Top cap torque - how much
Page 1 of 2 12 LastLast
Results 1 to 25 of 31
  1. #1
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032

    Top cap torque - how much

    Hi yall,

    I know this has been covered elsewhere, but Im just looking for a general idiot proof rule on top cap adjustment for a headset. The guildeline is tighten until no play. But im thinking if you do that, you then have virtually zero preload on the bearings. And Im also thinking for a bearing assembly that takes the shock of the road with full force, could be an issue.

    MY general rule is tighten until no play, then go an additional 1/2 turn on the nut. Is that right? Or too much? Maybe just go an additional 1/4 turn?

    Whats everyone's general rule here?

    Sorry for the "101" type of question. Sometimes its the simplest of things that vex us.

  2. #2
    Old and Fixed, Moderator
    Reputation: Dave Hickey's Avatar
    Join Date
    Jan 2002
    Posts
    20,473
    I tighten until the headset starts to bind up and than back off a 1/4 turn
    Dave Hickey/ Fort Worth

    My 3Rensho Blog: http://vintage3rensholove.blogspot.com/

  3. #3
    Rub it............
    Reputation: frdfandc's Avatar
    Join Date
    Nov 2007
    Posts
    3,827
    Quote Originally Posted by Dave Hickey
    I tighten until the headset starts to bind up and than back off a 1/4 turn

    This.


    You want tight, but not too tight. Depending on the headset/frame/fork combo, its usually between 1/4 to 1/2 turn off binding. It's a feel thing.

    Also, if using the integrated headsets that usually come equipped on the carbonious explodus bikes, they sometimes will seat further into the headtube - seen it once or 10 times - and a re-adjustment will need to be done, since sometimes the bearings don't bottom out at first. Its not a normal thing, but it does happen.
    You can't fix stupid.

    Quote Originally Posted by JoeDaddio

    I kind of wish it were legal to staple people in the face.

  4. #4
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032
    Thanks....I guess my chief concern is that by overtourquing the nut, Ill wreck the bearings, races, frame, etc.

    What Im hearing you say is that you really need to wrench that nut down before youre to the point of possibly causing damage. That right?

  5. #5
    wim
    wim is offline
    RoadBikeReview Member
    Reputation:
    Join Date
    Feb 2005
    Posts
    11,456
    Quote Originally Posted by RkFast
    you really need to wrench that nut down
    Well, that's a little harsh, in my view.

    What you want to do is tighten the headset ("apply bearing preload" if you like it a little more technical) until the headset just begins to bind up, then back off 1/4 turn or so. What I do is check freedom of movement by picking up the front end of the bike and tilting it slightly to one side or the other. When handlebar flop just begins to slow down ("gets draggy"), there's a little too much preload. You may have to tighten the top cap nut in small increments and check bar flop after each increment.

    What leads people astray sometimes is remembering how they adjusted an old threaded headset. With threadless headsets, just taking the play out is not tight enough. The old mechanic's rule of "you don't know the perfect setting until you've gone past it" applies here. But that doesn't mean to hose down on the top cap nut.

  6. #6
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by RkFast
    Hi yall,

    I know this has been covered elsewhere, but Im just looking for a general idiot proof rule on top cap adjustment for a headset. The guildeline is tighten until no play. But im thinking if you do that, you then have virtually zero preload on the bearings. And Im also thinking for a bearing assembly that takes the shock of the road with full force, could be an issue.

    MY general rule is tighten until no play, then go an additional 1/2 turn on the nut. Is that right? Or too much? Maybe just go an additional 1/4 turn?

    Whats everyone's general rule here?

    Sorry for the "101" type of question. Sometimes its the simplest of things that vex us.
    The top cap is an adjustment, not something that is "torqued". It is like a volume knob on the radio - don't go to '11'.

    With the stem bolts loose, adjust the top cap to the proper bearing tension, then lock that adjustment by tightening the stem bolts.

    Watch this video:
    http://www.canecreek.com/tech-headse...djustment2.flv
    Last edited by Kontact; 04-25-2011 at 05:52 AM.
    Get a better saddle: www.kontactbike.com

  7. #7
    RoadBikeReview Member
    Reputation: stevesbike's Avatar
    Join Date
    Jun 2002
    Posts
    5,602
    it also depends on the expander plug with a carbon steerer - the overtighten then loosen method works well enough for an alloy steerer with a starnut since it won't slip, but some expander plugs will start to slip if you overtighten them. Just tighten to eliminate play (without overtightening works fine).

