A Lithium-Ion Cell Low-Earth-Orbit Verification Test Program is being conducted by the NASA Glenn Research Center to assess the performance of lithium-ion (Li-ion) cells over a wide range of low-Earth-orbit (LEO) conditions. The data generated will be used to build an empirical model for Li-ion batteries. The goal of the modeling will be to develop a tool to predict the performance and cycle life of Li-ion batteries operating at a specified set of mission conditions. Using this tool, mission planners will be able to design operation points of the battery system while factoring in mission requirements and the expected life and performance of the batteries.
Test conditions were selected via a statistical design of experiments to span a range of feasible operational conditions for LEO aerospace applications. The variables under evaluation are temperature, depth-of-discharge (DOD), and end-of-charge voltage (EOCV). Four cells from each vendor are being LEO tested at each of 10 sets of test conditions. The LEO profile consists of a 90-min cycle: 55 min of charging, and 35 min of discharging. The required discharge (and charge) currents are calculated on the basis of the actual capacity of the cells.
The table shows the test matrix. As reflected in the table, test conditions for individual cells may vary slightly from the baseline test matrix depending upon the cell manufacturer’s recommended operating conditions. In such instances, deviation from the baseline matrix to another controlled set of operating conditions will not perturb the statistical relationships because the alternate conditions fall within parameter ranges that are bounded by the baseline matrix.
|
Temperature, °C |
End-of-charge voltage EOCV, V |
Depth of discharge, DOD, percent |
|---|---|---|
| 30 | 4.05c/4.0d | 20 |
| 30 | 3.85 | 20 |
| 10 | 3.85 | 20 |
| 30 | 3.95 | 30 |
| 20 | 3.95 | 20 |
| 10 | 3.85 | 40a/35b |
| 20 | 3.85 | 30 |
| 30 | 3.85 | 40a/35b |
| 20 | 4.05c/4.0d | 40a/35b |
| 10 | 4.05c/4.0d | 30 |
| aAll vendors except Saft. | ||
| bSaft. | ||
| cAll vendors except MSA. | ||
| dMSA. | ||
Currently, cells from four cell manufacturers are undergoing life-cycle tests. The program has the flexibility to accommodate additional test articles as it progresses, so the tests are at various stages. Life cycling on the first sets of cells began in September 2004. These cells consisted of Saft 40-A-hr cells and Lithion 30-A-hr cells. In the past year, the test program expanded to include the evaluation of Mine Safety Appliances (MSA) 50-A-hr cells and ABSL (Power Solutions, Inc.) battery modules. The ABSL battery modules consist of commercial Sony hard carbon 18650 lithium-ion cells configured in series and parallel combinations to create nominal 14.4-V, 6-A-hr packs (four cells in series and four cells in parallel). The modules contain individual cell voltage monitors for data collection purposes, but charging and discharging is controlled on the module level. One module (four strings) is being tested at each of the 10 conditions. The figure shows the results of capacity characterization on cells from each vendor.
Results of capacity characterization on ABSL modules and Lithion, MSA, and Saft cells at different temperatures.
In fiscal year 2006, Saft and Lithion cells achieved over 10,000 cycles each, equivalent to about 20 months in LEO. Preliminary results were reported in references 1 and 2. Saft cells generally have a higher end-of-discharge voltage (EODV) than Lithion cells operating at the same conditions. Saft cells also have a lower EODV dispersion among cells operating at the same conditions. For both vendors, when cells are operating at the highest DOD, the EODV of the cells varies more than the EODV for the cells operating at the lower DODs. Also, as a cell’s EODV gets lower and lower, operational capacity checks tend to have a more profound reconditioning effect on the cell. Upon return to LEO cycling, the cell tends to have a much higher EODV in comparison to the EODV before the capacity test.
Also in fiscal year 2006, ABSL modules accumulated approximately 3000 cycles. All 16 cells within each module are performing consistently and have shown very little EODV degradation. MSA cells will begin life cycling in October 2006. Actual capacity determination, open-circuit voltage stand, and capacity characterization at different temperatures have been performed on these cells thus far.
The life prediction and performance model for Li-ion cells in LEO will be built by analyzing the data statistically and performing regression analysis. Cells must be cycled to failure so that differences in performance trends that occur at different stages in the life of the cell can be observed and accurately modeled.
For this test program, a cell is considered to have failed once its discharge voltage reaches 3.0 V. By the end of fiscal year 2006, six cells had completed their life cycling by cycling to failure. All of these failures occurred in cells operating at the lowest EOCV and at the highest DOD for each vendor. Five of the six failures were in cells operating at 10 °C. The combination of the lowest EOCV, the highest DOD, and the lowest operating temperature (3.85-V EOCV, 35- or 40-percent DOD, and 10 °C) is considered to be the most stressful set of conditions in the test matrix, therefore, it was not unexpected that these cells would fail the earliest. The sixth failed cell was operating at 30 °C. Five of the cells were Yardney cells, including the cell cycling at 30 °C. The sixth cell was a Saft cell. These early failures will begin to provide input for the life prediction and performance modeling.
Cell testing is being performed at the Naval Surface Warfare Center in Crane, Indiana, through an Interagency Agreement. Statistical analysis is being performed by Harold S. Haller & Co. under contract. Additional background on the program, results of the initial characterization, and preliminary life-cycling results of the Lithion and Saft cells are discussed in more detail in references 1 and 2.
Find out more about the research of Glenn’s Electrochemistry Branch: http://www.grc.nasa.gov/WWW/Electrochemistry/
Glenn contact: Concha M. Reid, 216-433-8943, Concha.M.Reid@nasa.govLast updated: December 14, 2007
Responsible NASA Official:
Gynelle.C.Steele@nasa.gov
216-433-8258
Point of contact for NASA Glenn's Research & Technology reports:
Cynthia.L.Dreibelbis@nasa.gov
216-433-2912
SGT, Inc.
Web page curator:
Nancy.L.Obryan@nasa.gov
216-433-5793
Wyle Information Systems, LLC
