I have an off-grid solar system where this inverter/charge controller (ICC) is connected to a 48V battery (which is in fact 2 parallel quartets of 12V batteries, but this is non-relevant detail I guess). The ICC would feed current to the battery at as high as 56V (high sun, low load and when the battery needs charging), and draw it back (for making AC — when the sun is down and/or load is high) to as down as 42V.

Now, the hitch is that I have some DC appliances and would like to feed them DC without the double conversion through AC (the ICC does not have a DC output for connecting load). The idea is to get a DC-to-DC converter (which would deal with 42-56V floating input) and let it draw DC from the battery/ICC circuit directly — in parallel.

I am hereby trying to figure out if this setup would undermine the ICC's job to any such extent that the battery gets charged/discharged incorrectly and therefore passes away sooner than if I did not do this.

The extra parallel circuit will push down the overall battery/ICC circuit resistance so that:

  • when the ICC is charging the battery, the effect will be just higher current in the overall circuit — as the ICC will (attempt to) maintain the voltage;
  • when the ICC is drawing current from the battery, the effect will be just that the voltage drops faster — as the ICC won't be the only one consumer anymore.

So it looks like the only thing to watch out would be the battery's depth of discharge as the ICC won't be able to account all the current drawn from it anymore, and would only see the voltage as indicator of discharge.

How do the above understanding, reasoning and conclusions stand? Is there something I am fallaciously missing, or is it all reasonably sound?

  • \$\begingroup\$ You must say what battery chemistry is (Lead Acid assumed). LA at 42/4 = 10.5 per 12V battery is very deep discharge. If you want any lifetime you need to stop discharge well abiove this voltage. \$\endgroup\$
    – Russell McMahon
    Jun 24, 2018 at 11:34
  • \$\begingroup\$ @RussellMcMahon the low voltage cutout (42V) is given just as example. It can be configured via the ICC, and I guess it is not relevant to the question. The question is not about numbers but behaviour of the system when extra load is attached to the battery in parallel. As a side note, the batteries are lead carbon allowing 4000 cycles at 50% discharge and 1500 cycles at 80% discharge. \$\endgroup\$
    – Greendrake
    Jun 24, 2018 at 12:01
  • \$\begingroup\$ Nice battery spec. It would be nice to know how that compared to LiFePO4. || If I understand your description correctly, if you diode* feed your auxilliary inverter from the PV input and the battery it will be invisible to the ICC when solar charging but not when it is drawing battery power. That imbalance may your situation better oir worse :-). | I'm sure you don't wish to, but there is quite probably a point inside the ICC at the "true inverter" input where coulomb counting monitoring has been done. Drawing auxilliary low voltage from there plus the diode feed as above does what you want. \$\endgroup\$
    – Russell McMahon
    Jun 25, 2018 at 13:04
  • \$\begingroup\$ I said "diode" but you could use FET switch or relays if desired to lower energy loss. \$\endgroup\$
    – Russell McMahon
    Jun 25, 2018 at 13:04

2 Answers 2


As you are swinging the batteries between 56v and 42v, they are already getting a very hard life, especially the 42v end. I am assuming they are lead acid, and I am hoping they are deep cycle types. I doubt that the controller is doing any coulomb counting, there's no real need to do that with lead acid, so putting an extra load on the batteries is unlikely to worsen their cycle life, except to the obvious extent that more load possibly means more cycles. Beware that more load does not mean lower minimum voltage.

If you want longer cycle life, then limiting the bottom voltage to 44v or even 46v would be well worthwhile, if possible. You would win financially by limiting the low voltage, and putting an extra battery string in parallel with the existing to make up for the capacity being reduced by limiting the voltage swing. The better cycle life would more than make up for the higher investment in batteries.


If the ICC is doing coulomb counting It may mistake your side load for a damaged battery.

Your DC-DC converter should be equipped with a cut-out so that it doesn't draw the battery down below 42V. you could perhaps monitor the AC output and shut down the DC output if the AC fails.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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