0
\$\begingroup\$

I had an idea and wanted to present it here and ask if anyone foresees any problems with doing this... I have both a 48VDC input inverter and a 24VDC input inverter that I want to run from a 72V series bank of batteries (six identical 12V 100Ah AGM batteries in series). So imagine the batteries lettered sequentially A thru F, the 48V inverter would attach to A+ and D-, the 24V inverter would attach to E+ and F-, and the solar charge controller would attach to A+ and F-.

So the idea is to use a higher voltage battery bank since the solar charge controller (scc) I am using is a boost converter that requires the battery bank voltage to be higher than the solar panel input voltage. To get the maximum power out of the scc, I have to feed it close to 60V input which is the maximum allowed since there are 63V capacitors on the input side, which means my battery bank has to be 72V or higher (output capacitors are rated at 100V).

I will attempt to load the 48V and 24V inverters about equal (in proportion to each other), so that the 6 individual batteries are in a similar state of charge (SoC), and so that the scc should charge them all about equally.

Another advantage of using 72V that I just noticed is I also have an identical pair of 36V "golf cart" type chargers that I could use to top off the batteries on cloudy days. That is, at the end of the day if the batteries are still not charged from solar, to keep the batteries healthy, I can top them off by using a pair of 36V smart chargers each on a 3 battery subbank (charger 1 on ABC and charger 2 on DEF). I haven't tested to see if my chargers are compatible with that wiring arrangement but I should be able to test it soon.

So will/should this entire setup work? Anything to "watch out for"?

\$\endgroup\$
  • \$\begingroup\$ Could you please add your estimate of min max and average load wattage(nothing fancy just best guess) and also the wattage ratings of your panels, charger, and both inverters? \$\endgroup\$ – K H Jan 12 at 8:01
  • \$\begingroup\$ 48V inverter max wattage allowed is 1600W (2200VA). 24V inverter max wattage allowed is 450W (700VA). I will be using a pair of solar panels wired in parallel. Vmpp is 54.7V, Impp (of the pair) is slightly over 10A. Solar charger can output about 500W max. Output voltage is programmable but will be set at about 81.0V for the 72V battery bank so about 6A max. Generally speaking, the larger inverter will be loaded moreso than the smaller inverter. For example, the smaller inverter can run my laptop computer, some LED lights... while the larger inverter is for space heater, ceiling fan.... \$\endgroup\$ – David Jan 12 at 8:21
  • \$\begingroup\$ Also, another thing I can do is physically move the batteries in a set schedule so that they get about equal time in both subbanks. This should help lessen an "bias" in loading/charging. For example, the 48V subbank has batteries A,B,C, and D. The 24V subbank has batteries E and F. I could (in a roundrobin fashion), choose which 2 (of the 6 batteries) go into the 24V subbank, so sometimes it would be A and B, sometimes C and D... Perhaps once a week I could "rotate" them this way. Over years, they should be more "balanced" on average vs. if I never did this. 15 two battery combos exist. \$\endgroup\$ – David Jan 12 at 8:36
  • \$\begingroup\$ You're talking about hundreds of hours of effort now. Part time minimum wage job, dozen or so hours, buy correct part, figure out what to do with leftover dozens of hours. Then you can spend those hours on the fun parts of your hobby. You don't want to create a maintenance situation. \$\endgroup\$ – K H Jan 12 at 8:42
  • \$\begingroup\$ Please draw a schematic and/or block diagram of what you are proposing. \$\endgroup\$ – winny Jan 12 at 10:53
2
\$\begingroup\$

I had an idea and wanted to present it here and ask if anyone foresees any problems with doing this... I have both a 48VDC input inverter and a 24VDC input inverter that I want to run from a 72V series bank of batteries (six identical 12V batteries in series). So imagine the batteries lettered sequentially A thru F, the 48V inverter would attach to A+ and D-, the 24V inverter would attach to E+ and F-, and the solar charge controller would attach to A+ and F-.

Batteries to be used in series banks should always be matched and balanced before connection and discharging them in this manner will unbalance them unless you use some advanced battery balancing circuit. Testing an already used battery for match and balance is an advanced task. You would also have to design and build the circuit yourself as in almost every case this situation is to be avoided rather than worked around. Exceptions are things like taking a nanoamp voltage reference from some subset of cells. No currents of consequence though.

So the idea is to use a high voltage battery bank since the solar charge controller (scc) I am using is a boost converter that requires the battery bank voltage to be higher than the solar panel input voltage. To get the maximum power out of the scc, I have to feed it close to 60V input, which means my battery bank has to be 72V or higher.

Nothing wrong with using that voltage, but you would need to get an inverter with the correct input voltage or switch to a voltage that is compatible with one of your inverters.

I will attempt to load the 48V and 24V inverters about equal (in proportion to each other),

Your two inverters will have different loads and efficiency curves so even if you matched the loads you connected flawlessly the loads on the batteries would be different anyway. To produce adequate accuracy, you would need a specifically engineered solution, significant difficulty for someone with an EE degree.

so that the 6 individual batteries are in a similar state of charge (SoC), and so that the scc should charge them all about equally.

Because of permanent effects of charge and discharge on the batteries, it wouldn't be adequate to maintain a similar state of charge. You must maintain a similar state of charge, but also a near identical charge/discharge history.

So will/should this work? Anything to "watch out for"?

Funny you should phrase it that way. This connection method in itself is fraught with peril and is in itself something to "watch out for" ;D.

