I’ve long acknowledged a problem with my method of geometry development. I wasn’t prepared to deal with or thought that I could run the clock. The day has come to get it resolved.
Looking back in time, being forced to build my special handlebar systems in 2018, and even before that when I started working with the klunker platform, specific placement of the bicycle’s handgrips have required a huge amount of design effort. I’ve figured out new geometries, formulas, and methodologies to produce them at a very high level. I even pushed Brent to add quite a few dimensions and driving parameters to BikeCad.
This obsession carried over into may other areas of laying out a bicycle. My print detail and documentation advanced at a geometric rate. This evolved to produce Cyberdyne and Samantha, the most advanced frame fixtures that I know of developed to date. Handlebars will do this to a guy. Be careful.
What’s the problem that persisted? I’ve been describing the hand grip location at the handlebar referring to the end of the actual bar. We knew that this was not the most precise location to reference but it was the most expedient given the level of detail and description we had available. Since BikeCad is truly a two dimensional CAD system, solving in a more spatially nuanced way was going to be difficult and it remains very useful to have agreement between my BikeCad files and my SolidWorks files.
Today, this is getting fixed! Comparing divergent configurations has become more critical. Over time, the bar sweeps that I’m using have moved up to the 20 degree range. Remember that most commodity bars for mountain biking are in the 9 degree range. The correct location to reference the bar would be a bit inboard from the end off the handlebar. This is a little vague. As of today, I’m looking at 60mm. I’m very open to a kinesiologist helping me to find a better value. (LMK!!). Our goal here is to be dam close. Perfect is not really possible as a system.
60mm inboard is proximate to my 3rd metacarpophalangeal (MCP) joint. I think the real point would be about 15mm toward the wrist from here. This was chosen as it seemed to be a point that a changing wrist angle may have the least disruption in the body position. This could be any distance that the designer determines to be most advantageous. Since I have larger hands, folks may want to evaluate this for their hand size.
-
Old convention:
-
HGR/HGA → point at the end of the handlebar.
-
-
New convention:
-
HGR/HGA → point 60mm inboard from the grip end.
-
I’ve rebranded the HGA (hand grip angle) and HGR (hand grip radius) in my prints. They now point to a location 60mm inboard of the end of the bars. Below, we can see the Corsair described with a 842.0 HGR and 56.6 HGA. This is different from the earlier values posted (in update, note the 10mm spacer rather than 5mm) of 837.5 HGR and 58.0 HGA that pointed to the end of the bars. This will cause some difficulty in looking at older bike fits. That’s just the cost of forward change and part of why this change took so long to implement.
This might seem like a trivial detail to most. It’s shouldn’t be. When describing a bicycle (or anything really) we are looking for predictability and repeatability. This becomes even more crucial when you are making $600 (retail) dedicated and integrated handlebars that cannot be replaced with commodity parts. Here’s the rub, if I’m looking to replicate a bicycle with a specific riding position and the one value that I change is the bar sweep angle from 10 to 20 degrees, my true hand position can be over 10mm different using the end of the bars to refer to. This type of error can also happen if I’m changing the roll angle. This is unacceptable and not a good way to live.
-
Δy ≈ 60·(sin 20° − sin 10°) ≈ 10 mm
-
Δx ≈ 60·(cos 10° − cos 20°) ≈ 3 mm
Don’t confuse this as a method for fitting the bicycle. Different handlebar widths will change the location required for the handgrips in the elevation view. Testing will always be required to establish target locations and changes in development. I’m working to keep fit as predictable as possible within a few parameters as they change. Those being bar sweep and roll. These are parameters that should effect the feel and interaction with the bike rather than change body position.
I had gone through the process of accounting for this when designing the Corsair (from the M2-F2 and Starfighter) but I had been a bit quick and dirty. I was also changing the riding position some. This weekend I automated the process so that all future bikes will have a system in place.
I did pretty good with the Corsair, even after adding 5mm more spacer to it during testing. Hand grips are rising these days and the target moves. So the several months lag between the development of a bike and the initial testing sees change. Predicting the future is hard, they say.
Below, I’m showing some select equations from the SolidWorks master part file for the handlebars. From a construction perspective, the end of the bar is still critical but we want it expressed as dependent on the reference location. A little trigonometric slight of hand goes a long way here.
I’m going to make a little video to help clear this up for some people that can’t see it clearly yet.
It would be great if Brent could do some to help build this idea into BikeCad. Being able to fix the handlebar more precisely is very important to me…but I might be the only one.