In order to examine other battery alternatives, we must first get an idea of what the current usage (battery drain) is while using the 95LX. This is not as easy as it sounds because there are a wide range of operating conditions on the 95LX. While it is off, the 95LX consumes approximately 0.7-0.8 milliamps of current. If it is turned on but not performing any operations, it consumes 30-40 milliamps when the serial port is off and 40-50 milliamps when the port is powered. While the 95LX is actually computing, the current consumption jumps to 100ma-200ma. The exact amount depends on whether the 95LX is performing memory only operations or disk access operations. Since this represents quite a range of values, the determination of an absolute battery life figure is impossible.
Since we are looking at the suitability of various technologies, we are really only interested in their relative performance.
Therefore, for the purpose of this discussion I have picked 50 milliamps as an average consumption figure for a RAM card equipped 95LX.
In order to provide a repeatable test for the different battery types, I used a program that caused the 95LX to consume a relatively constant amount of current. Further, the program tracked results by monitoring the 95LX's internal battery management system and writing data to a file any time there was a change in battery level. The 95LX has an Analog-to-Digital converter built into its battery management system. This converter divides the 0-to-5 volt charge up into 256 discrete voltage levels. The ability to detect when the battery charge decreased one level and then write the results to a file allowed me to create a test that could be reliably repeated with the various battery types.
Since the low battery warning appears at 2.0v, this voltage represents the end of the useful life of batteries on the 95LX.
This is a significantly higher value than is commonly used in the battery industry for comparison testing; therefore, some of the battery life estimates you'll see are not always relevant to the 95LX.
An example will help clarify this. A typical 1.5v alkaline battery is rated at 2200 milliamps-hours. This means that if you were to run a 3.0v device that consumed 2200 milliamps, it would run for an hour before the two batteries it required would be considered dead. Using our average consumption figure for the 3.0 volt 95LX of 50 milliamps, this means that the Palmtop would run for 44 hours on a pair of 2200 ma batteries. Unfortunately, most battery companies consider a pair of batteries `dead' at approximately 1.4v, instead of the 2.0v used by HP. In my tests, this decreases the battery life estimate of a pair of 2200ma batteries to approximately 30 hours (at the 50 milliamps rate).