SA-7800: another one bites the dust.

Discussion in 'Pioneer Audio' started by USMC Spike, Jan 8, 2017.

  1. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    MTF - No worries, a double stack .45 and a 92 ain't bad.

    -- Spike
     
  2. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
    MT-200 will fit, takes some creative lead bending. I've seen 2SC2921/A1215 used. I was a little concerned about the saturation voltage vs. C2525/A1075, but they seem to work just fine. On that note, 2SC3519/A1386A (TO-3P) should also work, but you will need to drill a couple of holes in the heatsink.
     
  3. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Thanks tarior, 2sc3519/3519A looks to be pretty good. Data sheet very similar to the original
    part part. I don't know what was I thinking, It's not like I'm going to ruin a pristine amp drilling
    into the face plate, its a heat sink. Has any one else used this in like NSA amps?

    Folks have mounted at 90 degrees? or is that a different part number?
    So now that I'm looking at data sheets, I see the MJL21193/94, are
    similar. But their Saturation data is difference and I'm not sure
    how to compare them, or am I missing something here?

    Thanks,

    Spike
     
  4. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Thanks tarior, 2sc3519/3519A...have data sheet, pretty robust. Actually what was I thinking?
    It's not like I'm going to ruin a pristine amp drilling into the heat sink of this.
    Yes better specs than the 2SC921/A1215. But how does it compare to the
    Onsemi ones that folks are using. The MJL1302/3281s?

    If I knew I wouldn't ask.

    --Spike
     
  5. zebulon1

    zebulon1 Into Phase Linear's lately Subscriber

    Messages:
    3,660
    Location:
    Las Vegas Nevada
    The amp is looking good.
    Love the stories as well.
     
  6. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
    I think, all else being roughly equal, I would go with a TO-3P rather than a TO-247 package. You will have a much easier time fitting it in on the transformer side of the heatsink. Vce (sat) is two volts on 3519 and 2921 vs 0.7 on the originals. I have a 7800 here with 2921 on one channel, and judging by the dust, it has been there for a very long time. So, that number probably isn't gonna mean anything in this application.
    IMO, if you don't feel like shoehorning a MT-200 device in there, 2SC3519/A1386 look pretty damn good. In stock at Dig-Key. Don't forget to buy micas.
    Mark (thefixer) may have a good modern suggestion for drivers. I still have a stock of A1011/C2344 that I use, haven't figured out something for when those run out, but I'm sure there are some good choices out there.
     
  7. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Hi Tarior, You are pointing out the things I am noticing or at least has been
    in the back of my mind...simmering. I've read about some folks with problems
    getting the big Sankens in these...I also was concerned about bending the crap
    outta them, etc, etc, etc.

    Do you have any TO-3P devices that you can just
    place in yours to see if they will fit, centered, collector
    leg to collector trace? Is there head room to mount the
    device without machining it to hell and back?

    Any unknown fitment issues I should be aware of?

    Can you take a pic so that I can see too,
    Please?

    How do the A1011/C2344 work for the drivers?
    How are the distortion figures for
    2SC3519/3519A//2SA1386/A : A1011/C2344 combo?

    Where might a guy like me find a couple of matched sets
    of A1011/C2344 pairs for a project like this?

    What might come in handy is then we can take some measurements
    and see how this really perform. If y'all are up for it.

    Somewhere around here MTF has some recommened TO-126 pair?

    Trying to think with modern replacements for really close matched driver
    pair PnP, NpNs...10 each? If I can get them on a reel?

    Is that about the same count about 10 each for two pair
    of the actual Outputs as well?

    Anyone interested in stocking up on them?

    In closing for the moment,
    nothing quite like...okay got everything...and that
    gal comes whispering into you ear....sweet nothings.
    got mica?
    got mica?
    got micas big boy?

    -- Spike
     
    Last edited: Jan 11, 2017
  8. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Zebulon1,
    Thanks it's all good. It's so distracting seeing that scantily covered gal
    by that van. Where did they find all that grass in Vegas? OH wait,
    maybe it was green screened.

