Friday, January 19, 2007

Run Time?

Battery life is very important when you ride at night. I like to use the analogy that it is like water in the desert. The last thing that you need on a night ride is to have your light shut off before the ride is done, plunging you into darkness and an inopportune time -- like bombing down a rocky section of single track. Run times are always given by the light manufacturer, but you don't know what methodology they used to get that figure. It might taken from actually watching the light for a period of time, or they might calculate it from the Amp-Hour rating on the battery. However, batteries behave differently depending on the age of the battery, the time since the last charge, the ambient temperature, etc. I've particularly noticed a severe reduction in high beam run time with the Li-Ion batteries -- which most of the lights in this review use. They'll kick down to a lower setting sooner when the temperature is below freezing. That's just a fact of the battery chemistry, so it's not one light or the other, but across the board.

So, what do I do to determine run time?

Well, I don't have time to sit and watch 15, or so, lights. Instead, I let a computer do it for me. You saw the results last year in the graphs like the one below.

This graph is for the Blackburn X3 light. What does this tell us? Well, that the light output is fairly steady for a little over 200 minutes (or a little more than 3.3 hours). Then it cuts to low power for another 100 minutes, or so. After which, the light turns itself off to prevent damage to the battery. This is what a regulated power source looks like.

The graph below belongs to the Princeton Tec Eos. This is a 1W LED that is regulated for the first bit, then drops of like an unregulated LED, just for comparison.
As the batteries (Alkaline) run down, the light gets dimmer.

How I make these graphs is with the following: One photo-voltaic (solar) cell from Edmund Scientific, a Di-194RS terminal panal from Dataq Instruments and I also use their WinDaq data acquisition software to record the voltages.

I made a small stand out of PVC to hold the lights and I shine them on the photo cell from 7 feet away. I use a small (8" diameter) fan to keep the lights cool. If I neglected the fan, the lights would overheat, or their internal thermal management system would kick in and dim the light to keep them from overheating. Either one would throw off my measurement. The run times are done at room temp, approximately 71 degrees Fahrenheit. I do the test at night so that there isn't any fluctuation in the ambient light.

So, there it is. It's a simple set up that allows for trouble shooting of your lights. Do you think that the run time isn't what it should be? Maybe you have an older light system and you want to know what kind of run time is left, this is one way to find out. It's relatively inexpensive and very easy to set up.


Stephen O. said...


As my aged Cygolite faded into chthonic blackness during my last moonless commute, I kept telling myself that I'd soon see the light at least in regards to purchasing its replacement. The suspense is killing us (top say nothing of the streets here in unexpectedly Arctic Boulder, Colorado). Enlighten us, gentlemen! Your public awaits!

James Sharp said...


Honestly, I am not delaying this to build anticipation... though it's not too bad :).

I am working as fast as possible to compile everything. I am shooting for early next week. Bear with me here, it'll be worth it, I promise.