> I want to build a device that flashes a rear brake light at an intensity = depending on the G-force due to braking. > I want it to be super easy to install. Such as throwing it in the back w= indow and being done. This would possibly mean a battery and solar charger= .. > > #1 I do not want to use data from OBD2 ......... This would increase in= stallation issues. > #2 How might I differentiate between G-force due to braking and that due= to the vehicle tilting nose up or nose down ? > > I was thinking maybe a gyroscope might do the trick. The device you describe will have a significant affect on following driver behaviour only if it meets the Olin educational threshold. ie it must cause pain if itis to work. Practical experiments carried out 30+ years ago with such devices in a fleet of Australian taxis (in Melbourne) (and probably in numerous other places before and since) showed that if the light was bright enough to cause pain and discomfort to following drivers, such that people complained about the lights, then measurable reductions in rear-end accident rates occurred. At lower brightness levels sch lights had no significant effect on accident rate. From dim and distant memory, the lights were named "Cyberlight"s, had acceleration dependant brightness control and were not a vast success. ______________ If you do decide to try this regardless, a single single-axis accelerometer will suffice as a sensor.. Two may make life easier, but not by much. You could get enough information for a sensor pointing at almost any direction in a plane of a 2D image seen looking side on at the car. But easiest is a single accelerometer pointing front back. When horizontal A =3D 0. When on a front/back slope you get a steady reading. This changes relatively slowly with incline of road. Braking will typically cause a substantially faster change and will be reasonably transient. A 10 degree slope delivers 17% of gravity to the sensor. A 20 degree slope delivers 34% of gravity to the sensor. BUT A 10 degree slope still delivers ~=3D 98.5% of braking to the sensor. A 20 degree slope still delivers ~=3D 94% of braking to the sensor. __________________ Internal battery is easy. If mean operated power is 10 Watts and operating time is 10 seconds mean then 6 x AA cells (Alklaine, NimH 2500 mAh) will operate it over 500 times. Adjust assumptions to suit. _________________________ Solar powering in that environment is hard. The panel will only provide anything like full power when direct sunlight falls on the whole panel surface. (Through glass is OK). If the whole of the panel is not directly illuminated by sunlight then output will be far below full output - often only a few percent of maximum in such situations. When the panel is fully illuminated by direct sunlight the energy actually stored in the battery will be at about half the maximum rated power of the panel x transfer time. eg a 1 Watt panel will store about 0.5 Watt hours of energy in a battery in one hour of full direct illumination. A PV panel attached to (flat against) the glass is probably as effective as you are liable to achieve in that situation - average power out will be well below nominal rating. Russell --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .