Drone Equipment Forecasting Battery Life-Cycles

Drone Equipment: Forecasting Battery Life-Cycles

Can you predict when your equipment will fail? As a regular part of our maintenance program, we maintain a close eye on our batteries, one of the critical safety of flight elements of most unmanned aircraft.  The total cycles (a single charge and discharge) have proven to be the most useful single metric for tracking wear and tear, tracking these in DMS has proven invaluable.  HAZON, at our core is an end-to-end drone service provider.  Access to relevant data on the performance and lifecycle of the equipment that enables your work is absolutely critical to any business.

We recently took an opportunity to do a study on the useful life of a sampling of some of our batteries.  In this article, we will consider TB48D batteries in use on our DJI Matrice 100 fleet.  By way of background, we have over four dozen TB48D batteries in service; this study examines a random sampling of those beyond their 50% life expectancy.

The ten batteries we investigated in this study were flown exclusively on Matrice 100s with a Zenmuse Z30 camera installed and configured for a single TB48D battery. The usage period was about six months. Our configuration, flight operations tempo and business demands result in a harder than preferred life of our batteries.

In consideration of this study, it is important to note that HAZON Solutions specializes in providing drone based services and training and not on maximizing hardware life-cycle. Often times the needs of our business’s operations requires a harder than preferred usage of our equipment.

Data Set

SET 1:

Batteries that indicated 25% remaining

SET 2:

 Batteries that indicated 37% Remaining

TB48D #1, Current Cycles:135 TB48D #6, Current Cycles:110
TB48D #2, Current Cycles:125 TB48D #7, Current Cycles:108
TB48D #3, Current Cycles:132 TB48D #8, Current Cycles:108
TB48D #4, Current Cycles:127 TB48D #9, Current Cycles:120
TB48D #5, Current Cycles:126 TB48D #10, Current Cycles:108
 

*Calculations used for SET 1:

  • SUM:(TB48D#1-5 Current Cycles) =129 Avg. Used Cycles
  • 129 Avg. Used Cycles / 75% (*Battery life used*) = 172 Lifetime Cycles
  • 172 Lifetime Cycles X 25% (*Remaining Life*) = 43 Avg. Estimated Remaining Cycles
  • 172 Lifetime Cycles – 129 Avg. Used Cycles = 43 Avg. Estimated Remaining Cycles
 

*Calculations used for SET 2:

  • SUM:(TB48D#1-5 Current Cycles) =111 Avg. Used Cycles
  • 111 Avg Used Cycles / 63% *Battery life used* = 176 Lifetime Cycles
  • 176 Lifetime Cycles X 37% *Remaining Life* = 65 Avg. Estimated Remaining Cycles
  • 176 Lifetime Cycles – 111 Avg Used Cycles = 65 Avg. Estimated Remaining Cycles
Result: 172 Lifetime Cycles on a TB48D Result: 176 Lifetime Cycles on a TB48D
*Comparative Calculations

(172 *SET 1: Lifetime Cycles* + 176 *SET 2: Lifetime Cycles*) / 2 = 174 Avg. Total Cycles Before End of Battery Life (0%)

Final Result
TB48D Batteries will last approximately 174 Charging Cycles
before indicating 0% Battery Life Remaining

Contribution Factors

Multiple factors influenced the data set that we developed. One equipment related factor was that these batteries were charged using a DJI Hex Charger.  Hex chargers charge at a higher amperage than a standard 100-Watt charger, this could potentially reduce lifetime of the battery. From an environmental standpoint, the batteries were used in temperatures ranging from 20°F to 110°F. This factor should also impact battery life. Two additional factors to note; the operational tempo required the batteries to be cycled 2-3 times per day, and the batteries were often flown from a full charge down to 20-15% of battery power remaining.  As previously noted, this is not a study in how to extend the life of a battery, but rather a practical observation of the life cycle of a battery in a high-volume flight operation.

ObservationsDrone battery life-cycles

For the ten test case batteries, the endurance of the batteries remained within reasonable limits (approximately 15-20 minutes of flight time dependent on conditions) for approximately the first 100 cycles. Approaching 100 cycles of use the crews began noticing a degradation in battery performance. Crew members also reported reduced flight time and abnormal discharge rates. As crews approached 25% remaining battery life (as reported by the built in smart battery features), the endurance of the batteries dropped dramatically to the point where the batteries had to be pulled from service due to significant drops in voltage during flight.

From these observations, we determined that lithium polymer batteries follow the traditional wisdom of non-linear functions up to the 100-cycle point. The chart below shows that endurance remains relatively unchanged for approximately the first 100 cycles of usage. After 100 cycles the tolerance of the battery begins to diminish at an increasingly rapidly rate until the battery is no longer usable.

Conclusions

The overall result of our study affirmed the need for three critical operational functions:

  • A Remove from Service protocol
  • A battery life-cycle tracking system
  • A life cycle forecasting system

This study allowed HAZON to easily update our existing Remove from Service protocol.  We identified that TB48D batteries have a lifetime of 174 cycles through our usage cycle and that at approximately 100 cycles the battery performance began to decrease noticeably. Using these two data points, and applying an appropriate safety margin, our maintenance team decided that batteries should not be used beyond 90% of their total life. Based on the findings in this study we have amended our Standard Operating Procedure (SOP) to dictate that once a TB48D battery reaches 155 cycles that the battery will be immediately removed from operational service and relegated to low-risk training or testing.

We also identified a need for companies to track battery life cycles outside of the internal tracking of the batteries themselves. At HAZON, we rely on the HAZON Drone Management System (DMS) – a fleet management software for enterprise drone operations that provides operators and managers a single platform for all project-based planning and tracking of your fleet operations. Most importantly, HAZON DMS allows users to capitalize on their equipment investments while maximizing the efficiency of operations.

Resource management is a critical component of any business. We use the data provided by DMS to help us forecast when batteries need to be replaced and provide a cost estimate for that replacement within a monthly, annual, or biennial budget cycle. Having this type of information available to company leadership allows for proactive decision making on purchasing decisions that affect cash flow and enables them to define cost to customers more clearly. For more information on HAZON or HAZON DMS contact Ed Hine at ed@hazonsolutions.com.  To get in touch with the HAZON R&D team contact Garrett Scott at garrett@hazonsolutions.com.

Written by Garrett Scott, Training and Technology Division Manager & Todd Boward, PhD, Assistant Training and Technology Division Manager


 

Tags: , , , , , , , , , , , ,

Leave a Reply

Your email address will not be published. Required fields are marked *