I am thinking about drilling a hole in my Specialized Aerofly Carbon Handlebars. I am running Di2 and have my front junction box inside of my stem. What I am considering is putting a 1/4" hole in the center of the handlebars so that the cable could go right into the stem and be hidden.
At first the idea seems crazy, but the forces seem really small and if I drill a hole and get some epoxy around the edges of the hole so the the carbon doesn’t splinter I would think it would be safe.
I am running a 3T ARX stem, which has 4 bolts with the face plate being 1-1/2" wide. If you figure that the max forces on the bars come from sprinting, if you analyze the system as a beam that’s 1-1/2" wide with a 7" cantilever on each side with opposing 300# forces (seems high to me), then the only load on the bars where you drill the hold is a moment of 180 ft#. Shear forces are near zero. Now it doesn’t sound so crazy.
I think it would be OK but as mentioned, what about picking up the Cervelo bar with the hole already designed for this application. They have all of that extra material in the front which makes it stronger and it is pretty aero. Not sure about the weight of either but they should be fairly close I would believe.
This may be a little bit of a lottery and I’d be nervous about doing it.
The problem is you don’t know how the layup is designed, what the design assumptions were or how over-engineered it is. A bar designed to have a hole would undoubtedly take this into account in the layup. Whereas depending where you drill through the bar you could cut through a crucial section and destroy the structural properties of the component or you could have virtually no impact if it’s an area that is taking very little stress and really just providing a surface. If you have a good understanding of engineering you have a better shot at avoiding a problem but even then you’re working with assumptions and limited data, i.e. it’s a lottery. Given the risks if a bar were to fail, I don’t think I’d chance it.
I made the holes a shade bigger to run the cable plugs, first gen DI2 ( I did not drill new holes). The carbon handle bars tend to be pretty stoutly built - at least the ones I have. Slowly applied loads shouldn’t a problem. Forces applied very quickly - like a hockey stick, I’ve have carbon sticks snap in two like nothing. $couple hundred poof. So if you do decide to drill, you might want to seal the edges with epoxy.
I just got a 3D printer and will be making Garmin/GoPro/Di2 junction box mounts before the spring races roll around. It shouldn’t be too hard to modify the model to fit a SRM. I have a PCV I can use to test the fit.
Lots of assumptions in the works here, but size is your enemy. If you are willing to chop and solder your Di2 wires (such that you only need to pass the wire through and not the whole e-tube plug) you could potentially get the hole smaller and decrease your risk. Of course, that also becomes much more logistically difficult to feed the wires through…
I am on the original 10s Di2, which has the big plugs, hence the size of hole I want, but I might have to chop and solder regardless just to get the wire out the hole from the inside of the bars.
This is the main issue, if the bars were designed for open hole compression then, by definition, you could drill a 0.250" hole anywhere and the structure would still have positive strength margins. The issue is that layups which are really good at OHC, are not really stiff (say 6 Msi rather than the 18+ Msi that we would prefer to have for bike components). Typically frames/bars/wheels are designed for stiffness and contain a lot of unidirectional material oriented along the principal loading axis - ‘hard’ laminates like this carry substantial knockdowns for OHC. Matrix choice plays some part in this as well, but for the purposes of our discussion it is a lesser concern if ignored.
The other concern is stress concentration factor. For isotropic materials (metals) the Kt for an open circular hole is 3.0; meaning the stress at the edge of the hole is 3x the stress if there was no stress riser. Fun story, as the laminate becomes less isotropic, or more anisotropic, as is with composites the Kt goes up. It is layup dependent, but I’ve seen it in the 4-5 range for highly anisotropic hard laminates.
Using sigma = Mc/I the stress might be ‘low’, is it still low at 5Mc/I? So you got a stress, cool. What’s the stress allowable for that laminate (not the lamina, the laminate), conversely, we really should be doing Mc/EI to get a bending strain. E is so tricky though, that’s the problem. Furthermore, that’s for a specific load case - cantilever beam something something something. Except it is never that simple. The case you proposed also has a non trivial torsion component - unless you’re riding flat dirt bars… There are other fun load cases too… putting the bike in the car and the bars hit a seat, bike resting against the wall and falls over, etc… Different load cases, different stress directions, different structural problems.
As far as a resin wash of the hole after drilling, sure knock yourself out. If you use a clean sharp drill bit with a good spindle speed and low feed speed, that should generally minimize chipout or splintering. In either case, the damage would be done and the resin wash would really only serve to improve the aesthetics.
If you knew the layup and materials, the answer would be a simple yes or no. Since we don’t know that, its a tricky maybe with unquantified risk.
In addition to any stress concentrations, you don’t know if you’ve damaged the fiber or resin beyond the drilled hole. I doubt you have access to non-destructive evaluation equipment to tell if you have any sub surface delaminations.
Crazy to even think about it without knowing the carbon layup. I’ve seen too many bars break during testing and the last thing I would do is drill a hole in it.
What Shinny said. The original designer of those bars could do the math and tell you where you could drill, if you could drill. Anything less than that is a gamble. Why gamble with crashing?
Assume that the bars are going to snap at the worst possible time. Then what happens?
A few seconds isn’t worth your health. It isn’t like there aren’t plenty of reasonably priced aftermarket bars out there which do exactly what you want. (And they’re in the classifieds if you can’t afford new).
If, by some miracle, the OP had access to equipment (I’m thinking a material test machine, strain gages, and extensometer, as well as an ultrasound head) we could likely qualitatively (not quantitatively) assess the relative risk.
A couple assumptions could be made and we could probably get within, say… 30%. There’s a good chance the bars have a huge strength margin, since they’re likely stiffness driven rather than strength.
As far as nailing the bars on the centerline between the stem clamps… As much as we could make a section cut and assume the load reacts to zero at the clamp, we could also assume the system is represented as four-point bending - whereby there is no shear force in the beam between the clamps, but rather pure bending. It largely depends on the boundary condition which is assumed between the bars and clamp, and how realistic that is - I can almost guarantee it is not a pure fixed or pure pinned condition (it’s likely a fixed condition with some compliance since I do not believe we would see half the clamp react the bending from a half-bar). Again, this assumes only a simple bending case - we would expect there to also be axial, shear, and torsion loads to further muddy the waters.