  8. #8
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032
    Quote Originally Posted by Kontact
    The top cap is an adjustment, not something that is "torqued". It is like a volume knob on the radio - don't go to '11'.

    With the stem bolts loose, adjust the top cap to the proper bearing tension, then lock that adjustment by tightening the stem bolts.

    Watch this video:
    http://www.canecreek.com/tech-headse...djustment2.flv
    The bolded is the part Im struggling with. Whats the proper tension? Juuuuuust to the point where there is no play or should I get it to that point and go just a tad more?

    Like I said, I always adjusted to no play, then went another 1/4 turn. Good or too much?

  9. #9
    RoadBikeReview Member
    Reputation:
    Join Date
    Mar 2008
    Posts
    788
    Quote Originally Posted by RkFast
    The bolded is the part Im struggling with. Whats the proper tension? Juuuuuust to the point where there is no play or should I get it to that point and go just a tad more?

    Like I said, I always adjusted to no play, then went another 1/4 turn. Good or too much?
    In terms of torque...What I do is insert the long end of the hex wrench in, leaving the short end to tighten with...Turn until you feel a decent amount of tension, then make sure the handlebars turn freely but that there is no play..If that's the case, tighten the stem bolts and you're good to go. Using the short end of the hex wrench helps ensure you're not torquing the hell out of the headset bearings but at the same time, you're getting enough tension to remove play.If you are wanting a nm guess, I'd say like 1.5-2 nm

  10. #10
    RoadBikeReview Member
    Reputation:
    Join Date
    Feb 2004
    Posts
    5,297

    20-25 inch pounds works for me

    Quote Originally Posted by RkFast
    Hi yall,

    I know this has been covered elsewhere, but Im just looking for a general idiot proof rule on top cap adjustment for a headset. The guildeline is tighten until no play. But im thinking if you do that, you then have virtually zero preload on the bearings. And Im also thinking for a bearing assembly that takes the shock of the road with full force, could be an issue.

    MY general rule is tighten until no play, then go an additional 1/2 turn on the nut. Is that right? Or too much? Maybe just go an additional 1/4 turn?

    Whats everyone's general rule here?

    Sorry for the "101" type of question. Sometimes its the simplest of things that vex us.
    20-25 inch/lbs seems to work for me. I would ALWAYS check for tightness and play after the stem is attached and then adjust appropriately

  11. #11
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by DaveG
    20-25 inch/lbs seems to work for me. I would ALWAYS check for tightness and play after the stem is attached and then adjust appropriately
    A flat torque value makes no sense at all. The amount of torque necessary to preload one bearing/plug combination and another can very greatly. Advice like this is likely to kill the bearings since the actual preload after everything seats is pretty low.

    OP, go to no play, tighten just past until it is no longer smooth, back off 1/4 (or so) turn.

    Too loose and you'll hear a clunk while riding, too tight will destry the bearings. When in doubt, too loose.
    Get a better saddle: www.kontactbike.com

  12. #12
    wim
    wim is offline
    RoadBikeReview Member
    Reputation:
    Join Date
    Feb 2005
    Posts
    11,456
    Quote Originally Posted by RkFast
    The bolded is the part Im struggling with.
    IMO, you're making a mountain out of mole hill. Adjust, go for a brief ride and then check your work as per DaveG's suggestion. If you have problems riding the bike no-hands, the headset is too tight. If you hear a deep knocking sound every time you hit a good-size bump, it's still too loose. Correct accordingly.

  13. #13
    Cycling induced anoesis
    Reputation: PJ352's Avatar
    Join Date
    Dec 2007
    Posts
    13,003
    Quote Originally Posted by wim
    What you want to do is tighten the headset ("apply bearing preload" if you like it a little more technical) until the headset just begins to bind up, then back off 1/4 turn or so. What I do is check freedom of movement by picking up the front end of the bike and tilting it slightly to one side or the other. When handlebar flop just begins to slow down ("gets draggy"), there's a little too much preload. You may have to tighten the top cap nut in small increments and check bar flop after each increment.

