[RE-wrenches] LiFePo4

Starlight Solar Power Systems larry at starlightsolar.com
Tue Sep 16 10:13:35 PDT 2014


Jarmo,

I must say I disagree with most of what you have said. In the EV DIY industry, there are a growing number of conversions using LFP battery banks without any BMS or EMS at all. These pioneers are risking there multi-thousand banks because they have learned something about these cells. They will stay in balance and perform perfectly as long as they prevent over and under voltage or charging a frozen battery with high current. 

My reply numbers correlate to yours:
I agree for most customers a BMS is necessary.
Disagree: The battery systems (with BMS) that I sell do not regulate cell voltages during charge or discharge. They only adjust voltage after the cells reach 3.55 Vpc. However, at that 3.55 Vpc, I have terminated charge and entered a float/maintenance charge. On my personal battery bank, 24 cells, I have charged and discharged daily since May. My cell voltages still only vary by 0.02 volts per cell.
Disagree: There is no temperature adjustment for voltage during charge or discharge. All temperature compensation circuits are removed from the RE equipment per manufacturers specifications.
Disagree: We provide a continuous float voltage per manufacturer specifications. A float voltage can keep the battery at any SoC. I program for 90%.
Our CPU provides protection for over current BUT that will likely never happen because the cells can be charged and discharged up to 10C and continuously at 3C!
Disagree: No need to limit inverter current as the battery can produce 10C.
Disagree: Ditto above.
Disagree: No need to limit inverter charge current because of the 10C capability.
I agree that for most consumers a BMS is required. So far we install the BMS provided by the manufacturer on every system. I will be installing systems without BMS in the future but openly for select customers.

I have not found that LFP batteries are “amazingly sensitive”. I have discharged cells past 100% DoD, in fact as low as 0.35 Vpc, and recharged without any harm. I have started very complex loads like air conditioners with repeatable success from small battery banks. While I do not recommend this, I believe there is a robustness with modern LFP cells. 

The benefits of LFP batteries are many. Each year I recycle between 30,000 and 40,000 pounds of lead acid batteries. MOST of them are damaged by the consumer deficit charging them. I consider this to be a very unfavorable aspect of lead acid when compared to LFP batteries that never need to be fully charged or equalized or watered.

One thing that will damage LFP batteries is if a cell in a series string is ever discharged below 0 volts, it will take on reverse polarity and is usually not recoverable. But, you can simple replace it and balance it with the other cells. 

I welcome your comments.

Respectfully yours,

Larry Crutcher


On Sep 15, 2014, at 8:08 PM, Jarmo.Venalainen at schneider-electric.com wrote:

Hi: 

We've done some testing and operation of Lion battery banks in the 10 kWh range with our inverter chargers and so far the most important findings are: 

1. The battery pack must have its own battery management system to: 
a. regulate the individual cell voltages during charge and discharge 
b. balance the cell voltages during charge and discharge 
c. adjust cell charging voltages due to temperature variations 
d. not provide a continuous float voltage to the battery bank 
d. provide a built in safety cutoff disconnect, which turns off the current if for any reason either the charge or discharge current is too high. 

A series string of Lion cells, without the battery management system functions above is very likely to result in damaged cells or worse.  The most likely mechanism which ultimately causes damage is individual cell temperature or cell to cell voltage imbalance. 

2. The inverter system must be designed so that the inverter never draws so much current that the current limit circuit in the battery bank is triggered. 

3. It may be necessary to adjust the current limit circuit as they typically respond within milli-seconds and may be triggered on simple power up as the inverter cap bank charges up. 

4. When working with Lion, the inverter/charger should be configured so that it is a simple current limited voltage source when in charge mode.  The built in battery management system should take care of the required charge cycle operation. 

5. Whether the pack is LiFePO4, Lion polymer or other does not appear to be as important as a having a reputable brand and more importantly a well designed built in battery management system. 

The short version is that lead acid cells are amazingly forgiving with regards charging and discharging voltages, currents and temperatures.  Lion cells on the other hand are amazingly sensitive and going out of bounds on any of these can and will cause irreparable damage.   

Given the sensitivity, it doesn't make sense to take on responsibility for Lion battery management. 

Management of that sensitivity and responsibility is best left to the battery bank manufacturer. 

JARMO








-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.re-wrenches.org/pipermail/re-wrenches-re-wrenches.org/attachments/20140916/4afb5a23/attachment-0002.html>


More information about the RE-wrenches mailing list