Announcement

Collapse
No announcement yet.

DC offset on Gallien Krueger RB1000

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

  • DC offset on Gallien Krueger RB1000

    I recently bought a RB1000. The seller said that if ran hard will go into “protect”. I received it and put it on the bench and did a few tests. With no load the amp appears to be putting out a large healthy signal. I had a signal generator on the input and scope on output. The first sign of trouble was I found -2.5v DC on output. I am assuming this is playing a role in it faulting out. Visual inspection shows that r520 has been replaced and possibly Q513. I quickly (in circuit) ran through resistors and diodes, also tested all output devices for shorts. No problems found. Schematic shows rails at 40v and 80v pos and negative. They are actually measuring closer to 45 and 95v. I was checking some basic voltages at some of the transistors and found what to my novice eye seems incorrect (I wrote some notes in pencil in one of the pic I included, I also added the complete schematic for power amp section). Q534 base voltage is at 1.2v, emitter at 1.7v. Q539 base is at -8.25v with emitter at -7.68. Looking at the circuit(again to my novice eye) I would expect them to mirror each other? But they don’t. The 1 ohm collector resistor on each of them shows a .002v voltage drop. Suggesting that both transistors are flowing 2ma. I backed up to Q535 and Q538 and checked voltages. I made some notes on those too. Those seem to to mirror each other. Zener diodes d519 and d530 have 5v dropping across them and Q514 supporting resistors check ok. I did not test Q514 yet because I would need to pull it. Now that I’m written this I guess I could have checked the voltage drop from E to C. D522 and D527 also are showing voltage drops I would expect to see (can’t remember volts at this point). Checked voltage across bias adj circuit and appears to be changing with pot adjustment.
    With all this info I ask.....am I on to something or am I barking up the wrong tree?

    When reading about typical offset issues people often talk about the differential amp not being properly balanced. This amp is not using the typical configuration I am more familiar with. Are Q484 and Q485 (2 in 1 transistor packages) acting like the Diffy pair? I can see neg feedback coming in.

    Most of my back ground is with tube gear, but I am totally interested in any wisdom any of you might have share. This is a hobby for me but am always excited to expand my understanding of these solid state circuits. Thank you very much in advance.

    Fred
    Attached Files

  • #2
    If you write in the numbers on the other sides of the zeners (D519, D530) based on the 5V drops you stated, it doesn't make sense.
    Can you double check, and post voltages for Q514 bias transistor?
    (and yes, Q534 and Q539 should mirror each other as you suspected)
    Originally posted by Enzo
    I have a sign in my shop that says, "Never think up reasons not to check something."


    Comment


    • #3
      Thank you, I see your point, unfortunately I won’t be be able for a few days (out of town). Dang now I wish I could!

      Comment


      • #4
        Sorry about how long it took to respond. I gat back to the amp and began making some measurements when my test lead slipped and shorted out a hand full of components in the bias circuit including transistors, zeners and a couple resistors. I ordered parts and patched it back together.
        I then resumed my inspection, I ended up finding Q486 totally open. I also found its counterpart Q488 seemed a little odd when testing with DVOM in diode test mode. The junctions seemed to work as normal but the voltage drop was .8v instead of the typical.6 or so. The other MPSA06 and MPSA56 transistors did not test that way, nor did my replacement parts.
        This seems like an unusual failure mode for transistors. Does that suggest anything as far as what may have caused the failure?
        the amp now test good and is passing all performance tests.
        Thanks for your help
        Fred

        Comment

        Working...
        X