1. Time for some upgrades in server hardware and software to enhance security and take AK to the next level. Please contribute what you can to sales@audiokarma.org at PayPal.com - Thanks from the AK Team
    Dismiss Notice

McIntosh MC2100 - replace small signal transistors?

Discussion in 'McIntosh Audio' started by eedork, Jan 9, 2018.

  1. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    Hey guys -

    I've got a clean MC2100 that hadn't been turned on in over a year. This amp has had all of the electrolytic caps replaced recently, aside from the multi-cap and the main filter caps.

    This past weekend I decided to fire it up and rotate it into the system. Unfortunately after a few minutes of good clean sound I heard some faint crackling and lost the left channel. I figured there was a bad transistor on the input board or driver board, and after some troubleshooting thought I had traced it down to one of Q1/Q3 (132-056).

    Reading many repair/restoration threads on the 2100/2105 led me to an acceptable modern replacement for 132-056, KSA992, which I have plenty of on-hand. I hfe matched a pair, popped them in (note - pinout is different), and fired it up with the same issue - sound for maybe five minutes and then faint crackling and the channel drops out.

    After more testing and troubleshooting, I now believe the REAL issue was either a dirty stereo/mono switch or input potentiometer. I've applied DeOxit D5 to the stereo/mono switch and F5 to the pots and worked all controls many times. The amp now plays perfectly and sounds great (I've got two hours of run time on it). It also runs cool as a cucumber.

    I have not popped the original 132-056 transistors back into the left channel driver board, and started wondering if I even should. Transistor technology has surely improved significantly since the 70s when this amp was produced. Is it worth replacing the small signal transistors with modern replacements even if the originals are still working?

    I've also seen many threads where people recommend automatically replacing the differential pair on the input board (132-092, can be replaced with KSC1845 I believe). Some also recommend replacing all of the outputs, even if working, with new MJ15003.

    What's the current feeling on this? Leave it alone and wait for it to fail, or update some of the silicon as a preventative maintenance measure?

    -Matt
     
    Last edited: Jan 9, 2018

     

    Please register to disable this ad.

  2. c_dk

    c_dk Addicted Member

    Messages:
    6,182
    Location:
    West Michigan
    What does the distortion wave form look like......

    I don't know that transistors got so much better, but they certainly got a lot more consistent with tighter tolerances.

    When the Toshiba sc2240bl and sa970bl came out DOB made a comment to me that Mac's continuing weeding through batchs had become redundant they matched so closely. That was when the paint designations stopped.

    The recent 2240 and 970s I have purchased for much longer money no longer match well with the hoarded ones I have had for 25 years. They barely meet the BL spec.
     
    Last edited: Jan 9, 2018
  3. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    Hi c_dk,

    Thanks for the response! I have read many of your posts and appreciate your knowledge and insight. Unfortunately I do not have a distortion analyzer so I can't really answer your question.

    If everything looked and measured fine (distortion wise) would you just leave it alone or would you still replace silicon? Should the differential input pairs on the input board be replaced regardless (many recommend doing this even if the amp works)?

    -Matt
     
  4. c_dk

    c_dk Addicted Member

    Messages:
    6,182
    Location:
    West Michigan
    Without a proper bench test all is speculation......have the driver transistors been updated from the original RCAs with the attached square heatsinks or are they the replacements with the radial top hat heatsinks?

    I have been noticing that the bias resistors in the 250, 2505, 2100, and 2105 have been drifting now that they are over 50 years old and even 2 or 3 ohms will drive the distortion up.
     
  5. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    The driver transistors have not been updated - they are still the original RCAs with the integrated square heatsinks. I have the driver board service bulletin and have read that Mc parts sells update kits. This may be something I investigate at some point, although as you've said, without the proper test gear I won't really know if its needed or if has made anything better.

    I will at least test the bias resistors and make sure that they are within spec. I suppose I should check all of the resistors since many are old carbon comp that I have had problems with in the past on other projects.

    While this unit is very clean and the boards show no scorch marks, one thing I have noticed is that the power transformer hums ever so slightly when plugged directly into the wall (~123V). If I knock the voltage down a little to ~115V using a varian the hum goes away completely. I may end up building a bucking transformer so that I don't have to run it off my variac all the time.

    -Matt
     
  6. c_dk

    c_dk Addicted Member

    Messages:
    6,182
    Location:
    West Michigan
    You will spend a lot of wasted time lifting all the resistors to find them off less than 10%......the bias resistors however, being off by 2 ohms will allow the distortion to rise out of spec. I just keep some 2% 18s or 22s on hand and replace them because by the time you unsolder to test you might as well replace.
     

     

    Please register to disable this ad.

  7. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    Thanks again c_dk.

    Which resistors are the bias resistors specifically? I assume it is R15 and R16?

    -Matt
     
  8. c_dk

    c_dk Addicted Member

    Messages:
    6,182
    Location:
    West Michigan
    R17 and R18 should be 18 ohms with the original driver transistors......22 if they have been replaced with the mod kit parts.
     
  9. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    Thanks c_dk!

    Out of curiosity, have you ever used Hayseed Hamfest to reproduce a new multicap for a 2100/2105?

    -Matt
     
  10. c_dk

    c_dk Addicted Member

    Messages:
    6,182
    Location:
    West Michigan
    Jeff79 caught me in a lack of clarity.....depending on your serial number the bias transistors for a MC2100 could be 18 or 22 ohms.......in units below 99W26 they are 22 ohms, above 18 ohms.

    The schematic in the AKdatabase shows 18 on the schematic.....

    You must always remember to read all the service notes anytime you work on one of these units with long production runs.
     
  11. eedork

    eedork AK Subscriber Subscriber

    Messages:
    1,529
    Location:
    Southern Maine
    Hi c_dk,

    I was *just* going to write to you about this because I noticed the same discrepancy a few minutes ago.

    The schematic in the service manual available here shows 18, you said 18, but my unit has 22 ohm bias resistors installed. My serial number is partly rubbed off, but I believe it is 56W** (I can't read the last two digits) so 22 ohms for R17 and R18 is indeed correct according to schematic note #9 in the service manual. So all is good.

    The bias resistors, as well as a few others, have already been replaced with new 1/2 W parts. So my board should be in good shape even though they don't have the mod kit updates. I've attached a picture of one of my boards (with the new KSA992 transistors installed) just for reference.

    I'm going to order a new multi cap from Hayseed. Seems like a nice, clean solution.

    -Matt
     

    Attached Files:

     

    Please register to disable this ad.

Share This Page