WattFlyer RC Electric Flight Forums - Discuss radio control eflight - View Single Post - FMA Cellpro PowerLab 8 regenerative discharge power source?
Old 04-15-2011, 06:21 PM
Tim Marks
President FMA, Inc.
Join Date: Dec 2009
Posts: 112

Originally Posted by miernik View Post
What are the limitations on the power source used in case of using the regenerative discharge function?

Does it absolutely have to be a lead-acid battery, or can I have a chain of 8 LiFePo4 batteries as the power source and it will do fine? Eight LiFePO4 have the same charging voltage (28.8V) as two lead-acid batteries, so it should be possible to do? There will be no BMS, but if I make the source battery larger then the one on the other side of the charger, there should be no risk of overcharging, no?

I read that the Cellpro auto detects the power source, and will not allow me to use regenerative discharge if its not a lead-acid battery. Will it refuse a LiFePO4 instead of a lead-acid?

And what if the power source would be a battery and another 28.8V DC power supply connected to it in parallel, with a small limited current (say C/20) to replace the energy lost in charging inefficiency? Is that possible to do, or will the PowerLab refuse to do regenerative discharge in such setup?

I want to do long-term cycling of 10000 cycles at 4C of A123 and Headway cells. Headways are 10Ah per cell, and I want to charge/discharge them at 4C, which is a full 40A. It'd be problematic to get a large-enough lead-acid battery to accept so much charge, buying a 400Ah lead acid... no way! Much easier and cheaper to get another LiFePO4 and flush the energy back and forth between two LiFePO4 packs.

Although I didn't intend for our chief engineeer to spend SO MUCH time on this answer, he did. So I'll just post his reply back without any edits. Here you go:

The setup is pretty simple. A power supply will be required, because of the 10,000 cycle duration of the test. A battery needs to be connected to the power supply to allow the 40 amp discharge current.

Here is the setup.

1) Since the output battery is 10Ah, the supply battery should be 30Ah or higher. This will leave plenty of room to hold the discharge.
2) Connect the power supply in parallel with the supply battery.
3) Set the power supply voltage to 50% state of charge in the supply battery.
For example: The nominal LiFePO4 voltage is 3.3V per cell. This is the 50% SOC voltage.
A 7S supply battery should have the power supply set at 23.1V
4) The supply battery should be drained to 50% SOC before the test begins.
5) Set the data log rate to 120 sec. This will support data files that are over a year long. I haven't tested 1 year data files yet. I did test 1 month long files. I think you should be good to 65,535 cycles, but I am not sure.
6) Of course, you are going to need to dedicate a PC to this task. The PC should be connected to a UPS for power outages, which will surly happen. The Charger should be ok if power is lost because of the supply battery.
7) Turn off windows automatic updates and virus protection updates so that the PC can run for a year without rebooting. It might be a good idea to disconnect from the internet to guarantee that the PC will never try to re-boot.
8) Periodically save the gathered data to the hard drive using the File SaveGraph option in the main screen.

You will find that cycle life capacity decay is quite linear from 100% to 80%. Then cycle capacity drops off a cliff quickly. PL8 can actually predict cycle life based on 20 cycles by comparing the capacity of the 2nd cycle to the last cycle. However, PL8 will not predict cycle life beyond 500 cycles. If you find that after 100 cycles the battery has decayed 1%, the expected cycle life will be 2000 cycles to 80% capacity. If you find that after 100 cycles the battery has decayed 5%, the expected cycle life is 400 cycles.

9) It is very important to do the testing in a temperature controlled room. Cold temperatures really destroy lithium cycle life. Also, if the room isn't temperature controlled, you will see the capacity change from nighttime to daytime and back again. This isn't a big deal if you are running an end of life test. However, if you are trying to predict cycle life based on a linear decay, you will need account for the daily temperature changes and only compare cycles that were done at the same temperature.
10) Testing should be stopped at 80% capacity. This is the industry standard for all type of batteries, because the growth rate of the dendrites in the cell is cubic and very exponential. You will see what I mean if you run to end of life.

I believe the genuine A123 cells will go 10,000 cycles at 4C charge/discharge and room temperature. I have no idea what the Chinese copy cells will do.

I would be interested in seeing your data. I have found that what the cell manufacture says and what really happens are two different things.
Attached Thumbnails Click image for larger version

Name:	PL8 CCS Cycle Test.jpg
Views:	354
Size:	149.8 KB
ID:	146620  
Tim Marks is offline  
Page generated in 0.04962 seconds with 12 queries