Bendy FSA-SLK Crank chainrings ?

After having a power meter installed (Ergomo) I had to change my cranks from Dura Ace to FSA or more to the point the choice was limited at the time. This in itself did not seem to be much of an issue at the time, however strange things have occured since and the chainrings seem to have a life of thier own.

I did not have any means to directly measure the chape of the chainring profile, so thought the camera resolution would provide a good view of what I am seeing.

To clarify what was going on I setup my camera on a tripod (to eliminate camera shift) and then took pictures at 6 tooth increments around the chainring, which then gave me 9 shots (well about 9 in the end after about 40-50 trial shots).

From each of the pictures I took a slice at 800% enlargement (so that the pixel size was hightly visible) at the location shown on the top image and then shrunk it width ways into Excel so that the tooth profile was visible again. Each of the 9 images was processed in this way to give an end picture as shown to the right.


As a secondary means of measuring the distortion I then shrunk the row height to 0.75 and marked off each row position that correlated to the tip of the tooth on the chainring, took the row location and normalised the row numbers to a zero base, which gives the figures in the base of the picture.

The first image represents the right hand crank (chainring side) pointing upwards to 12 o'clock.

What stands out is the profile is non eliptical and distorted in 2 areas with the main distortion with the crank position around 90 degrees (each picture is a 40.75 degree inrement from (6/53)*360 degrees) that resulted from the chain in the 12 tooth sprocket on the cassette when I was training as this either creates a chain angle just enough for the pedal force to bend the chainring or the carbon fiber crank arm is distorting ?

I'm a bit puzzled as to why they are bending, although I can bend them out of shape in the opposite direction if I end up climbing in a 53-23 combination and put on a lot of pressure. That said, the doo seem to be very stable sub 1000W, it's just when you go above this some jelly seems to become involved.