AC millivolt test on battery banks

[Machine Gun Kelly]

I am trying to understand how to preform and interpret AC millivolt on industrial battery banks. I have searched on line and I can't find a good explanation of the test.

When I preform the test with a Fluke meter I touch the battery with the probes and I will get a high reading maybe 50 mV or higher. The longer I hold the meter on the battery the lower the reading goes. After 15 or 20 seconds it is down to 4 or 5 mV.

The question is what is the true millivolt reading the initial 50 mV or the later reading and what is this test telling me.

Thanks, MGK

[madMAX]

Hey MGK, help me understand exactly what you are trying to do, I have never heard of an AC millivolt test for batteries. Are you trying to measure impedance? If so you will need a test set specifically designed for testing batteries, such as a Megger BITE or Alber Cellcorder. Check out IEEE Standard 450 for a reference.

[Machine Gun Kelly]

I have seen the UPS service techs do this test and I should have ask one of them but I did not. The little bit that I know about it is that basically the batteries in the bank are in series across the output of the charger. By doing the AC millivolt drop test you are measuring the ripple out of the charger and how it is divided across each battery is indication of that particular battery's health. I think it might be a rule of thumb test but i cannot say for sure.

I do not have access to the IEEE standards.

Thanks MGK

[veracon0700]

I don't have much experience with this test but from what I can gather you will want to compare each battery to the others in the string. Go with the initial number and investigate any value that has a ripple voltage twice that of the others in the string.

Your best bet would to rent a battery impedance tester and test each cell with that. The Alber cellcorder for example can take voltage, cell impedance and strap resistance all in a single test. Do this once a year.

Here is a little write-up on the theory behind the test in question:

Usually, communications rectifiers are highly filtered and there is very little AC ripple voltage impressed on the connected battery. However, substation and UPS battery chargers typically impress a significant AC ripple voltage on the connected battery causing a measurable AC current to flow through the battery.

If all cells had identical impedance, this AC ripple voltage would be evenly divided across the cells. However, since the same AC current flows through all the cells, those cells with higher impedance will exhibit a greater AC voltage across the terminals (E = I x Z) while those with lower impedance will exhibit a lower AC voltage.

While the AC current could be measured and the impedance calculated as Z = E / I, for troubleshooting purposes the measured AC voltages per battery can be compared directly and treated as though they were impedance measurements since the same current flows through all units.

The calculated values of impedance using this method will not be the same as that derived using commercial test equipment since the frequency of the AC ripple is different and it may also vary from time to time.

Also, the value of the measured AC voltage per battery may vary at different times due to variation of the amplitude of the AC ripple voltage across the total string. However, with these constraints in mind, this technique can still provide a valuable troubleshooting tool.

Reference: Impedance and Conductance Testing (C&D)

Hope this helps!

[Machine Gun Kelly]

That is a big help. Thanks MGK

[Jay Knight]

compare each battery to the others in the string. Go with the initial number and investigate any value that has a ripple voltage twice that of the others in the string.

That's pretty much what I was taught when learning how to test batteries. Great post.

[Rick_Tressler]

Just a point of clarification. The Alber Cellcorder measures internal resistance as opposed to impedance as stated previously in this thread. There are three types of ohmic test methods today; Resistance, Impedance and Conductance. The term "ohmic" covers all three as a generic term. If you have further interest in ohmic testing, I suggest reading a very informative, white paper on the subject. It can be found on the Alber website at www.alber.com.

[unimployed]

This thread was very helpful. Great info