    What leads people astray sometimes is remembering how they adjusted an old threaded headset. With threadless headsets, just taking the play out is not tight enough. The old mechanic's rule of "you don't know the perfect setting until you've gone past it" applies here. But that doesn't mean to hose down on the top cap nut.
    This.

  14. #14
    RoadBikeReview Member
    Reputation:
    Join Date
    Feb 2004
    Posts
    5,297

    not in agreement

    Quote Originally Posted by Kontact
    A flat torque value makes no sense at all. The amount of torque necessary to preload one bearing/plug combination and another can very greatly. Advice like this is likely to kill the bearings since the actual preload after everything seats is pretty low.

    OP, go to no play, tighten just past until it is no longer smooth, back off 1/4 (or so) turn.

    Too loose and you'll hear a clunk while riding, too tight will destry the bearings. When in doubt, too loose.
    An actual torque measurement seems a lot less arbitray than a subjective measurement like "tighten just past smooth". Try it your way and see what the torque value is; I'd be intereted to see if its all about the same. BTW, I think (but not postive) that I got that torque advice from Zinn

  15. #15
    RoadBikeReview Member
    Reputation: looigi's Avatar
    Join Date
    Nov 2010
    Posts
    7,162
    20-25 inch-lb seems a bit too much to me, considering 45 is the typical limit for the bar and stem pinch bolts. I go by feel but if I had to guess I'd say it more like 10-12 inch-lb.

    (BTW the units are inches x pounds ... not inches/pounds)

  16. #16
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by DaveG
    An actual torque measurement seems a lot less arbitray than a subjective measurement like "tighten just past smooth". Try it your way and see what the torque value is; I'd be intereted to see if its all about the same. BTW, I think (but not postive) that I got that torque advice from Zinn
    You're not getting me at all, I'm afraid.

    The actual load on the bearing is low - a few inch pounds. The torque you feel as you tighten the cap is the torque it takes to overcome the friction in the system and the weight of the fork (if it is in a workstand). Once you've compressed the stack of bearings, spacers, grease, compression cap, etc, the torque stops being necessary - now you're down to actual bearing adjustment.

    To give you an example, a new bike that has never had bearing sit in the built in cups may take more initial torque to compress down than the next time you take it apart. And a bike with an integrated headset is going to need different compression from a tradtional, and a 1 1/8" is going to require different than a 1". So one torque value isn't going to work, especially when the seating torque is higher than the bearing preload torque.

    Bearings are one of those things that are mostly adjusted by feel - doesn't matter if it is a car axle or a bike headset - that's the way the things work and are designed. Same thing with wheels - the torque needed to turn a spoke nipple is not the same as the resulting spoke tension. That's just the unfortunate way the universe works - you have to think when you're doing some of this stuff.


    I set headsets all day. I'm not making this up, and if you use a torque setting your are probably over-compressing your bearings, which will eat them up.


    Instead of working off a mis-memory of what a third party may or may not have said, how about reading a manual:
    http://www.canecreek.com/manuals/Hea...structions.pdf
    Do you see a torque value anywhere in there? Who's right - me and Cane Creek, or you?
    Get a better saddle: www.kontactbike.com

  17. #17
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032
    Quote Originally Posted by Kontact
    You're not getting me at all, I'm afraid.

    The actual load on the bearing is low - a few inch pounds. The torque you feel as you tighten the cap is the torque it takes to overcome the friction in the system and the weight of the fork (if it is in a workstand). Once you've compressed the stack of bearings, spacers, grease, compression cap, etc, the torque stops being necessary - now you're down to actual bearing adjustment.

    To give you an example, a new bike that has never had bearing sit in the built in cups may take more initial torque to compress down than the next time you take it apart. And a bike with an integrated headset is going to need different compression from a tradtional, and a 1 1/8" is going to require different than a 1". So one torque value isn't going to work, especially when the seating torque is higher than the bearing preload torque.

    Bearings are one of those things that are mostly adjusted by feel - doesn't matter if it is a car axle or a bike headset - that's the way the things work and are designed. Same thing with wheels - the torque needed to turn a spoke nipple is not the same as the resulting spoke tension. That's just the unfortunate way the universe works - you have to think when you're doing some of this stuff.


    I set headsets all day. I'm not making this up, and if you use a torque setting your are probably over-compressing your bearings, which will eat them up.


    Instead of working off a mis-memory of what a third party may or may not have said, how about reading a manual:
    http://www.canecreek.com/manuals/Hea...structions.pdf
    Do you see a torque value anywhere in there? Who's right - me and Cane Creek, or you?

    But what youre seeing on the CC site is different than what people here are saying, so now Im MORE confused!

    Park, King, Cane Creek, Zinn all say to tight till you hit the point of no play...then leave it there.

    Lots of people here are saying tighten past the point of no play...until actual binding occurs (MUCH tighter) and then back OFF 1/4 turn.

    Which is better? Or are they both OK.

    My priamry concern and why I wrote the thread in the first place was to validate if MY method (tighten to no play, then tighten an additional 1/4 turn) is OK or not.

  18. #18
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by RkFast
    But what youre seeing on the CC site is different than what people here are saying, so now Im MORE confused!

    Park, King, Cane Creek, Zinn all say to tight till you hit the point of no play...then leave it there.

    Lots of people here are saying tighten past the point of no play...until actual binding occurs (MUCH tighter) and then back OFF 1/4 turn.

    Which is better? Or are they both OK.

    My priamry concern and why I wrote the thread in the first place was to validate if MY method (tighten to no play, then tighten an additional 1/4 turn) is OK or not.
    They are not really different. Going past and backing off is a method to produce zero play - it works because going past accounts for other friction, then you take it off until it is smooth but without play. It is a "bracketing" method - by finding too much and too little you can estimate exactly right.

    The 1/4 turn loose is just a method for learning how to adjust it - not the whole goal. Attempt to make your headset as smooth as possible, whether it take 1/8, 3/16 or 1/4 turn off too-tight to accomplish it.
    Get a better saddle: www.kontactbike.com

  19. #19
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032
    Quote Originally Posted by Kontact
    They are not really different. Going past and backing off is a method to produce zero play - it works because going past accounts for other friction, then you take it off until it is smooth but without play. It is a "bracketing" method - by finding too much and too little you can estimate exactly right.

    The 1/4 turn loose is just a method for learning how to adjust it - not the whole goal. Attempt to make your headset as smooth as possible, whether it take 1/8, 3/16 or 1/4 turn off too-tight to accomplish it.
    I do understand. But agaiin...to make sure.....you really have to WRENCH the sucker down to do damage, correct?

    I guess I read these manuals that are vauge and say I can do damage by going too tight and when combined with working on parts and bikes than cost so much, I get worried.

    Guess Im trying to figure out just where the threshold of doing real damage and premature wear is.

  20. #20
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by RkFast
    I do understand. But agaiin...to make sure.....you really have to WRENCH the sucker down to do damage, correct?

    I guess I read these manuals that are vauge and say I can do damage by going too tight and when combined with working on parts and bikes than cost so much, I get worried.

    Guess Im trying to figure out just where the threshold of doing real damage and premature wear is.
    Unfortunately, if the headset is binding, it is doing damage to the bearings. When you get it right the bars will swing without any notchiness. Once you've done it right, it becomes pretty obvious what is right and wrong.

    If you aren't sure, take it to a shop and ask (or pay) them to show you how to adjust it correctly.
    Get a better saddle: www.kontactbike.com

  21. #21
    RoadBikeReview Member
    Reputation:
    Join Date
    Dec 2004
    Posts
    2,032
    Quote Originally Posted by Kontact
    Unfortunately, if the headset is binding, it is doing damage to the bearings. When you get it right the bars will swing without any notchiness. Once you've done it right, it becomes pretty obvious what is right and wrong.

    If you aren't sure, take it to a shop and ask (or pay) them to show you how to adjust it correctly.
    OK...binding = damage. Got it!

  22. #22
    RoadBikeReview Member
    Reputation: AvantDale's Avatar
    Join Date
    Dec 2008
    Posts
    1,294
    I just keep turning till these no more play in the headset. I'll check by grabbing the front brake and rock the bike back and forth. I've never tightened it to the point of the headset binding. Thats alot of force. Lol...its not that complicated.

    If you keep tightening and the headset still has play, check the spacers above the stem. There might be a chance that you don't have enough spacers and the top of the steerer is hitting the top cap.

  23. #23
    RoadBikeReview Member
    Reputation:
    Join Date
    Feb 2004
    Posts
    5,297

    last post on this

    Quote Originally Posted by Kontact
    You're not getting me at all, I'm afraid.

    The actual load on the bearing is low - a few inch pounds. The torque you feel as you tighten the cap is the torque it takes to overcome the friction in the system and the weight of the fork (if it is in a workstand). Once you've compressed the stack of bearings, spacers, grease, compression cap, etc, the torque stops being necessary - now you're down to actual bearing adjustment.

    To give you an example, a new bike that has never had bearing sit in the built in cups may take more initial torque to compress down than the next time you take it apart. And a bike with an integrated headset is going to need different compression from a tradtional, and a 1 1/8" is going to require different than a 1". So one torque value isn't going to work, especially when the seating torque is higher than the bearing preload torque.

    Bearings are one of those things that are mostly adjusted by feel - doesn't matter if it is a car axle or a bike headset - that's the way the things work and are designed. Same thing with wheels - the torque needed to turn a spoke nipple is not the same as the resulting spoke tension. That's just the unfortunate way the universe works - you have to think when you're doing some of this stuff.


    I set headsets all day. I'm not making this up, and if you use a torque setting your are probably over-compressing your bearings, which will eat them up.


    Instead of working off a mis-memory of what a third party may or may not have said, how about reading a manual:
    http://www.canecreek.com/manuals/Hea...structions.pdf
    Do you see a torque value anywhere in there? Who's right - me and Cane Creek, or you?

    Not sure why this is such a hot button issue for you, but Zinn does indeed suggest a value of 22 inch.pounds for preload. He does state that is a starting point. Maybe Zinn is just wrong here but he is a respected mechanic and author. I think everyone would agree that checking for tightness/looseness after the top cap is installed and again after the stem is torqued is the right thing. I don't think there is anything wrong with your method but the OP did ask for a value and I gave him one from a recognized source

  24. #24
    RoadBikeReview Member
    Reputation: cyclesport45's Avatar
    Join Date
    Dec 2007
    Posts
    544
    Any time your headset needs adjusting, buy a new bike. The only way to have a properly adjusted headset is to have it done by the "new bike headset gremlins". New bike headset gremlins can only be found in Taiwan, or sometimes in China. No one else will ever get a headset adjusted properly ever again.

    Ever. Nobody here ever actually touches our own headsets. We just like to pretend.
    Just ride.

  25. #25
    Happily absent RBR Member
    Reputation: Kontact's Avatar
    Join Date
    Apr 2011
    Posts
    3,940
    Quote Originally Posted by DaveG
    Not sure why this is such a hot button issue for you, but Zinn does indeed suggest a value of 22 inch.pounds for preload. He does state that is a starting point. Maybe Zinn is just wrong here but he is a respected mechanic and author. I think everyone would agree that checking for tightness/looseness after the top cap is installed and again after the stem is torqued is the right thing. I don't think there is anything wrong with your method but the OP did ask for a value and I gave him one from a recognized source
    Starting point. Which isn't "a general idiot proof rule on top cap adjustment for a headset."

    The second part is also wrong. There is no reason to do any tightening of the cap after the stem. It doesn't do anything at that point:
    http://velonews.competitor.com/2010/...reaking_136734

    Hot button? Aside from all the people with toasted too-tight headsets, not really. I'm not able to find Zinn's 22 thing online, but there's no reason to use a torque wrench to overtighten something before backing it off, so I don't understand the reasoning of stating a torque value. And when trying to explain to a novice how to adjust a headset, it just confuses matters to suggest that a torque wrench increases the ease or precision of the adjustment.

    May I ask why it is important for you to put out and defend information that is absent from or in conflict with the manufacturer's service instructions? Like the weird post stem tightening thing.
    Get a better saddle: www.kontactbike.com

Page 1 of 2 12 LastLast

Posting Permissions

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

THE SITE

ABOUT ROADBIKEREVIEW

VISIT US AT

© Copyright 2019 VerticalScope Inc. All rights reserved.