Principles for Evaluating the BatteryInformer
The BatteryInformer® is designed to measure the decay of a battery’s state of health over the lifetime of the battery. While long-term observation of actual BatteryInformer® performance in a real world application is the ideal method of evaluation, it is obviously impractical to spend 3+ years observing and testing. Here we will describe a simple methodology and 5-step process for evaluating the primary functions of the BatteryInformer® using comparative tools and procedures to simulate the battery aging process. Additionally, a BatteryInformer® test protocol is included for in depth larger scale laboratory testing in Appendix A. Independent test laboratory results are available upon request from Mega-Power, Inc.
There are three (3) characteristics of operation that can be evaluated on the BatteryInformer® :
- Accuracy, precision and overall reliability of the measured ohmic values;
- Distinguishing the condition of a battery on a scale of ‘good to bad’; and
- Determining if the alarm threshold for indicating end-of-life is appropriate for the application.
A suggested approach for each evaluation is discussed below:
The underlying ability of the device to detect health decay is the inherent capability of the ohmic test. If the device can measure Mho (conductance) in a precise, accurate and reliable way, then it is relatively simple to compare measurements to determine changes in ohmic health.
Connect the sensor to a charged battery of appropriate size. After the 30-second countdown delay, document the displayed conductance value (Mho). The conductance value displayed on the sensor represents a moving average of the past 16 periodic measurements. From simple observation of the display every minute, it can be observed that the measurements are consistent (unless the battery is being charged or discharged actively, or experiences significant temperature changes). Repeatability (precision) of the conductance value is expected because a fixed connection to the battery eliminates ‘tool error’ that results from variability in the resistance of the connection between test attempts using a hand held tester. The test battery need not be ‘new’ to use the above method to evaluate precision.
Obtain a new and fully charged battery for which a manufacturer’s published internal resistance is available, and attach the sensor. Note the published internal resistance specification must be one derived by the manufacturer, not through benchmark testing using a 3rd party handheld ohmic tester. Observe the displayed conductance value of the battery (in Mho) on the BatteryInformer® after a few minutes of operation. Now compare this conductance value to the manufacturers published resistance specification for the battery as follows: First, convert the displayed conductance value to internal resistance, using the formula Resistance (in mOhm) = 1000*(1/X Mho) where X is the displayed conductance value. Note: It is normal for the resistance measured by the sensor to be somewhat lower than the published value (i.e. sensor value = 4.3mohm, spec value = 4.5mohm). This favorable difference is the result of manufacturers ensuring by design that batteries exceed the performance specifications when new.
Note that conductance measurements from the sensor may also be compared to conductance measurements from other ohmic test devices such as hand-held testers. However, it must be noted that ohmic measurements are a function of the frequency of the signal used to make the measurement, and therefore many ohmic methods (particularly AC conductance methods) used in handheld testers yield ohmic values that are sometimes not comparable to battery manufacturer’s published internal resistance values. Thus, comparisons between the Smart-500 (as well as Manufacturer’s published specifications) and the results of some hand held testers could be problematic.
The BatteryInformer learns (stores) the maximum conductance value of the battery and then compares subsequent conductance values to the maximum value to determine decay. For a typical battery a period of 6 to 18 months will be necessary for the device to detect significant health decay in normal service. To simulate this, follow the ‘Simulation Protocol’ (See Exhibit A).
The BatteryInformer is programmed to report SOH = 0% when the measured conductance drops to 60% of the measured/stored maximum value. This level of ohmic decay is typical of a battery that retains less than 80% of its specified storage capacity. This threshold approximates the IEEE guideline for determining the end-of-life of a battery using the discharge method. This threshold is designed to indicate end of life at a similar level of decay as used in hand held test methods.
The ‘standard’ threshold level is appropriate for most applications. However, some designs have very large operational safety margins and can experience significantly more decay and still meet the minimum runtime requirement, in which case the alarm threshold could be further lowered. In a very few other applications, it may be appropriate to raise the threshold such that the end of life alarm will be activated earlier in the life of the battery.
If the IEEE guideline is not appropriate for the target application / design, it is advisable to establish a benchmark and request a customized threshold setting for your application. Use the following process to determine the target benchmark value.
A. Identify a number (preferably more than 5 pieces) of batteries in the target application that have decayed through normal aging to the target end of life condition. Two important considerations:
i. It is important that the batteries are chosen carefully to represent as closely as possible the desired end of life condition. Batteries that have not yet reached the target end of life condition, and batteries that have aged beyond the target condition should be disqualified.
ii. It is critical that new batteries (preferably 2 or more) of the identical model are also available to be measured for a reference value.
B. Fully charge the selected end-of-life batteries and connect them to a charger typical for the application.
C. Install a Smart-500 on each of the selected batteries and after one-hour document the Mho value. Calculate the average of the Mho values. This is the target end of life value for this battery model, Mho(eol).
D. Determine the ‘reference’ value for this battery model by attaching Smart-500 sensors to new and fully charged batteries of the same model. After one hour, document and average the Mho values, Mho(new).
E. Calculate the target end of life threshold for this application: (Mho(new) – Mho(eol)) / Mho(new) *100. This target can be used to create firmware customized for a particular application.
BatteryInformer® Evaluation Process
This guide will help expedite your evaluation process and ensure reliable results. This guide provides a framework for simulating, in only a few hours, the battery lifecycle decay the BatteryInformer is intended to measure normally over many months and years. The simulation is achieved by establishing a baseline value for a known good battery, and then attaching the BatteryInformer to a known older battery and observing the ‘decayed life’ of the older battery compared to the new battery.
The 5-Step Evaluation Process
Attach the BatteryInformer® Smart-500 sensor to a new and known good battery (“battery A”). Each terminal should be firmly installed between the post face (copper or lead) and a washer to ensure proper electrical contact for reliable sensor readings.
