistors is probably a must.... anything that looks to be burnt or damaged is automatically changed and anything within the same circuit needs to be checked out of circuit to make sure it is within spec.

    If you already know all this, I'm sorry. The A-9 (IIRC) is a NSA and so very very twitchy so great care is needed to bring it back from what sounds like the brink... but totally worth it. Good luck!

    :lurk:
     
    Tangerine likes this.
  6. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Here are three pics that roughly show what my tech is up against.
    IMAG0967.jpg IMAG0970.jpg IMAG0972.jpg
     
  7. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Update 3

    Well progress has been made.

    With all boards disconnected both regulators are up and running with the centre indicator panel lights all glowing.
    The L channel driver is ok but the R has damaged active components.

    Boards have been cleaned in a mentholated spirits bath.

    Next steps are:

    1. Do a static test of components on the input board.
    2. Apply a test signal to the front panel drivers to check indicator panel function.

    A separate thread is running, which asks about compatibility of A-7 boards for use in an A-9.

    C
     
  8. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Pic of input board.
    IMG_4828.JPG
    IMG_4829.JPG
     
  9. QSilver

    QSilver Super Member

    Messages:
    1,036
    There's nothing in that amplifier that cant be fixed. I'd only consider swapping boards as a last resort.... or if the board was very badly damaged. I'm not sure about the meths bath though... That could damage some components. But if you just check the resistors, replace any that looked to have a "baked clay" look to them, replace transistors and the electrolytic capacitors you should be fine. Don't replace or remove the styrene caps (tall and on long legs) don't even heat them. They'll be fine and a bit of heat for too long will destroy them electrically.
     
  10. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Update 4
    Right channel driver transistors 2SC2526 and 2SA1076 have died.
    There's a substitution thread that lists replacements as 2SC3858 and 2SA1494 respectively however they aren't direct replacements and I'm unable to determine if they are really suitable for the A-9.
    Having said that the original tranny's are available via eBay, from a US seller and hopefully not knock-offs.
    Have original transistors always been available or was there a need to substitute with those given above?
    Are the substitutes really suitable to use in lieu of originals.

    Cheers,

    Les.
     
  11. redk9258

    redk9258 Well-Known Member

    Messages:
    979
    Location:
    Illinois, close to St. Louis.
    Those transistors are most likely fakes. The last ones made were many, many moons ago! Never buy obsolete transistors from eBay. Use a modern substitute.
     
  12. redk9258

    redk9258 Well-Known Member

    Messages:
    979
    Location:
    Illinois, close to St. Louis.
  13. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    I'm afraid I've made the transaction. I knew to stay away from Chinese based sellers for the reasons stated. They may come in handy during the continuing fault finding process.
    I'll look for modern substitutes to use in the Amp as the final repair.
     
  14. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Noted from a thread on substitutes:

    "Sanken 2SA1494 and 2SC3858 from B+D Enterprise or Digikey for the MT-200. I use these in Yamaha M-80/85 amplifiers.
    2SC 2526 —> 2SC3858
    2SA1076 —> 2SA1494
    Correct? if so ill add that to the list"

    Are these still a valid substitute?

    Or, are the MJL3281/1302 a better choice?

    Les
     
  15. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    Update 5

    I've settled on these Sanken tranny's:
    2SA1295
    2SC3264

    to replace both left and right channels. Getting them from Digi-key.

    My basis was a minimum Hfe of 50 {OEM 2SC2526 was between 40-60 depending on Ic of 1A (60) - 7A(40)}; Ft of 60Mhz typical (OEM was 50min, 80 typ); and Vce/Vcb of 230V (OEM 120V). The rail voltages in mine are slightly higher that +/- 60V so I doubt the new ones will get stressed.
     
  16. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    This post jumps over an info gap on progress to date but you'll understand why in a sec.

    Having fitted the new transistors to the power amp (that's a post on its own) and recapping the driver board we now have parasitic oscillation (PO) present. BIG TIME. Notwithstanding -ve feedback paths for the servos there's the protection path that may be the cause. These three elements are the only boards in play at this stage.

    With the protection ccts disconnected hopefully this will cure it, otherwise we're going to chase the RF to gnd and inject an audio tone to trace it to a point where the oscillation is having an effect.

    My obvious question to the better technically adept (than I, have forgotten more than I know) here is have you come across PO in a NS Amp before, even better if it's an A-9??

    Cheers,

    Les.
     
  17. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    After some persistent efforts, and a touch of professional pride, my A-9 sings again. DC is flat as a pancake with no oscillation.

    More to follow after protection reed relay cct gets s design overhaul to provide the correct coil voltage.
     
  18. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    UPDATE 5
    Well it's now down to the Power Amp Assy (GWH-148). At the moment I'm running my fluke over the board unpowered.
    This machine once in fault condition has proven to be a difficult thing to master given the protection ccts and the DC Servo loops.
    What we have now is:
    All voltage regulators are solid and clean (i.e. no noise on the rails)
    Driver Amps are solid and clean (i.e. DC bias is flat)
    It's had at least 200 power cycles with no faults occurring.
    The Power Amp has been re-capped and I'm waiting on some higher voltage caps to come in to finish off the outputs.
    Function switches and pots work, with some requiring a DeOxit spray once the machine is fault free.

    In another thread I asked for anyone with a full, original service manual. Seems like I had one all along, and upon in-packing from a recent move, I found it amongst the owner manuals the previous owner had given me. This has been a god send, that and the Service Guide "Troubleshooting Power Amplifiers" a Tuning Fork Supplement that shows fault finding processes albeit based on the A-7.

    IMG_4879.JPG IMG_4880.JPG IMG_4881.JPG
     
    Last edited: Mar 31, 2017
  19. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    To add to yesterdays update, the Power Amp part of this board has been decoupled from the Protection Cct, so that when it's powered back on the protection cct fault won't drag the board down.
    RL3 had been shorted out by the previous owner or his/her technician and a resistor put across the coil.
    Why I cannot fathom, must have had a reason but inability to fault find is possibly one of them.

    This machine has been quite a so and so to get to this state and I'm very grateful to my colleague for coming to grips with how the circuitry works, how it feeds back on itself and how that protection weaves its way into everything.

    Might start a poll on what track/album I should play to debut the A-9.
     
  20. Tangerine

    Tangerine Thanks for all the fish!

    Messages:
    97
    Location:
    Torrens, ACT, AUSTRALIA
    I finished off the previous post without saying that the Protection Ccts are also in fault, and have been decoupled from the Power Amp.
    RL3 has been shorted out and a resistor placed across the coil, done by the previous owner or their tech.
    Why is beyond me but since we've gone through a lot to get the A-9 this far I suspect that it was beyond them and easier to put that quick fix in place.
     

Share This Page