Battery Internal Ohmic Measurements – Part 3

  • Home
  • Blog
  • Battery Internal Ohmic Measurements – Part 3

Battery Internal Ohmic Measurements – Part 3

Introduction

In part 1 there was the introduction of ohmic measurement and some of its challenges. In part 2 the four wire connection was explained and the influence the connection can have on the reading.

This article will look wider and consider what the battery may be connected to. We explain that an online battery ohmic reading can be influenced by parallel paths to the battery string – for example the charger, load or other batteries. This influence is typically only significant in strings with a small number of batteries i.e. two or four batteries in a string.

Explanation

To measure battery impedance, a known test signal is applied to the battery and its voltage response measured.

This influence is usually only significant in small battery strings, for example two batteries in a string. The level of influence also depends on the cable resistance, battery charger, other parallel battery strings, load and anything else in the circuit.

To understand why the battery impedance reading can be influenced, one way is to take a look at the path of the test signal. In figure 1, battery #2 is being measured and the test signal current is shown in red. The string breaker open, the total test signal current is flowing through battery #2.

battery parallel impedance one

Figure 1

When the string breaker is closed the current path of the test signal has changed and shown in Figure 2.

battery parallel impedance two

Figure 2

From Figure 2 you can see that the test signal current through Battery #2 is reduced because some of the test signal is flowing through other parts of the circuit.

For the impedance measurement device to provide a reading the formula is: \(|{Z}|={V\over I}\)

| Z | is impedance, V is voltage and I is current.  The Volt meter measures V and the Test signal is I.

For example, let’s say that Battery #2 is 5mΩ and that the impedance of the parallel path causes the test signal current to have a 90/10 split. Battery #2 has 0.9A test signal going through it.

In Figure 1 when the test signal of 1A is applied, a battery a voltage is measured of 5mV.

Battery #2 impedance \({5mV\over 1A} = 5mΩ\)

In Figure 2 when the test signal of 1A is applied 0.9A flows through the battery so a voltage is measured of 4.5mV.

Battery #2 impedance \({4.5mV\over 1A} = 4.5mΩ\)

The impedance of Battery #2 has not changed but closing the breaker, the reading has changed because the circuit has changed and the test signal through Battery #2 is less.

Experiment

To demonstrate the effect with some actual measurements, two measurements are taken on a battery in the following configuration where the battery is out of the circuit and in circuit. This configuration is where the effect is most significant.

Equipment:

  • 4 x VRLA batteries 32Ah
  • Interlink cables
  • Hioki BT3554-01 battery tester

Configuration Result

Battery out of circuit

Battery out of the circuit


Battery = 6.06mΩ

Battery in circuit

Battery in circuit


Battery = 4.68mΩ

 

The change in reading is (6.06 - 4.68)/6.06 x 100 = 23%

Another example is a configuration with two UPS battery strings of 32 12V batteries per string.

A battery out of circuit = 6.53mΩ, a battery in circuit/online = 6.36mΩ

The change in reading is (6.53 – 6.36)/6.53 x 100 = 2.6%

As you can see the amount of change in the impedance measured largely depends on the number of batteries in a string.

Conclusion

As mentioned, the battery impedance reading is influenced when a measurement is taken when the battery is in service/online however the level of influence mainly depends on the number of batteries in the string, in the experiment there was a 2.6% change for the string of 32 batteries and a 23% change for a string of 2 batteries.

As has been highlighted in Parts 1 & 2, good impedance trending relies on a consistent measurement context. A change from offline/disconnected to online measurement, a change in installation arrangement etc. may affect individual measurements and create data that is out of context. Consideration also have to be taken with the baseline and the percentage of change for the alarm threshold.

When trending battery impedance measurements for installations with small number of batteries per string, any changes to the installation [e.g. addition of new parallel string] must be recorded and that information made available to the person interpreting the results. A change in battery impedance may be due to an installation change and not the battery itself.

If you have any questions about battery ohmic measurments, don’t hesitate to contact us.

About the Author

With over 20 years experience in the battery monitoring industry as a Technical Manager at PowerShield, Paul Hectors is an accomplished expert providing valuable battery management insights. Paul's background is in Electronics Engineering and in the design and development of sensors for batteries in standby applications. 

1-2

By PowerShield | March 11th, 2020 | Technical series | 0 Comments

About the Author: