Note that this blog post is a companion piece to an article on the Dandyhorse blog where the city has furnished quite a bit of additional information.
I’ve heard anecdotal evidence that there is less traction on the green patches that have been incorporated into several recent examples of bike infrastructure around town. The City of Toronto advises that the green paint is a thermoplastic material that has been blended with grit to give traction similar to asphalt.
Having had training in both Physics and Engineering, I decided to see if I could measure the difference in traction between ordinary asphalt, and asphalt covered in two different kinds of road markings. After some thought, I decided to repurpose some equipment that our undergraduate students have been using to measure the tensile strength of materials. What is relevant to the present discussion is that there is a force sensor that can be hooked to a computer to record force as a function of time.
Here is a picture of the apparatus.
The red string pulls on a lever that presses against a silver plunger that sticks out from one of two blue boxes (the one to the left that is labelled “Force Sensor” in small green type.) The apparatus has four small rubber feet underneath. The idea is to drag the whole thing to the right by pulling on the red string while recording the force measured, which would be a measure of the frictional force.
The first set of measurements were done on a crosswalk on Runnymede Rd. Here is a picture of the setup being dragged.
I will note that this was done on pavement that was slightly wet from rain earlier in the day. I started the measurement on asphalt, dragged the meter all the way across the white stripe, and then some distance on the asphalt, just so that I could see if there was a different on and off the marking. The data for three separate runs are shown below. The horizontal axis is time in seconds, and I’m not going to quantify the force reading on the vertical axis.
As you can see, the measurements are fairly consistent, and that there is a dip in each curve (from 4-8 seconds), corresponding to when the meter is being dragged across the white paint as opposed to the asphalt. Clearly, there is a reduction in friction on the white paint.
Since each of the curves has quite a bit of scatter, I take the average value for the asphalt and the paint sections (with the standard deviation as the error). The results are as follows:
- asphalt: 5.4 ±0.5
- white paint: 4.4 ± 0.5
The net result is a 20% ±10% reduction in friction on the white paint.
Now onto the green bike boxes. The difficult here was that my ability to take measurements was somewhat hampered by the fact that I was dodging traffic while doing so. Here is a picture where I took the data, on the Annette bike lane at Dundas St. West.
The data is taken where I draft the meter off of the green paint. Here is the data:
If you are charitable, you might imagine that the first part of the curve, say from 1.5 – 3 seconds is a little higher than from 4-6. This would indicate that the traction on the green paint is actually higher than on the asphalt. The numbers are as follows:
- green paint: 5.6 ±.5
- asphalt: 4.8 ± 0.6
Therefore, in this case the variation in the data is comparable to the difference measured between the two surfaces, and so it is not possible to conclude that there is a clear difference. If fact, if there is a difference, the green paint might actually provide slightly better traction than bare asphalt. I will note that the asphalt section adjacent to the green paint was very rough (much rougher than the asphalt on the Runnymede bike lane), and this probably reduced the contact area between the rubber feet and the pavement, which would account for the comparatively friction reading for asphalt. At any rate, the green paint is very comparable in traction to bare pavement, and appears to give better grip that the white paint used for crosswalks (and I assume for bike lane markings etc).