Reset the sensor (to erase any stored memory) by depressing the button underneath the right side of the rear label with a pen tip until the screen goes blank and begins a 30 second countdown. Connect battery A to a charger of appropriate size and type and allow for at least 2 hours of charging. This process establishes the correct baseline reference values against which all later measurements are compared.
After two hours of charging, observe and document the indicated battery state-of-health percentage (SOH%) and the conductance value (Mho) reading. The SOH% should be at or near 100%. If the SOH% is below 95%, repeat step 2.
Remove the Smart-500 sensor from battery A and attach it to an identical battery model known to be significantly aged (“battery B”). Do NOT reset the sensor memory. Ensure battery B is fully charged and also connected to the same charger as used on battery A.
After 2 hours, observe and document battery B SOH% and Mho readings. If battery B SOH% is significantly decayed relative to battery A, the display will indicate a low SOH%. If battery B is at end-of-life (SOH% = 0%) relative to the reference value established on Battery A, the Smart500’s contact closure will close indicating the battery needs replacement, and the red LED will flash rapidly. See Smart-500 user manual for more detail on display functions.
Yes, just as they would be in the target installation environment.
No, doing so will result in faulty data and false alarms. The BatteryInformer® Smart-500 is intended to monitor the SOH% decay of batteries in their normally charged, standby state.
The red light will double flash and an LCD ‘warning’ icon will be visible when SOH% drops below 30%. When SOH% reaches 0%, LED will rapid flash, the contact closure will close to send an alarm to a connected device, and a ‘fail’ icon will be visible on the display.
Battery conductance (1/r) decreases with discharge. The Smart-500 is designed to track the decay of the ‘full charge’ conductance value as it decays over time. Mho values and SOH% calculated when the battery is discharged should not be considered reflective of State of Health. Any alarms, LED and icon indications that activate during a discharge will clear upon fully charging the battery, as the conductance value rebounds.
Failure is indicated when conductance (Mho) drops below 60% of the learned reference value. Warning is indicated when conductance drops below 70% of the learned reference value. When conductance is 70% or above, the condition is ‘pass’. Using this ohmic comparison, the Smart-500 is designed to indicate a failure approximately when a discharge test would indicate 80% of rated capacity (IEEE definition of end of life.)
A discharge test is the only method to determine storage capacity (in Ah) or runtime (in minutes). Ohmic measurements have been used and accepted globally for almost 20 years as a proven measure of battery health. BatteryInformer® uses ohmic testing to provide real-time, cost optimized, and reliable estimate of battery health without the complications, cost, stress and service interruptions imposed by discharge testing.
BatteryInformer® utilizes a DC pulse load technique to produce accurate ohmic battery measurements. BatteryInformer®’s Mho measurement results closely correlate to battery manufacturers published ohmic values for new batteries.
Ohmic measurements obtained using AC methods are not typically directly comparable. BatteryInformer®’s Mho values may be significantly different than conductance values from “high frequency” testing technology.
Note that SOH% is calculated from the relative difference between the measured Mho value and the learned Mho maximum reference value. Therefore in practice an absolute Mho value is not of particular importance to determine SOH. By nature of the permanent connection and periodic testing, BatteryInformer® inevitably measures with superior precision.
Appendix A: BatteryInfomer Laboratory Test Protocol
- Compare indicated ohmic values with known ohmic values for the battery. Determine the manufacturers published internal resistance specifications for the particular battery models under test. Using new batteries, document the Mho values from the Smart-500 after two hours of charging, average the observed values based on the number of batteries being tested. Convert Mho to mohm: mohm = (1/Mho)*100. The internal resistance of the Smart-500 should be close to the published value for a new and fully charged battery. Note that published specifications are often conservative compared to the actual ohmic measurement. For example, if the specification is 3.5 mohm, a typical new battery measurement might be 3.2 ohm.
- Note: Some hand held testers will indicate values that also agree closely with manufacturer specifications, while other hand held tests may indicate values that vary widely from published specifications. Therefore, comparison of Smart-500 results, while correlating closely with the battery manufacturer specifications, may not correlate with the results of some other hand held testers. Important notice: please do not try to compare Smart-500 benchmark (baseline) to benchmark values produced with other hand-held testers, as some of these derived benchmarks are known to be different because of frequency and testing methodology.
- Obtain a population of batteries that are known to be close to end of life and have been in the field for some period of time (ideally 3+ years) or batteries that have been recently retired (don’t use batteries that have been retired and sitting idle for more than one month.) These batteries must be chosen such that new batteries of the same manufacturer and model are available for comparative testing.
- Apply the “5 Step Evaluation Process“ (detailed in main document) to the population of older batteries, using the new battery models as the ‘A battery’ reference.
- Document the State of Health of the older batteries as indicated by the Smart-500. In addition, document the displayed Mho value, and also record the conductance or internal resistance as measured with a hand-held tester.
- Subsequently perform full discharge tests (at least two tests per battery) to determine remaining % capacity using specified capacity rating.
- Statistically compare the Smart-500 indicated SOH values to the % capacity as indicated by the discharge test.
- Perform a “control” discharge test on the new ‘reference’ batteries to confirm that the % capacity comparisons (of the aged batteries) are valid.
- For any of the aged batteries that are indicated to be at or near 0% after performing the ‘Ohmic Decay Simulation Process’, document the Mho value as displayed on the Smart-500, and document the conductance value as measured by a reference hand-held ohmic tester that is certified to be compatible as described in test 1.
- Compare the Smart-500 indicated Mho values to the reference Mho values to confirm proper indication of end-of-life. Compare hand held conductance values to reference battery conductance values as measured by the hand-held tester. This comparison should also confirm end-of-life indication.