question

Al.McLuckie avatar image

Uneven Battery Bank Discharge Current

I have two battery banks of different sizes, the first supplying a DC panel with 2 X 100AH gel batteries and a Skylla 24VDC 50A charger, the second supplying a DC panel with 8 X 100AH gel batteries and a Skylla 24VDC 100A charger.


Coincidently, the larger of the two banks has the least working load at 17A and the first and smaller bank has a larger working load at 22.5A.


There is a consumer which is supplied a main supply from the first panel and a backup from the second panel. Both supplies are terminated and "on" at the same time through separate breakers. When I measure the DC discharge current at this consumer at each breaker I thought I would see an even load as the end consumer load is the same but this is not the case. The higher capacity battery bank and charger is supplying this consumer with 11A and the smaller capacity battery bank and charger is supplying the same consumer with 6.5A.


The only conclusion that I can infer from the tests I have done is that the larger capacity bank will supply the load unevenly and greater than the smaller capacity bank. Could someone confirm that, possibly show a formula that one could calculate these differences?

paralleldischarge level
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

3 Answers
Al.McLuckie avatar image

Hi Graham

Cable size is identical but the length is certainly not the same. My guess would be 15M or so

Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Graham avatar image

If I have got this correct two battery banks joined at a consumer type unit via separate fuses.

The most likely cause is the bank supplying least has a higher voltdrop between the batteries and the fuse. My guess is the feed cables are different sizes and lengths. If you want balanced use, the two feed cables need to be identical so that the voltdrop at the joining point is the same for both routes.


1 comment Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Thanks Graham

Cable size is identical however the length is not equal, probably around 15M difference

WKirby avatar image

It's difficult to answer this accurately.
What is the consumer that is using bot supplies? Are its supplies isolated from one another?

It's usually quite difficult to balance the current from two separate supplies evenly. If the Voltage of one supply is even slightly higher than the other then more current can flow from that supply versus the lower Voltage one.
Resistance or impedance of the two circuits would have to be exactly the same. A few milliohms here and there will make a difference. This is almost impossible to achieve. Cable lengths and screw torque ofthe terminals makes a difference and not really practical to balance out.

Again, without knowing exactly what and how the consumer works, an accurate answer will be difficult to provide.

1 comment Share
10 |3000 characters needed characters left characters exceeded

Up to 8 attachments (including images) can be used with a maximum of 190.8 MiB each and 286.6 MiB total.

Hi WKirby

The consumer is a control system for an Air Con rack. Funny you mention that when the voltage of one supply is even slightly higher than the other there is more current flow. When the larger bank charger goes into Boost, the imbalance is so great that it trips the breaker (two supplies are not isolated, other than MCB's.

When one trips that in turn means the other trips. Load is too great for the breakers and heat inside the switchboard is a factor too. It's obvious that there needs to be an upgrade to the MCB's and some internal wiring.

Trouble is trying to explain this to the customer. The system has worked for years and it's only happening now, the calculated load from as built drawings leaves little room for error. Compound that with varying temps in the switchboard, old components and new variations, but the customer can't get passed the imbalance and thinks there are underlying issues. I was hoping that there was a magic formula that I could reference.