Measuring gearbox end-play

Discussion in 'Airheads' started by DiabloADV, Jun 3, 2012.

  1. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    I spent 3 hours Saturday trying to get the shim measurements right after an overhaul of my 5sp gearbox. We had the right plates and a handful of depth measuring gadgets. I (and about everyone else there) kept getting different measurements. The box would get closed up, and be too tight. Then, do it again and the output shaft shim (which is visible before the seal goes on) would be loose. One guy had been at it for 4 hours by that time. I finally gave up and had to get home. I think using those indicators takes more skill than most of us at the tech day possessed.

    Today, I tried a method I'd read about. Place little pieces of hollow-core solder so the ends sit across the outer races. Heat the cover and install and torque. Then take the cover back off. Measure the squashed solder, subtract 0.05mm, and you have your shim specs.

    [​IMG]

    It worked perfect, and took 15 minutes instead of the tedious afternoon of ultimate failure. Plus you don't need a $70 flat plate and a $100 indicator. Or much skill.
    #1
  2. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    Also...here's why you don't use the kick starter on a '74 90/6...

    [​IMG]

    Those gear nubs are almost completely worn through.
    #2
  3. Rucksta

    Rucksta SS Blowhard

    Joined:
    Oct 26, 2008
    Oddometer:
    2,724
    Location:
    Gold Coast
    Thank you.
    It seems to make perfect sense.
    #3
  4. hardwaregrrl

    hardwaregrrl ignore list

    Joined:
    Jan 26, 2006
    Oddometer:
    8,141
    Location:
    Atlanta
    Wow!! Thats nuts!!! Thanks for the pics!


    #4
  5. ozmoses

    ozmoses Ride On

    Joined:
    Jul 3, 2009
    Oddometer:
    8,159
    Location:
    USA
    Being mathematically challenged, my first reaction when numbers are involved is to panic :eek1.


    Is the .05mm a tolerance or...?
    #5
  6. Wirespokes

    Wirespokes Beemerholics Anonymous

    Joined:
    Jul 26, 2002
    Oddometer:
    8,691
    Location:
    Jackson's Bottom Oregon
    Correct!

    The thickness of the solder is the exact distance between the bearings and case. The .05mm (.002") is the clearance needed.
    #6
  7. Airhead Wrangler

    Airhead Wrangler Long timer

    Joined:
    Mar 4, 2008
    Oddometer:
    6,821
    Location:
    Back in Seattle, FINALLY
    Hmmmm. That's pretty slick. I've never rebuilt a trans before, at least not on an airhead, but now I kind of want to just try that trick. Cool. :clap
    #7
  8. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    Yup. That's what gap is left over so it's not tight and oil can circulate.
    #8
  9. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    And...I bet this would work on a crankshaft. Place the solder, torque up the flywheel (with the old bolts). Remove flywheel, measure solder, install proper thrust washer/shim and install flywheel with new bolts.
    #9
  10. supershaft

    supershaft because I can

    Joined:
    Sep 9, 2010
    Oddometer:
    9,116
    Location:
    San Francisco Bay area
    I advise people to do that with pistons and valves all the time and I had never thought about doing it in the tranny. I work on them semi-regularly. I might try that next time. I always look up the specs but I think they go to zero. For my own nit picky self, I think they shift better shimmed to the bare minimum. IMO, with all the right stuff and a lot of experience the measurements are still a bit of a guess for measuring the bearing height. I measure them from the inside race but you are never quite sure if the gauge isn't rocking.

    Flywheels? I will probably stick with a dial indicator but you could do it that way. It's real accurate and a real time saver with pistons. If everything is right you don't even have to take it back apart! I do it all through the spark plug hole. I think it's a much more solid measurement than clay.
    #10
  11. Bill Harris

    Bill Harris Confirmed Curmudgeon

    Joined:
    Sep 20, 2008
    Oddometer:
    9,738
    Location:
    backwoods Alabama
    So this will give you a clearance of .002" at 'room temperature'. Has anyone ever reckoned that clearance at operating temperature? That is, at the temperature of the gearbox while running, the steel gearshaft will heat up and expand in length, the aluminum case will will heat up and expand in length, and the delta-length will be more fo the aluminum than for the steel (different "coefficient of expansion" of the metals, about 3x more for the aluminum than the steel). For example, the .002" cold clearance may be ~.009" hot.

    Just curious. Been meaning to do the same figurin' for the crankshaft endplay. But if some (other) techno-geek has beat me to it, I'm no glory-hog... :rofl
    #11
  12. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    Yes...and I know exactly who....

    The engineers at BMW that decided that 0.05mm was the right number. :wink: They get paid to think about such things.
    #12
  13. 100RT

    100RT Long timer

    Joined:
    Nov 4, 2009
    Oddometer:
    1,190
    Location:
    N. E. OHIO

    Are they the same guys that eliminated the output shaft circlip and then put it back on years later?
    #13
  14. Wirespokes

    Wirespokes Beemerholics Anonymous

    Joined:
    Jul 26, 2002
    Oddometer:
    8,691
    Location:
    Jackson's Bottom Oregon
    :rofl:rofl:rofl:rofl

    Who knows? The guys who designed the transmission may have been long gone by then!
    #14
  15. Bill Harris

    Bill Harris Confirmed Curmudgeon

    Joined:
    Sep 20, 2008
    Oddometer:
    9,738
    Location:
    backwoods Alabama
    Nah, I suspect that the old guys who designed the tranny retired and came back to the office for a visit, found out what the whippersnappers had done with the circlip and screamed "vat vur you tinkink???" and got it put back in.
    #15
  16. DiabloADV

    DiabloADV Semi-Occasional

    Joined:
    Oct 17, 2010
    Oddometer:
    2,405
    Location:
    Nor Cal, USA
    While there is a long trail of tears around no-circlip problems, I've yet to come across an archive of owners complaining that the 0.05mm axial play clearance of the gearbox shafts is causing people's RTW trips to come to a halt. Perhaps maybe they got that one right.

    Besides, given the shim sizes they made available (when they were all available), the 0.05mm figure obviously has some play in it as well. A little more; a little less. Probably OK. I'm shooting for a fit in the 0.05mm to 0.10mm range since shims are not available in an infinity of sizes.
    #16
  17. supershaft

    supershaft because I can

    Joined:
    Sep 9, 2010
    Oddometer:
    9,116
    Location:
    San Francisco Bay area
    Where did they say that? Everywhere I have looked they said 0 - .1mm. You can usually get it a lot closer to zero if you stack shims. That is if you want to.
    #17
  18. bmwrench

    bmwrench Long timer

    Joined:
    Nov 10, 2006
    Oddometer:
    3,444
    In a Service Bulletin about 1985. I suspect it was needed with the 17.5 degree gears.
    #18
  19. supershaft

    supershaft because I can

    Joined:
    Sep 9, 2010
    Oddometer:
    9,116
    Location:
    San Francisco Bay area
    I don't remember that one. Never seen it mentioned in other manuals. I wonder what their reasoning was? I know from experience that less works just fine on the later PFA gearsets. I wish I could read that one. Somebody around here knows how to look them up. I believe trannys are in group 23?
    #19
  20. Olliew72

    Olliew72 Adventurer

    Joined:
    Mar 30, 2011
    Oddometer:
    81
    Location:
    Inner Eastside of Portland, OR
    Well I'd just like to say that I tried this solder trick for measuring the end play and it worked perfectly. I just wish that I was smart enough to have thought of it myself. I'm glad I stumbled on this thread. Thanks.
    #20