Battery Discharge Test

[shkim2000]

I am preparing commissioning procedure for battery discharge test. I have read tech manual from Saft-Nife Battery. During discharge test do I record current and voltage at certain interval ? If so, what is time interval for recording ?
Why dischangre test is required for battry ?

 

[alehman]

If the system is large, it would probably be best to hire a professional engineer experienced in such testing. There is a lot of knowledge and preparation involved in getting a meaningful test and avoiding damaging the battery.

Yes, you should record voltage (of every cell if possible) and current as a minimum. The time interval is up to you. You should have at least 5 readings during the discharge. 15 or more is much better.

A discharge test will help you determine the condition of the battery. You can determine the total capacity, find bad cells, bad connectors, etc. It is the best (but not definitive) way to know the battery is likely to perform when called upon by the UPS.

 

[supraskate13]

I myself have performed Load test and I'm not sure if it is the same thing as the test you are talking about but here is my experience anyway. I connect the whole battery bank to a rented load bank and ran the bank for 8 hours taking voltage and hydrometer readings every 15 minutes for the first hour, every half hour during the second hour and then every hour on the hour after that for the last 6 hours. The reason for so many readings during the first hours is because the battery bank will drain rapidly at first (to about 80% of its rated voltage I believe) and after that the battery should maintain at least 80% for the rest of the test. The reason for running the test 8 hours is because a battery's rating (amp-hours) are based on 8 hours. so a battery bank with a 80 amp-hour rating can supply 10 amps for 8 hours, or 20 amps for 4 hours.....etc. This test proves the capacity of every cell.

 

[nukeman]

IEEE standards (450 lead calcium, 1188 VRLA, etc) provide the recommended frequency for testing batteries. The vendor instructions normally follow the IEEE guidelines. Keep in mind there are two separate discharge tests that are specified, a capacity/performance test and a service test. The capacity test is normally performed every 5 years whereas the service test is performed more frequently. A capacity test is a constant current test to vendor specified parameters (discharge current and end voltage) and the service test emmulates your designed load profile.

Per IEEE-450 a new lead-calcium station battery is required to be discharge tested within the first two years after installation, some users require a discharge test immediately upon installation.

The best way to monitor the battery bank during a discharge test is to use a system that is designed to monitor the voltage on each cell and the discharge current during the entire test. The data can easily be graphed and any suspect cell identified (i.e. low cell voltage). The test duration should be the same length as what your system is designed for i.e. if you require a 2 hour back-up capability you would run the test for 2 hours at the manufactures rated value at 2 hours. You also need to determine what your end voltage is, this value is normally 1.75 volts per cell but it can be any value it just depends on the low voltage limit of your loads and what type of battery.

If you do not have access to a recording device to monitor all the cells take cell voltage and current readings every 15 minutes of the first hour and every 30 minutes thereafter.

I am not sure why you would want to take hydrometer readings during a discharge test since the specific gravity value will change throughout the discharge. I guess you could theoretically compare the value between the cells but on a large system chances are the values would be different from cell 1 and the last cell recorded.

 

[alehman]

Here's an article I wrote on the subject if you want to get into the details.

ftp://ftp.gbutler.com/pub/CFG/battery-cx-article.pdf