Re: UPS as backup power

[K9AUB at aol_com]

Well, this EE not-much-of-a-guru will make an attempt to explain it.  Much of this phenomena can be explained by the type and design of the lead/acid battery used in the UPS.  Batteries are designed for different purposes and different types of loads by varying the thickness and number of their plates, and thus the area of lead presented to the electrolyte.  For example, an automobile battery is designed to deliver a very large amount of current over a very short period of time, to handle the load presented by an automobile starter.  Thus, due to the design of its plates, a car battery can dump a lot of current very quickly.  A boat trolling motor would present a different type of load, where a small amount of current needs to be supplied over a long period of time.  If you attempted to use a marine battery (type 27, for example) to start a car, you would find it is very inefficient because a marine battery cannot dump huge amounts of current very quickly for any length of t!
ime.  If the car did not start immediately, the marine battery would quickly taper off in delivering its load and might fail to start the car, even though it had lots of charge left in the battery.  It isn't a question of how MUCH power was in the battery; it's a question of how quickly it can be delivered.  Similarly, if you used a car battery on a trolling motor, you would find you got significantly less amounts of current delivered because a slow draw is not the purpose for which a car battery was designed, and you might get perhaps as little as 20% of the charge out of the battery before it failed.  The batteries in a UPS have similar varying discharge curves, according to their design.  A UPS is designed to deliver X-amount of power over a short period of time (e.g., 350w for 10 or 12 minutes), to enable the computer user to save his files at time of power failure, and then shut down.  If you pull a certain amount of power from the battery very quickly (high current dra!
w), the battery is not capable of delivering ALL of the power!
 it might hold, and its useful life would be shortened.  However, if you draw a small load over an extended period of time, that same battery is capable of delivering much more of its charge.  This would explain why a computer UPS used on an aquarium would deliver varying amounts of power according to how many devices you are running on it.  And thus, your experience was that you got only 2 minutes of service at a 200w load, but you got 21 minutes with only a 50w load.  Had you continued your experiment, you might have found that the same UPS delivered, say, perhaps 80 minutes at a 25w load.  If you were to upsize the UPS from 350w to 700w, you would find that you receive much more than the expected 2x2=4 minutes of service, but might instead get 20 minues of power, and would probably get an hour or two's service at 50w load.  
By the way, if you can access the 12v leads internally feeding the UPS, you can extend those out to something like a type 27 marine battery, and get days and days of service driving your 50w aquarium load.  A freshly charged pair of 6v golf cart batteries in series (to deliver 12v) would drive your aquarium for a week or more.  If you had a solar panel charging them at about twice the rate of discharge, it's possible that you would never have need of any other power source.
> load=200W, runtime=2 min> load=100W, runtime=8 min> load=50W, runtime=21 min> > So I thought a filter at 20-25W would only run 60min> max.  But you have gotten 12+ hrs!  This is very good> news for some of us.> > Maybe an EE guru can explain why.  My guess is, it has> to do with 1) Wattage rating on many equipment are> peak values so the actual consumption varies & is much> lower, 2) the UPS has to invert DC back to AC so maybe> the higher the load the less efficient it gets.