TPB #90 – Analyzing Error in the Projectile Sensor

Posted by:

The error of the Camera Axe’s projectile sensor is analyzed. Experiments are done in software to isolate the error which turns out to come almost entirely from the 4 microsecond granularity of the Camera Axe’s timer hardware. Then math is done on that 4 microsecond granularity to determine how that affects the reported speed of the bullet in the Camera Axe and the positioning of the bullet in a photo.

For a typical pellet gun with 950 ft/second the reported speed has a granularity of +- 20 feet and the positioning of the bullet is to within a quarter of an inch. The videos has details on how this is derived and data points for other speeds of projectiles.

Remember all sensors have error in the real world. This video shows the projectile is useful for it’s designed use case, but also gives insight into it’s limitations so people know what to expect from it.

If you want to see more about this projectile sensor in action watch one of my other videos such as:


About the Author:

Maurice is a computer engineer, photographer, and all around renaissance man.


  1. Yves  May 25, 2015

    How about increasing the distance between the two photosensors? That should help increase accuracy, no?

    It’s not hard to imagine a sensor with 8 or 12 inches distance.


  2. Maurice  May 25, 2015

    Yes, that would help, but it would add cost.

  3. Dan  June 28, 2015

    Just an FYI: I have discovered that the projectile sensor can be problematic when used with arrows. I believe this is due to the arrow being longer than the sensor, and the inherent flex of the arrow when fired.