Just a further note, batteries are a significant part of your system cost. If you wanted a 6s(72V) battery bank, if you want to replace a battery and you don't know how to properly match, maintain and test them, you're stuck with buying batteries in sets of 6. In many cases battery cost is greater than converter cost for a system. I'm telling you not to connect this way so you don't end up with a house fire, but even if you just damaged your batteries or wore them out faster you might spend more money than the correct equipment.

Edits/Additions: Regarding your proposition regarding load balancing, note that if your small inverter provides 500W to a load, this will require a different amount of total battery power per battery than if your large inverter provides 1000W. Wear on the batteries will be based more or less on the total currents, both charge and discharge, with larger currents wearing out the battery faster. If you draw 10A for 1/2 hour, that will wear the battery out more than if you draw 5A for 1 hour. Given this, it is more or less impossible to balance battery wear properly and difficult and labor intensive to actively match batteries. The labor cost exceeds the cost of new equipment by an order of magnitude, even at minimum wage in most areas. Your time has value. Also if your system just happens to charge when your batteries are a bit imbalanced, it will cause battery wear on those that are overcharged. If it happens when they are more imbalanced, rapid failure of 2-4 batteries, possible outgassing/fire, so you had better be sure to check and charge them frequently enough.

Regarding connection of equipment in unintended manner: The casing and ground contacts of your inverters have to be grounded on at least one of the input or the output. It likely has multiple internal connections to ground through it's electronics in addition to the bonding wire. Unless it is intended specifically to be used on an isolated-from-ground battery. In most cases the negative DC terminal will be grounded, so when you connect both inverters and their grounds, you will be creating a 48V direct short.

Series arrangement of batteries - This is very important for large batteries/banks. The batteries should be matched in characteristics, before being combined, not just matched in voltage, but ideally brand new identical batteries, obviously from the same manufacturer. They should be grouped up and never charged or discharged separately. Maintaining a bank where load is uneven or batteries are rotated requires a professional level of skill, aside from the large effort.

As of early 2019, $120 for the crappiest versions of a 3kw 72V inverter or $20 and $65 for a 500W and a 2000W if you actually want two inverters for efficiency reasons. I'd consider shopping around for a 2-3kW inverter with good efficiency and a flattish efficiency curve.

I won't tell you not to learn the way you want, but do so knowing this is a very bad idea and do everything possible to mitigate the dangers.

Be sure your battery storage is up to code or exceeds it if necessary. Depending on where you live there may be no code or you may be required to have a metal box or a specific method of venting. Nothing crazy. If you were to have a house fire that started at your battery bank and this setup was found to be the cause, it will in all likelihood void your homeowner's insurance unless the insurance company neglected to check.

\$\endgroup\$
  • \$\begingroup\$ Remember this is for a solar charging application and the loads will be common household loads such as a laptop computer, LED lights, perhaps even an occasional space heater. I am pretty good with batteries and getting them to "balance". I suppose at night, when there is no solar charging, I can "pre-balance" the lower bank using a regular wallpowered charger, let it sit overnight, then when the sun comes back up the next morning, the bank of 6 should be charged fairly evenly. Load balancing in a system like this is tricky but doing what I described should help a lot with "even" charging. \$\endgroup\$ – David Jan 12 at 7:31
  • \$\begingroup\$ Still a huge nono. Lead acid batteries are usually used in home solar setups, and they are one of the most tolerant types to this type of treatment, yet this should still not be done. In order to pull it off, you have to balance your loads flawlessly and in real time. If you put a tremendous amount of time and effort in over the years, you could conceivably minimise the risk, but your time would be better spent working the hours at a regular job to get the proper equipment. I've taken the angle to discuss the batteries here, but perhaps I should add a section about equipment grounding. \$\endgroup\$ – K H Jan 12 at 7:56
  • \$\begingroup\$ I disagree wit the "must always be balanced" since it would depend on how they are drained. For example, in my 72V battery bank setup, if instead of using 48V and 24V inverters, I used 48V and 24V inverter chargers, I could then drain the 24V subbank more (or less) than the 48V subbank. I could even have a 72V load across the entire bank. Whatever imbalance there is say in the 24V subbank, the 24V inverter/charger will handle that. So the more accurate statement is series battery banks containing subbanks need each subbank to be balanced but independently of other subbanks. \$\endgroup\$ – David Jan 12 at 14:55
  • \$\begingroup\$ What you say is true if your equipment is designed and rated specifically for the purpose you describe. Specifically you would need isolated inputs/outputs from the battery side and load sharing capable outputs in order to share a load. This type of specialized equipment is called a "grid tie inverter". The question in this scenario would be if you had two separate ungrounded battery banks, why you would bother making a completely unnecessary connection between them as they are not in series in any meaningful way. \$\endgroup\$ – K H Jan 13 at 0:36
  • \$\begingroup\$ Part of the problem is the tolerance of lead acid batteries, which forces me to concede that given enough diligence, you will experience aggravated battery wear rather than battery destruction. If you really want to take a risk, by all means connect the equipment without checking its relevant properties and see if the smoke comes out. It's your money. If on the other hand you'd like to confirm what I say on a shorter timeframe (probably similar budged just due to how cheap the inverters you need are) try designing yourself a lithium ion charge/discharge circuit that behaves as you wish. \$\endgroup\$ – K H Jan 13 at 0:44

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.