I'm just wondering if anyone makes a computer UPS that runs on Lithium Ion batteries? I'm having a hard time finding that information. I'm tempted to convert my UPS to LiFePO4 but it looks like it might be too much work. I'm just tired of buying lead batteries every few years and I hope that Lithium batteries would last longer in a UPS.
Li would not be suitable for a UPS because they do not work once discharged. NiCad and Ni Mh can both be discharged quite deeply without damage, Li ion cannot.
Li Ion battery does not offer any advantage over Pb Acid batteries in terms of shelve live which the UPS is mainly siting there most of the time. Li Ion typically have a shelve life of 3-5 years, so you will be replacing the battery about the same time period. But Li Ion is 5-10 time expensive and thats why no body is selling one except for special application where longer backup time is needed. How to prolong lithium-based batteries
Well just as an experiment I bought a LiFePO4 drop in replacement for a 7Ah lead battery. Here is the link. LiFePO4 26650 Battery: 12.8V 6.8AH (87 Wh, 25A rate) built in PCB , Replace SLA with 10 times longer life & 61.77% lighter weight This battery has over voltage protection, under voltage protection, and over current protection. So I'm pretty much covered from damaging the battery. I ran some tests on it and the UPS will charge it to about 75%. A 115 watt load will run for about 33 minutes. A 310 watt load will run for about 8 minutes but the terminals will get very hot. About 185F. 310 watts is more than I would normally put on it so it looks like this battery will work just fine. The only issue now is how long will it last in terms of years? Hopefully it will be a long time before I find out.
Please post updates when you have any, there are a lot of people interested in this topic, but unable to experiment at the current prices. Thanks, Aaron
There is really not much to say about my upgrade to LiFePO4. I bought 4 more of the 6.8Ah batteries for my rack mount UPS and a 40Ah battery for my large UPS. So far they are working fine. I wont know how well they will work until I get at least 4 years of use out of them. Thats about how long the lead batteries last before they bloat and outgas. If my Hymotion battery in my Prius is any indication of how well these batteries in the UPS will do I think the batteries will out last the UPS. The only thing I'm concerned about is the BMS board on these batteries. They use shunts to balance the batteries and Jack Rickard has convinced me that these BMS's are very bad. But since the UPS's are designed for lead that means the LiFePO4 batteries are being under charged. So I think I might be OK. Since it will be about a year since I bought these when I get some time I will try to get some measurements of the cell voltages and capacity.
I was under the impression that lithium's advantage is lighter weight compared to lead. In a car you can fit in more kWh with lithium than with lead. But on a shelf, weight is not an issue.
The biggest issue with lead/acid batteries and shelf life is maintaining a proper float charge. Most small batteries are "maintenance free" which really means that over time, they will run out of water, exposing the plates to air, killing the batteries. Most small trickle chargers are not very well suited (in spite of claims to the contrary) for long term storage of FLA batteries. In the off grid solar world, we use four stage chargers, that modify currents and voltages to maximize battery life. Bulk, absorb, float and equalize stages, each with different voltage/current requirements. Matching the charge characteristics with battery size and chemistry. Icarus
Sorry I left out the links to the website that I got the idea from, I meant to add them at the end... A123RC Battery
OK op, it's been 5 years and I joined just to find out how your project went. I'm dying to find out how the Li batteries held up inside your UPS - care to share?
The lithium batteries still work in my UPS but I dont know if the capacity had dropped. Two other UPS's I had batteries in either killed some of the cells or the BMS killed the cells. From the looks of one dissected battery it was the fault of the BMS. The BMS also causes issues with my APC rack mount UPS's. When the voltage gets to less than 10v the BMS will cut the voltage to the terminals. Only a 12v source can reactivate the BMS. Since the UPS can operate just below 10v this means if I dont shut off the UPS before it drains the battery I have to open the UPS up and apply 12v the batteries to turn them on. The UPS will not turn on if it does not have a 12v from the battery and it wont "kick start" it either. Only option is to open it up and apply 12v to the battery. So as far as longevity goes its looking good. As far as BMS goes dont bother. A good bottom balanced LiFePO4 battery pack without a BMS is just fine as long as the UPS does its own cutoff voltage of 10v which most if not all should do. And the UPS should not over charge lithium batteries either since SLA maximum charge voltage is less than a 4 cell LiFePO4 pack. I plan on redoing my UPS's with no BMS sometime to see how well that will work but I have too many things to work on at the moment.
Awesome, thanks for the reply. There's actually a manufacturer that makes a UPS with an Li battery built in (APC, I believe), but it's in a rather small capacity. Good to know for my bigger project though.
A lithium-type battery should always have low-voltage protection, particularly in a UPS application. While deep discharges are not the lithium battery's strong suit, remaining at a constant level of charge for long periods of time should contribute to fairly good battery longevity. An added bonus is the much greater power-to-weight ratio of lithium cells.
LiFePO4 batteries can be discharged to 0v just fine. They may loose some capacity but they wont catch fire. Reverse voltage can be an issue but thats where bottom balancing comes in and a controller that wont let a pack voltage drop below 2v per cell. Over voltage is what cause LiFePO4 cells to catch fire and its not so much a fire but more like a lot of heat and out gassing. Still can be a fire hazard though. Active top balancing BMS's should be avoided.
