The APC might not charge the LiFePO4 batteries to a high enough voltage to trigger the BMS balancer, and also probably won't float the LiFePO4 batteries at the proper voltage. You also may have serious issues bootstrapping the LiFePO4 batteries in that series configuration if one of them disconnects on low voltage. Generally speaking, BMSs start balancing at 3.45V/cell (x8 = 27.6V). Charge target should be 3.50V-3.55V/cell (28.0V to 28.4V), and the float needs to be 3.35V-3.375V/cell (26.8V to 27.0V). If the APC doesn't charge to a high enough voltage the LiFePO4 batteries will slowly go out of balance and eventually stop working properly in 6-12 months. If the APC doesn't float the battery properly, the LiFePO4 batteries will remain stressed 100% of the time which is not good for them at all, shortening its cycle life. -Matt
This one charges to 27.5 3.43/cell. Time will tell. I’ve cycled it a few more times since this with no issues yet. I have another video coming with a 24V jar which I think is a better solution. I’m of the “one BMS for the voltage” mindset.
Recycled power haha - Thanks for doing the test. It's all about max amp draw with these LifePO4 batteries - UPS max is probably around 45A, and these batteries likely have 50A BMSs.
That's all well and dandy but what I really want to know is what happens when the UPS finishes charging the batteries and the BMS shuts off and the UPS no longer sees voltage or any kind of load. I doubt it'll be happy and most likely will go into an emergency mode or a bad battery alarm so let us know how that works out.
This UPS charges to 27.5V. The combined battery BMSs don’t cut off until 28.8. It’s also designed for hot swap of batteries, you can remove them and reconnect with no issue as long as the power does not go out in the meantime.
@@DougKremer Maybe I'll try to find a UPS that has a port for a external battery. My cyber power 1500 W internal battery UPS will not work with LiFePO4. Seems like it's looking for that .25 to .5amp load a lead acid battery has or something like that because as soon as the current falls off after charging it throws an error message that the battery is bad. I wonder what would happen if I use a dummy load resistor to hold a little bit of charging current. Penny for your thoughts.
Think of switchable "perfect" diodes setup in opposite directions for charge and discharge. The BMS Charge mosfets turn off, but the Discharge mosfets stay on. The battery terminals will float at whatever the UPS float voltage may be, but if that's over the BMS charge cutoff, then the cells settle to their resting voltage with discharge mosfets on and the battery will pickup the load instantly when a discharge state drops the terminal voltage to match their level. Always a possibility a given BMS and UPS combo won't behave well together, but the ones I've used simply stop accepting a charge and appear full while ready to discharge. They don't disconnect entirely like a battery switch would.
I’m curious what the float voltage of your UPS is when connected to lead cells, and how balanced your LFP jar is. I spent a day or two slowly dialing up the voltage on these with my adjustable power supply so the internal BMS could balance. I eventually got them up to 14.3V without the BMS shutting them off on a high cell, then left them connected at that for a couple more. Could be your battery isn’t balanced enough to get to the float voltage the UPS is trying to achieve, so it jumps off, causing the UPS alarm. I have another video coming soon with this UPS and a different battery where I show the power supply a little more, but did not go into the balance process and theory. That might make for a good video. It’d be ideal if I had one with Bluetooth or something so we can see what is happening on the inside while it happens instead of me just waving my hands around trying to describe 🤣.
What kind of UPS are you using? Touching energized parts is generally not a good idea, which is why they are designed with internal batteries, not exposed as I am testing here.
The APC might not charge the LiFePO4 batteries to a high enough voltage to trigger the BMS balancer, and also probably won't float the LiFePO4 batteries at the proper voltage. You also may have serious issues bootstrapping the LiFePO4 batteries in that series configuration if one of them disconnects on low voltage.
Generally speaking, BMSs start balancing at 3.45V/cell (x8 = 27.6V). Charge target should be 3.50V-3.55V/cell (28.0V to 28.4V), and the float needs to be 3.35V-3.375V/cell (26.8V to 27.0V).
If the APC doesn't charge to a high enough voltage the LiFePO4 batteries will slowly go out of balance and eventually stop working properly in 6-12 months. If the APC doesn't float the battery properly, the LiFePO4 batteries will remain stressed 100% of the time which is not good for them at all, shortening its cycle life.
-Matt
This one charges to 27.5 3.43/cell. Time will tell. I’ve cycled it a few more times since this with no issues yet. I have another video coming with a 24V jar which I think is a better solution. I’m of the “one BMS for the voltage” mindset.
@@junkerzn7312 Not in series, in parallel.
Great video Doug!
I like your workbench 😉
Thanks! Multitasker! It will also cook a bunch of mean shashburgers.
I did the same with one 100Ah battery. Those old UPS are very good. New APC ones are bad (catch fire).
Recycled power haha - Thanks for doing the test. It's all about max amp draw with these LifePO4 batteries - UPS max is probably around 45A, and these batteries likely have 50A BMSs.
Great point!
How much time elapsed till shutoff?
I think +/- two hours. I was after power delivered, not time.
That's all well and dandy but what I really want to know is what happens when the UPS finishes charging the batteries and the BMS shuts off and the UPS no longer sees voltage or any kind of load. I doubt it'll be happy and most likely will go into an emergency mode or a bad battery alarm so let us know how that works out.
This UPS charges to 27.5V. The combined battery BMSs don’t cut off until 28.8. It’s also designed for hot swap of batteries, you can remove them and reconnect with no issue as long as the power does not go out in the meantime.
@@DougKremer Maybe I'll try to find a UPS that has a port for a external battery. My cyber power 1500 W internal battery UPS will not work with LiFePO4. Seems like it's looking for that .25 to .5amp load a lead acid battery has or something like that because as soon as the current falls off after charging it throws an error message that the battery is bad. I wonder what would happen if I use a dummy load resistor to hold a little bit of charging current. Penny for your thoughts.
Think of switchable "perfect" diodes setup in opposite directions for charge and discharge. The BMS Charge mosfets turn off, but the Discharge mosfets stay on. The battery terminals will float at whatever the UPS float voltage may be, but if that's over the BMS charge cutoff, then the cells settle to their resting voltage with discharge mosfets on and the battery will pickup the load instantly when a discharge state drops the terminal voltage to match their level.
Always a possibility a given BMS and UPS combo won't behave well together, but the ones I've used simply stop accepting a charge and appear full while ready to discharge. They don't disconnect entirely like a battery switch would.
I’m curious what the float voltage of your UPS is when connected to lead cells, and how balanced your LFP jar is. I spent a day or two slowly dialing up the voltage on these with my adjustable power supply so the internal BMS could balance. I eventually got them up to 14.3V without the BMS shutting them off on a high cell, then left them connected at that for a couple more. Could be your battery isn’t balanced enough to get to the float voltage the UPS is trying to achieve, so it jumps off, causing the UPS alarm. I have another video coming soon with this UPS and a different battery where I show the power supply a little more, but did not go into the balance process and theory. That might make for a good video. It’d be ideal if I had one with Bluetooth or something so we can see what is happening on the inside while it happens instead of me just waving my hands around trying to describe 🤣.
I tried it but why I'm getting shock when i touch the battery terminal while UPS is plugged in
What kind of UPS are you using? Touching energized parts is generally not a good idea, which is why they are designed with internal batteries, not exposed as I am testing here.