January 24, 2012 by John_Eggers
Posted in
In building carbon bicycle frames we use the traditional tube to tube construction method. Most builders tack the tubes together to get the right geometry, then wrap each tube junction with either carbon/and or Kevlar® woven tape to create a lug. This can be done using either wer layup or prepreg.
I have noticed over the past few year a few companies are simply gluing the tubes together to form the frame skipping the lug reinforcement step. Time in France and Stitch in Italy are using this approach and it does appear to be working for them without problems.
Is the glue out there now really strong enough to bond the tubes into the frame without the reinforcement lugs?
How can I test this using the software to see if is strong enough?
If this is possible which epoxy would be the best choice for this application?
- John_Eggers's blog
- Login to post comments
Internet Explorer 7 + preferred.
John, Most glued joints are
John,
Most glued joints are shear loaded. So you need to have reliable shear data on your glue. Most such glues are epoxy with a thixotropic additive such as Cab-o-sil and some fiber or nano-fiber as reinforcement. You could make your own by adding milled carbon fibers, for instance, then test it. Milled carbon will make the joint more stiffer and as a consequence will give less than if you put milled glass, but glass will be weaker even if it may "give" more.
To estimate strength you could use rule of mixtures for a homogeneous composite (taking the milled carbon in consideration), and use that to compare with tests.
Once you have the shear strength, then do a FEA model of the joint and use that shear strength between the joint and tube and have enough shear area at the glue-joint for the transfer of all loads between the tubes joint. I would use a high factor of safety for the shear transfer through the glue.
I'm sure someone else may have a different idea, but the above works for us. We have very few joints, but they're all made from our own formulation in the aero-grade epoxy we use.
best regards,
Naresh Sharma, Ph.D.
www.nashero.com
Thank you
Naresh,
Thank you so much for your input, it is greatly appreciated. I will move forward using the approach you outlined.
John Eggers, Ph.D.
American Artisan Bicycles Inc.
Carbon Frame Tubing Bonding
Hi John,
I have some ideas in this area also. I have been working on properties of some specific epoxy resins which might be useful for this type of study.
My opinion - joints and connections are typically the weak link in most structures. They are areas with greatest stress concentrations, greatest nonuniformity of load, and greatest difficulty in fabrication. In general, I would not rely on just gluing a joint together to carry the day on a critical structure where safety is a concern.
Regards,
Kim Parnell
-----------------
T. Kim Parnell, Ph.D.,P.E.
PEC - Parnell Engineering and Consulting
Sunnyvale, CA 94087
E-mail: kim.parnell@stanfordalumni.org
http://www.parnell-eng.com
http://www.linkedin.com/in/parnellpec
Skype: kim.parnell_msc
Kim,Thats interesting.
Kim,
Thats interesting. Fokker 100, Fokker 70, american airlines bought around 200 of those 98 seat commuter jets in 1992 and some are still in operation, and Fokker 28 had their thick machined aluminum wing skins "GLUED" to their aluminum spar. No rivets or bolts. There's been no squawk from those in operation for well over 35 years (considering F28).
If you do a proper analysis of the joint in FEA including service loads with a proper safety factor, why shouldn't one use loaded-epoxy to join two parts together? Its done every day in a variety of application, the above was just one case where its critical.
best regards,
Naresh Sharma, Ph.D.
www.nashero.com