WHETHER PARALLELING OF DC BATTERIES IS RECOMMENDED.

[rajashri]

DEAR ENGINEERS,
I HAVE TO ENGINEER A 24 V DC SYSTEM FOR I@C SYSTEM OF A POWER PLANT. THE AH CAPACITY OF BATTERY IS COMING TO 2500 AH. SINCE THIS CAPACITY BATTERY IS NOT AVAILABLE IN PLANTE TYPE OF LEAD ACID BATTERIES, WE INTEND TO GO FOR A 2*1250 AH BATTERIES. SO IN SUCH CASE WHETHER, IT IS RECOMMENDED TO GO FOR PARALLELING OF TWO BATTERIES? WHAT EFFECT IT WILL HAVE ON FAULT LEVEL,? AND WHAT ABOUT THE CIRCULATING CURRENTS? AND WHAT SORT OF COMBINATION CHARGER SHOULD BE SELECTED? IE SINGLE FLOAT AND BOOST OR FLOAT CUM BOOST AND A BOOST CHARGER. CAN ANY BODY ADISE AND SHARE THEIR EXPERIENCES ON THIS SUBJECT?
IS THERE ANY ALTERNATE WAY FOR THIS ?

 

[jbartos]

Suggestions:
1. If possible by some different battery type, avoid the battery paralleling
2. If batteries are in parallel connection, it is better to parallel them cell by cell, i.e. two cells in parallel are then connected in series to obtain 24 VDC. This way if there is a short in one of those parallel cells, about 2Volts will be missed from 24VDC power supply. Remaining cells will not be negatively affected except by slightly higher discharge rate.
3. If for some reason, there will be two 24VDC battery branches connected in parallel, then the current monitoring in the paralleling conductors is recommended. This is to avoid discharging and overcharging one parallel battery branch due to some shorted cells.
4. Usually, a battery manufacturer has the preferred type of charger for their batteries. This may be linked to battery warranty.
5. Follow some industry standards for Batteries, e.g. IEEE Stds 450, 484, 485, 946, etc., as applicable.
6. Paralleled batteries will produce about one half of the unparalleled battery branch internal resistance. This will increase the short circuit level, approximately twice.

 

[busbar]

24-volt battery sets in generating stations can be true disasters with supposedly “compatible” instrumentation and control during borderline undervoltage conditions. The reason it is sometimes “required” is for starting-motor loads. Carefully consider 48V minimum.

May want to review:

IEEE 484-1996 …Installation Design and Installation of Vented Lead-Acid Batteries for Stationary Applications

1375-1998 …Protection of Stationary Battery Systems

 

[alehman]

Paralleling battery strings is common practice for telecom and UPS applications. Telecom facilities nomally have many parallel strings in their 48V systems which can get as large as 10's of thousands of ampere-hours. Eash string must be individually fused. Paralleling is undesirable from the point of view that you have to separate the strings to test them properly, which must be done periodically.

As jbartos said, if you parallel multiple strings they should be nearly identical. One frequently overlooked factor is to keep them at the same temperature. If they are at different temperatures they will have different float voltages and different charge levels. I agree with busbar - use a higher voltage if possible.

 

[schwarz]

I would continue on alehman's comments. It's common to parallel even 4000AH,5000AH battery banks in telecom installations.Since batteries will be new and from same batch/mfr paralleling won't be a problem.In our case cell to cell paralleling is not allowed.
You should also specify :
Auto Low Volts disconnect(LVD) ( contactor in each battery path ) to protect from over discharge
Battery temp. compensation if VRLA ( Sealed maintenance free ) batteries are used.
Manual Isolation (through switches ) for maintenance /testing.
DC/DC Converters can be used if different voltages are reqd.

 

[buzzp]

One other note not clearly stated is to use batteries from the same mfg purchased at the same time. When replacement is needed, all parallel cells should be replaced at the same time to prevent unequal load sharing between batteries.

 

[TWW]

Amen to that, I tried to save money (get cheap) by replacing weak cells in a 24 cell bank. slowly and painfully the hole system went to h*ll and scrambling I resorted to only using the new ones. This reduced sized bank rapidly deteriated and forced the purchase of full new set. Big waste of time and money. Less time wrestling those sinking lead acid batteries the better.

 

[peebee]

To take the UPS example one further:

Battery strings are commonly intentionally parallelled in UPS applications for reliability purposes. That is, if you have two strings and one string fails, well at least you still get half your run time from the remaining string.

I don't quite understand why that reasoning would not apply to any battery installation.

Individual cell impedance monitoring is not unusual (although not exactly commonplace) to detect individual failed jars within parallel strings. They typically inject an AC current into each cell to check its impedance. Again, I don't see why such a system could not be employed with any battery installation.

 

[jcai]

GOOD MORNING FOLKS - ALL BATTERIES ARE D.C. - I LIKE THE COMMENTS ABOVE - ALL CLOSE TO THE MONEY - IDENTICAL BATTERIES CAN BE PARALLELED - BUT MUST HAVE VERY CLOSELY RELATED INTERAL IMPEDANCE - SO THE LIFE CYCLES THAT HAVE BEEN TAKEN OUT MUST BE VERY SIMILAR - YOU SHOULD NOT PARALLEL AN OLDER BATT WITH NEW ONE IN OTHER WORDS . THE LOW IMPEDANCE BATT WILL DISCHARGE INTO THE HIGHER IMPEDANCE ONE IF IT HAS HIGHER VOLTAGE - CAN START THERMAL RUNAWAY !
WHY DONT YOU TALK TO THE MAIN BATT MANUFACTURERS LIKE VARTA HAGEN GERMANY OR SAFT ROMAINVILLE FRANCE OR HAWKER ENERGY NEWPORT WALES - NOW INVENSYS OWNED .
IN U-BOATS WE USE MAIN BATTS WHICH ARE A HEAP LARGER THAN MOST FOLKS HAVE SEEN - IN AIRCRAFT WE USE SINTERED PLATE NI CAD BATTS - IN POWER STATION APPLICATION I WOULD USE POCKET PLATE NI CADS FOR COST EFFECTIVENESS + RELIABILITY - LEAD ACIDS ARE AN O.H.+ S HAZARD IN ANYONE,S LANGUAGE .
NI CADS NEED DIFFERENT CHARGER DESIGN TO LEAD ACID BATT .
MAYBE S.L.A.B.s ARE A POSSIBILITY ALSO. ALL THE BEST J.C.

 

[alehman]

I concur with peebee and jcai. Individual cell or jar impedance tests are a good quick check (that can be done on-line) to see if you have any serious problems, but it doesn't tell you a lot about the capacity of the cell. It's a good idea to do an impedance before a full discharge test since a cell with much higher than average impedance will probably fail in a discharge test.