Results 1 to 15 of 15
  1. #1
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897

    Torque for certain bolts? And accuracy of torque wrenches?

    What should a rack bolt (the ones that go into the eyelets of a frame) be torqued to?

    What should a seatpost clamp bolt be torqued to?

    I've never found instructions on those bolts torque ratings.

    Also, how accurate are the larger torque wrenches at lower torque ratings?

  2. #2
    mtbr member
    Reputation:
    Join Date
    Apr 2012
    Posts
    1,625
    1) The manufacturer may give you a value, but 'tight enough not to move under vibration' (aka, not terribly tight...less than 5nm?) is just about right. It depends heavily on whether the hole is tapped, or is a rivnut set in the frame.

    2) Depends on the seatpost/frame/clamp material. Typically 5-8nm. You can go lower if you use friction paste instead of grease when installing the seatpost. The correct value is 'tight enough that the seatpost doesn't slip when you hit a bump seated'

    3) Not very. You typically see an error range of 5-10% with torque wrenches from the printed value. The larger the scale, the larger the discrepancy will be. Pick up a small torque wrench (a beam type by park, intended for bicycles, is good). Better yet, get a torque key...these are small, $20-30 one value click-type torque wrenches. Some have interchangeable bits, but regardless, usually have the appropriate bit for the range. They can be found, usually, in 4, 5, and 6nm ranges, with either 4 or 5mm hex keys. I know for a fact that the hex bit in the Ritchey version can be pried out, and it will take a standard 1/4" bit.

    Torque values are almost always printed on components (seatpost collars, stems, cranks, etc).

  3. #3
    mtbr member
    Reputation: One Pivot's Avatar
    Join Date
    Nov 2009
    Posts
    4,123
    Its generally accepted that torque wrenches are "useless" in the lower 20% of their range. You should only use them from 20-100%.. so if you have a 0-100 in lb wrench, you can really only use it from 20-100 in lb.

    It gets much worse when using a larger foot pound torque wrench on things rated in inch pounds, and converting. 60 in lb is 5 ft lb, but if you have a torque wrench that goes 5 to 100 ft lb, you're going to bust fasteners trying to torque them to 60 in lb. This is a major reason why people bust components trying to torque them properly.

    My worst torque wrench was less than 1% inaccurate, in some ranges it was around 0.5%. The good stuff is really pretty accurate! Even the cheaper stuff is pretty good, as long as its not too cheap and you're using it properly.

    For bikes, I dont torque anything at all and Ive never had a problem at all. The only objective for a bike is to not over stress the fastener, and have it tight enough to not slip. Theres no heat and warping from inaccurate torque like some automotive/industrial applications.

    A rack bolt, id use a short hex wrench and go "tight enough". Long tools make torque drastically harder to feel, but doing it reasonable with a standard L hex wrench you'll be fine.

  4. #4
    mtbr member
    Reputation: cockroach's Avatar
    Join Date
    Jun 2009
    Posts
    161
    Keep it simple, there's a reason allen keys / hex wrenches are short for small bolts & long for bigger bolts. Small bolts need a lot less torque, just use the standard allen key & don't overtighten.

  5. #5
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897
    Quote Originally Posted by One Pivot View Post
    Its generally accepted that torque wrenches are "useless" in the lower 20% of their range. You should only use them from 20-100%.. so if you have a 0-100 in lb wrench, you can really only use it from 20-100 in lb.

    It gets much worse when using a larger foot pound torque wrench on things rated in inch pounds, and converting. 60 in lb is 5 ft lb, but if you have a torque wrench that goes 5 to 100 ft lb, you're going to bust fasteners trying to torque them to 60 in lb. This is a major reason why people bust components trying to torque them properly.

    My worst torque wrench was less than 1% inaccurate, in some ranges it was around 0.5%. The good stuff is really pretty accurate! Even the cheaper stuff is pretty good, as long as its not too cheap and you're using it properly.

    For bikes, I dont torque anything at all and Ive never had a problem at all. The only objective for a bike is to not over stress the fastener, and have it tight enough to not slip. Theres no heat and warping from inaccurate torque like some automotive/industrial applications.

    A rack bolt, id use a short hex wrench and go "tight enough". Long tools make torque drastically harder to feel, but doing it reasonable with a standard L hex wrench you'll be fine.
    I was hoping I'd be under-torquing things. I have a US made craftsman one that goes to 95 newton meters but tend to use it for things that are about 5-14 newton meters.

  6. #6
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897
    Okay, now I'm really confounded. Park offers one wrench that does 0-7 Nm which makes it basically useless for everything but tightening cable pincher bolts (even my derailleur hanger bolt is 8-10 Nm) and the other is 0-70 Nm (which is still smaller than mine by about 25 Nm).

    If they're that inaccurate then why the gap in torque wrench sizes?

  7. #7
    mtbr member
    Reputation: breckenridge's Avatar
    Join Date
    Jul 2012
    Posts
    216
    The easy answer is the well-calibrated mechanic's hand on the appropriate length allen wrench

    But no, I'd say an inch-lb torque wrench should do you fine for most medium-tight applications (brake calipers to frame, stem pinch bolts, etc. 5-10% error is only a few in-lbs for most of these, so you should be fine. Specific torque values are a guideline and I wouldn't worry about a few in-lb in variance. Just don't be double or half of what it should be. Just don't ham-fist it and you'll be fine.

  8. #8
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897
    Are the percentages of the total range or of the amount of torque being applied? Mines supposed to be accurate to 4%.

  9. #9
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897
    Quote Originally Posted by aBicycle View Post
    Okay, now I'm really confounded. Park offers one wrench that does 0-7 Nm which makes it basically useless for everything but tightening cable pincher bolts (even my derailleur hanger bolt is 8-10 Nm) and the other is 0-70 Nm (which is still smaller than mine by about 25 Nm).

    If they're that inaccurate then why the gap in torque wrench sizes?
    On this note, is it just me or is the park tw-2 the exact same wrench as the craftsman 3/8" beam wrench, but with a different set of measurement plates on it? How could it be any more accurate within its reduced range than the craftsman given that they both use the same beam (the craftsman goes up to 100 Nm now that I look at it, but the scale covers much more of the width of the face plate to accomplish this)?

    And why didn't they make a wrench for the 1-20 Nm range? Technically anything below 20 Nm is in the bottom 20% of the park tw-2 range as well as that tw-2 is the same 100 Nm max wrench that the craftsman is. So how can it be accurate down to the above 7 Nm and to 20 Nm range that the tw-1 doesn't cover?

    And is there any way to get a tw-1 for cheap? It's pretty obviously a TD tools 1/4" beam wrench, but that thing costs just as much as the park does.


    Also, why do torque wrenches have a bottom 20% of their range if it is useless?
    Last edited by aBicycle; 01-20-2013 at 06:24 PM.

  10. #10
    mtbr member
    Reputation: One Pivot's Avatar
    Join Date
    Nov 2009
    Posts
    4,123
    Oops, I should have stated that the 20% thing is only for clicker torque wrenches. A bending beam type is accurate in its full range. I think dial torque wrenches are too, but no ones going to use those on a bike ($$$$).

  11. #11
    mtbr member
    Reputation:
    Join Date
    Jun 2012
    Posts
    897
    Quote Originally Posted by One Pivot View Post
    Oops, I should have stated that the 20% thing is only for clicker torque wrenches. A bending beam type is accurate in its full range. I think dial torque wrenches are too, but no ones going to use those on a bike ($$$$).
    Oh thank god.

    What happens if you do over-tighten something a bit, whilst not paying attention as well as you should, say a 6 Nm max to around 10 Nm probably, but then back off really quickly? This would be for an aluminum clamp part.

  12. #12
    mtbr member
    Reputation: splatworldwide's Avatar
    Join Date
    Aug 2005
    Posts
    198
    For anything without a spec, and especially for smaller fasteners, I always stick to the mantra, "Monkey tight, not gorilla tight."
    Loud hubs save lives. http://about.me/splat/

  13. #13
    mtbr member
    Reputation:
    Join Date
    Apr 2012
    Posts
    1,625
    Quote Originally Posted by aBicycle View Post
    Oh thank god.

    What happens if you do over-tighten something a bit, whilst not paying attention as well as you should, say a 6 Nm max to around 10 Nm probably, but then back off really quickly? This would be for an aluminum clamp part.
    If the threads don't strip, or the part doesn't collapse? Nothing happens.

  14. #14
    Plays with tools
    Reputation: customfab's Avatar
    Join Date
    Jun 2008
    Posts
    4,245
    Quote Originally Posted by One Pivot View Post
    Oops, I should have stated that the 20% thing is only for clicker torque wrenches. A bending beam type is accurate in its full range. I think dial torque wrenches are too, but no ones going to use those on a bike ($$$$).
    What wrenches have you seen that have a margin of error as wide as 20%? The vast majority of torque wrenches I've looked at are more like 2-4% in the upper 80% of their range.

  15. #15
    mtbr member
    Reputation: squareback's Avatar
    Join Date
    Sep 2011
    Posts
    1,054
    Quote Originally Posted by customfab View Post
    What wrenches have you seen that have a margin of error as wide as 20%? The vast majority of torque wrenches I've looked at are more like 2-4% in the upper 80% of their range.
    This is what he said:

    Quote Originally Posted by One Pivot
    Its generally accepted that torque wrenches are "useless" in the lower 20% of their range. You should only use them from 20-100%.. so if you have a 0-100 in lb wrench, you can really only use it from 20-100 in lb.
    .... and I would say even greater than 20%. ( I bought a calibration tool)



    I blame broken bolts on a generation of parents who threw the bike in the minivan to bring to LBS everytime there was a flat, instead of teaching a kid how to R and R a wheel with a crescent wrench, and no torque wrench within 5 miles.

    To the original poster: I think the "snug with a short arm allen key" is great advice.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •