Thursday, December 31, 2009

19 Thomson Masterpiece Seatpost Failure

In this post, we will explore the mechanics of a simple no-nonsense bicycle seatpost collar. What encourages me you ask?


I was alerted to a Thomson seatpost failure at the Weightweenies forum. The item is a "Masterpiece" brand, the lighter cousin of the utilitarian "Elite". The post severed at the seatpost clamp as an image provided shows. Thomson makes everyone's favorite seat-posts.

As we often have it in such cases, the claim from the poster is that a torque wrench was indeed used to install the product and it was done so by a "pro mechanic". By that, I take it that he knew what he was doing. By that, I meant that he did not give a "few extra" turns than required to tighten the screw.

You may know that I have explored two other Thomson seatpost breakages in detail this year. Those efforts involved a background study of Thomson's seat post design and interviews with the user and the product person over at Thomson. Statistically, in the end we all had to agree to the idea that such failures result more from a common human condition called the "Over-Tightening Syndrome". I have written about it here with the help of Thomson's Dave Parrett.

As for this new breakage, it is advisable to let Thomson know so they can test it for signs of extraneous use of force. There's also a good chance they might replace it for the customer on goodwill/warranty.


THE MECHANICS OF A SEATPOST CLAMP


The common cyclist's seatpost clamp may be closely approximated by what they call in engineering circles as a "Split-Ring Clamp Type Shaft Collar" (there are all kinds of weird shapes and designs out there for bicycle seatpost collars but we will limit this discussion to circular). A hex socket cap screw is used to clamp the collar around the seatpost so it constrains it from motion/slippage due to seated weight acting downwards.

When the collar is set with the cap screw, three brilliant things are taking place in your equipment that you probably never heard of :

1) Pre-Loading The Cap Screw : As the cap screw is tightened in the collar, it is stretched to produce the clamping force regardless of the external load. This force, or initial tension, is called a "Pre-load". So without considering the external load, the torque necessary to achieve this pre-load is given by the relation :

Fig 1 : Pre-Load Screw Equation. Courtesy for K table : Bowman Fasteners

The seat post manufacturer or supplier is to select/supply capscrews that have sufficient pre-load capability. They specify diameter, pitch, threads per inch etc of said screw for the particular service. Pre-load force is found from the torque desired.

The pre-load stress is found by dividing pre-load tension over the tensile stress area for the fastener from reference tables. The same process is used to calculate service load stresses, i.e external stress divided over the tensile stress area of the fastener. When you have these two quantities, a pre-load stress and a service load stress, you compare them both to the known proof strength of the said fastener from reference tables. Ideally, they should be lower from the proof strength by a sufficient safety margin because proof strength is the maximum load a screw/fastener/bolt can take before acquiring a permanent set (plastic deformation, no good).

2) Axial Holding Force : When tight, an internal pressure acts on the collar surface due to contact with the seatpost through its internal diameter. There exists an axial holding force from the collar that is proportional to the co-efficient of friction between the collar and seatpost and the torque in the screw. Avoiding lengthy integral based derivations, axial holding force is given by :

Fig 2 : Axial holding force of a seatpost clamp/collar

3) Hoop Tension : A tangential component of stress acts inside the collar at any radius due to the action of internal pressure of the seat post mentioned above. It is called "Hoop Tension". By relating this tangential stress αt to hoop tension, we arrive at the following relation :

Fig 3 : Collar hoop tension

When the collar is set by tightening the cap screw to some torque value, the hoop tension is equal to the screw tension. This is the role of the screw - to induce hoop tension so that it produces clamping for the seatpost.

Armed with that knowledge and the above 3 equations, one can easily find the tension in the screw and the internal pressure of the seat post on the collar. While out of the scope of this blog, you can even calculate the maximum shear stresses in the collar and find out your factor of safety for the design so that it is safe for a rider to use.

You can take this information further and ask - what happens if I screw my cap in too much, vastly exceeding specified torque values? The result could be too much hoop tension, too much clamping force, too much shear stress, too much pre-load. Its a combination of these things and/or manufacturing related defects and the kind of external loadings induced by the weight of the rider and his usage patterns which all set a clear path to failure.

So be cognizant of these things... and why, Happy New Year too! :)


19 comments:

  1. Anonymous9:45 AM

    This is a helpful post, but honestly the moment I saw your approach to handling the OP's comment, I went 'oh no, here we go slapping the blame game on him again' What extra tools should a pro-mechanic arm himself with besides the fact that all he needs is a simple torque wrench and some experience to dial it in? It is slightly ridiculous. While I know Thomson make reliable products, like I posted before in your previous articles...its statistically impossible to make perfect designs all the time. So you can keep giving the user a free post as bandaid everytime, but what takes care of his character assassination?

    ReplyDelete
  2. Anon : I did not engage in any character assassination. That's an amusing comment. All I said was there while there is a chance for manufacturing defect, there is a higher likelihood for the user unaware of it, to overtighten. It need not happen the first time durign installation. Subsequent riding induces vibration in the join and can loosen the screw a little everytime. The chances that the rider takes their Allen key and gives it a few extra turns for better safety may actually not going give them safety. Good day to you...

    ReplyDelete
  3. Scienceguy1:34 PM

    The owner needs to check the calibration of the torque wrench he's using. I was trying to look for the model he used but as I found out, this isn't even his post, it's a friend's and he's speaking in third person tongue.

    Anon above : Chill the **** out. It's the new year for chrissake.

    ReplyDelete
  4. Ron, you are one smart cookie. I have been lurking here for a long time and finally feel confident enough to post. For what it is worth, there is no character assassination here. There is no publishing of his name, phone number, or address. You could have also published his/her email address.

    Thanks for all the great information. You are a great resource.

    ReplyDelete
  5. Mike : Thanks for the support of the material presented here. I don't want to sideline/persecute the first comment but that's the last thing that crossed my mind, that I need to do a character assassination in today's post. Happy New Year and always come back more for something interesting! ;)

    ReplyDelete
  6. Ron,

    These are the kind of posts I come to this blog for. Have a sound new year ahead.

    ReplyDelete
  7. the seat post seems surprisingly thin to me. maybe its because im more used to the elite. another visual clue that you should torque that thing too much.

    ReplyDelete
  8. Thirty-Six, Rec.Bicycles.Tech6:00 PM

    The overtightening comes as the result of using grease which allows
    the post to slip despite greater pressures. Using thin oil lets it be squeezed out allowing the clamp to grip the post adequately at lower pressures. You only need to tighten the clamp sufficiently that you cannot twist the saddle using reasonable force. Using thin oil in the seat tube and on the seat pin keeps the required clamping force low and failures unlikely. If you wish to seal the seat tube from water ingress, drip molten wax around the top and the pinch area.

    ReplyDelete
  9. Anonymous4:27 PM

    I was thinking of getting a Thompson seatpost, then I saw the Ayn Rand quote on their website..... Hello Nitto!

    ReplyDelete
  10. “Throughout The Centuries There Were Men Who Took First Steps Down New Roads Armed With Nothing But Their Own Vision”

    ReplyDelete
  11. Thirty-Six :

    Interesting comment from you on grease. Thomson's instructions for both stems and seatposts call for grease, not anti-seize, only on the threads of the bolts. According to them, greasing under the heads of the bolt is not needed and can greatly change the amount of force generated by the bolt at our recommended torque, this can lead to premature failure of your components. With all Thomson products they plead with the user to take special care to grease only the threads of the bolt and use a torque wrench.

    ReplyDelete
  12. Nice description of the "static" physics on a seatpost clamp. Thanks Ron, always great for an engineer/amateur bike racer.

    Rod

    ReplyDelete
  13. Thirty-Six, Rec.Bicycles.Tech7:46 PM

    They are arguing the same point from a different angle. If , as they
    suggest, you do not provide temporary lubrication under the bolt head it may tear the the surface upon which it seats. Providing a MACHINE oil on the thread and under the head prevents damage and allows the bolt to stay secure with the correct torque to provide adequate
    clamping force to the dry seatpin.

    ReplyDelete
  14. Patrick4:59 PM

    This will be great resource for "pro" mechanics and amateur mechanics alike. Follow instructions, and know whats going on in the fastener as you wrench it.

    ReplyDelete
  15. Also you can round out the hexagonal slot in the bolt head before it's tight enough to stop the seat slipping down. The point is if you don't grease under the
    bolt head, it can feel like you've done the bolt up much tighter than
    you have because it's just sticking against the collar.

    Then the seat keeps falling down, so you do it up tighter, to the point where you've rounded out the bolt head.

    I'm surprised therefore at Thomson's instructions. Seems to me it's more sensible to specify all torque settings on the assumption that everything is greased.

    ReplyDelete
  16. If you use grease rather than oil to grease your parts then something
    will get damaged if the post is retained fully during riding. I have used grease on seatposts and they slip, so out it comes, is cleaned and put in dry or with a thin oil and there is no further problem. Handlebar stems are not so fussy, a little give can be usefull to mitigate injury in the event of a collision. I set handlebars so that I can force them to a new position without resorting to tools.

    ReplyDelete
  17. Anonymous1:52 AM

    Hi Ron,

    Great information regarding on Torque Specs. and Bolt Thread Strengths.
    It may be very valuable to your readers to point out that this product has been on the market for many years with little to no significant quality issues.

    It may also be very valuable to go into exactly HOW the product was manufactured.
    I believe this product is made from and extruded tube which controls mainly the ID of the product.
    It is then CNC Lathed to create the outer dimensions of the Post and then basic shape of the head.
    Lastly it should be CNC Machined in order to crate the specific dimensions of the head.

    Now, all that being said, you must remember that since the material is simply extruded, there is a likely hood for it to not be exactly the same every time as it is essentially billet and therefore does not have the same structural stability or properties as a Forged Unit.
    It is this lack of structural stability which, when the design gets optimized, can lead to a mode of failure just as what you are seeing.

    In reality, the likelihood of the Root Cause actually being the Torque applied to the Seat Clamp is quite low for if that were true, there would be truckloads of busted Thompson Posts going back to Macon, GA every week.
    Rather, it is far more likely that the Root Cause is due to the raw materials used in the part.
    Without actually examining these materials, this cannot be confirmed, but as I stated above, this model has been in use for many years with little to no significant quality issues.
    - Ryan

    ReplyDelete
  18. Ryan : Thanks for the insight and it is interesting you point to a chance of manufacturing related defect. Yes there hasn't been any widespread breakages in the seat posts but neither have most seatposts shown "widespread" failures. The images like the one you see in the this post maybe outliers, extreme case scenarious. Thomson maintains that for the tube to break as in the image and not bend, there was a stress riser. They apparently know how much stress it takes to break the tube. Hence they don't see it as their fault as long as the user doesn't send the seatpost back to them for inspection.

    ReplyDelete

Thank you. I read every single comment.