    -- Spike
     
  9. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
    I did check on the 7800 that I have here, and the TO-3P will fit just fine and look right. Just need to install some holes for 3mm (pretty sure, check it) self tapping screws.. A1011/2344 are just fine, distortion will meet spec I'm certain. Both are showing in stock at Consolidated. http://www.ceitron.com/
    Steve has a $25 minumum order, which is kind of a PITA, but he's good to deal with and I've never heard of anyone ever receiving counterfeit parts from him.
     
  10. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    tarior, thanks for the link as I think that will be really helpful.

    -- Spike
     
  11. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    In the mean time getting ready for transistor order, w/mica, heatsinks where needed.
    then double checking the lead spacing for caps. Compiling and verifying list from other
    threads here.


    Do we have any RF folks here?
    Questions:
    1. It's been my understanding that we have to be careful about using stacked
    capacitors around transistors because the stacking acts as antenna and
    picks up HF signals, hash, etc.

    2. Metalized Films are fine except for their use in crossovers.
    Yes I know many people use them, the problem is the the
    spray on metal moves, vibrates oscillates, etc causing distortions.

    3. If we can get them to fit...film and foils are great.

    4. Then there are always the Paper in oils (PIO).

    I won't even get into resistor types.

    In general, tube stages perform with film coupling caps and
    SS perform with 'lytic coupling caps.

    Preaching to the choir here from what I know:
    A. We want tighly matched diff pair, is most critical in the amp!
    along with tight thermal coupling.

    B. High quality matched temp co resistors here and in
    the feedback path.

    Let's not forget matching the output transistors either,
    if we read the applicable data sheets a 10x reduction in
    distortion with tightly matched outputs.

    So I"m going to be paying very close attention to these areas.

    Another question then for my transistor educated forumites...
    When we look that the gain structure of the circuits we are using,
    Where or how do we need to put the GainBandwidth Product used
    between the amp stages and what is optimum?

    I appreciate discussion along the way if y'all don't mind.

    Thanks,

    -- Spike
     
  12. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
    1. Stacked films are fine. Not many places to use them in this amp any way.
    Do not even think about re-capping or anything else until the original fault is repaired.
    2. see answer #1.
    3.There is pretty much no place in this amp where a film and foil will be useful. There are polystyrene caps on the phono board. Leave them be.
    4. Same as #3.
    Don't replace any resistors unless they are damaged.
    This unit uses monolithic diff inputs. Leave them be unless they are screwy.
    Matching the outputs is a total waste of time. The difference between .001% and .01% distortion is academic at best.
    Gain bandwidth product is a transistor specific parameter. The outputs recommended previously are a perfect match. Gain is a circuit specification, and it will not be changed whatsoever during the course of repairs.
     
  13. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    taroir,
    got it. will keep focused on
    1. Fix first, then stabilize.
    2. Then upgrade one side at a time.
    3. Maybe three or four places to use film caps.
    in those spots then the stacked are okay.
    I take it you aren't too hot on doing too much
    to these amps?

    -- Spike
     
  14. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
    Outside of taking care of recurring issues, I'm not hot on modding any amplifier. The engineers at Pioneer (or Sansui, Marantz, Kenwood, and so on and so forth....) knew what they were doing.
     
  15. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    With that in mind tarior, it kind of makes the servicing of amplifiers
    pre amplifiers and the like more focused and reliable.
     
  16. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
  17. tarior

    tarior Dirty pool, old man?

    Messages:
    5,249
    Location:
    Kelso, Washington
  18. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Tarior, Nicely done I'm sure anyone reading it will benefit from it.

    --Spike
     
  19. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Almost...there, it took me a bit to find the RM-66 transistors....

    ....whenever they get here I'll finished up the testing and intstall and
    trouble shoot them.

    Then, and only then (once verified working properly) do the caps
    AND some basic upgrades.

    -- Spike
     
  20. USMC Spike

    USMC Spike Active Member

    Messages:
    134
    Why RM-66 and not TM-200 or TO-3P or TO-247?

    I don't have enough head room to mount on the heatsink.
    I'll take a pic next time the 3800 is on the bench.

    Currently I'm fixing a Keithley 2001 Meter.
    Can we say thank you UPS.

    There is a reason for the brown uniforms...
    ...it has to do with the state of the gear you receive!

    OR

    If it ain't broke, it will be.

    -- Spike
     

Share This Page