0
\$\begingroup\$

I have a tornado shelter that I have wired for both 120V AC and 12V DC. All of the lighting is via 12V DC while the AC is used for recharging the batteries and running a 1500 VA UPS system that will power a small TV and radio. The UPS system normally uses two 12V SLA batteries in series. To increase the run time of the UPS I want to use auto batteries instead of the SLA batteries. In order to cut down on the cost of batteries I want to use only two auto batteries instead of three. One 100AH battery used for the main lighting and a 50AH battery in series with the 100AH battery to provide the 24 volts for the UPS system. The UPS system will only use the two batteries when the AC is out which is normally rare. Do you see any problems with this setup providing I install a diode in the UPS battery feed to prevent the UPS from recharging the batteries and use separate smart chargers constantly connected and recharging (as long as AC is available) each individual battery? I feel I have overlooked something but I can't seem to put my finger on it.

\$\endgroup\$
  • \$\begingroup\$ A simplified schematic would be helpful. \$\endgroup\$ – Caleb Reister Oct 21 '19 at 20:14
  • \$\begingroup\$ With non-sealed batteries, are you concerned about the gasses ( hydrogen-sulfide ) produced by the batteries building up in the shelter? \$\endgroup\$ – Aaron Oct 21 '19 at 20:19
  • \$\begingroup\$ No, the shelter is well ventilated so build up should not be a problem. \$\endgroup\$ – Ned Olav Oct 22 '19 at 3:36
1
\$\begingroup\$

I would avoid all the possible issues of isolated chargers etc by having the 24V battery bank and running the lights from a 24v to 12v converter.

They are not expensive and it simplifies things if someone else has to conduct repairs.

| improve this answer | |
\$\endgroup\$
0
\$\begingroup\$

Verify that your charger outputs are isolated from ground.

You probably don't want to use generic automotive batteries -- you want to use deep discharge batteries, of the sort designed for RV lighting, golf cart driving, etc.

| improve this answer | |
\$\endgroup\$
  • 1
    \$\begingroup\$ I'd say that the choice of charger plays an important role here also. SLA batteries can be destroyed slowly by a poor charger, and Murphy's law dictates this will only be discovered when they are actually needed (the worst possible time.) \$\endgroup\$ – rdtsc Oct 21 '19 at 20:02
0
\$\begingroup\$

What you are proposing would not usually be deemed "good practice" BUT sounds reasonably workable in your case. Two batteries in series with different loads on each will often lead to mismatches in aging and charging. Different capacities = worse again. But in your application this seems "reasonably well catered for".

The two mains based chargers need to have isolated outputs - ie not share a ground (or any other) connection between their output and input circuits. This would usually be what you'd expect. Chargers with a ground input lead MAY have noise filter capacitors from both inputs to AC ground and MAY connect this to DC negative out - but usually not. A simple ohmmeter check will allow isolation to be checked.

The diode in the UPS lead will probably not cause a problem. A Schottky diode will give less loss. A P Channel MOSFET with low Rdson and appropriate connection* will allow a low loss diode to be formed. [Drain to B+, Source to output, gate to ground as long as Vgs max >> 12V).

If thelights never discharge the battery to more than say 90% of full charge then an automotive battery may last a long time. If discharge is often much lower than that then a deep cycle battery may be wise. Even deep cycle batteries should not be taken below say 60% SOC (higher better, much higher much better) for longish life.

As Tim says - try to ensure that the chargers treat the batteries well.
Worst case capacity can be much reduced but this is unlikely.
Mor likely is lack of a properly applied topping / boost charge after use and so a reduction in capacity with time, but usually not a vast one.

A VERY OCCASIONAL deep discharge test - maybe annually - will give you a reasonable indication of batter health. Serious requirements end up with SOC monitoring and acid checking and .... but that is probably beyond what you need here.

| improve this answer | |
\$\endgroup\$
  • \$\begingroup\$ Thanks for the thoughtful review! I realize that it isn't "good practice" for a number of reasons but since the usage is very low and I am trying not to oberly invest in this project I thought it would suffice. Under normal circumstances the batteries should never discharge below 40% to 50%. Only if there is a complete disaster and I am trapped in the place for days should there ever be a problem in which case I have a DPDT switch to place both batteries in parallel and run the lights only (which are individually switched) for several days if necessary. The chargers \$\endgroup\$ – Ned Olav Oct 24 '19 at 18:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.