After selling my heavy SLA, I have turned to those portable Lithium polymer battery and I have got one with solar charger and also a few LED as flashlight. It's rated at 2600mmAh with 4.5V, 5.5V and 6.5V and 9V. It has a mini-USB socket both for charging and powering.
The solar charging panel seems quite good and even with an overcast sky, the charger started working. Of course, expect extremely slow charging under this situation but then I won't rely on it but treat it as a bonus during solar imaging or observation.
This CXD-1 controller requires 160mA at 9V and it has a center negative pin. In order to reduce the complexity of the wiring, I've soldered another cable into the circuit board directly. That power cable is a retractable USB cable from a dead device:
I assume that the 2600mmAh rating is for 4.5V and thus, at 9V, it means 1300mmAh. Given these products were usually over-rated and also with the efficiency reduction due to voltage conversion, I guess it would mean 700mmAh?
So, it should work for 700 / 160 which means over 4 hours. At the time of writing this blog entry, the mount has been running for nearly 4 hours and there is no sign of low power yet.
After 5 hours of continuous tracking with some occasional 4x fast movement, it's still going strong. Actually, if it's really rated at 2600mmAh, we should have enough juice for 8 hours of continuous usage. Frankly, 5 hours are more than enough for me for I rarely observe more than 5 hours. I simply do not have the luxury to do so.
For day time events like solar eclipse or the upcoming Venus transit, the solar panel should help to extend the usable tracking time. For night time events like a total lunar eclipse, it's still pretty enough for I will only do half of the whole process.
For very long observation session, I've a backup solution. It consists of 8AA re-chargeables (Eneloop) which deliver 9.6V with 1900mmAh guarantee capacity, so it translates to another 11+ hours, more than a night worth of observation.