Glen Apollo

Auto turnoff is challenging. If the inverter draws 75W idleing you will have to have a sensor that only turns on when maybe 100w is asked for.
I don't think this can simply be added to the inverter input wiring?
Doing some math, 75w at 48v = 1.5A idle current.
15A for 10 hours of darkness.
A simple option would add a power relay in the inverter input that can switch 100A. If my math is correct 4000W is 89A at 48v.
The relay can be switched via a wired or inexpensive wireless switch.
Ebay or Banggood sells 12v coil 100-150A power relays. Use a regulator to drop 48v to 12v.  The relay coil draws maybe 1/4A 250 ma. When energized.

The power relay would simply be installed in series with the + inverter input wire.
These power relays come with connection typebolts with nuts to accept cable lugs.
The relay coil sw would switch power to the Buck regulator set for 12v output from the 48v battery. When the sw is off, no current would be drawn for the relay coil control.
Another option is to use the voltage from one 12v in the bank for the 12v relay coil voltage source eliminating the regulator.

Typical 100A power 12v relay.
https://www.ebay.com/itm/282068085904

48v in:12v out adjustable regulator
https://www.ebay.com/itm/171247024835

Thanks for the reply. What I've found out with testing. Using a pwm controller setup for 12v, the solar panels want to be in the 18-20v unloaded range. Yes, this type of controller dios pulse the charging buss with pulses up to the 18-20v level when charging a battery that's low voltage.
These pulses are ok for many lighting and motor/pump requirements, but not for sensitive electronics.
I am going to try another $20 brand pwm controller just to verify mine isn't defective.
My work around I'm trying with the controller I have is to use a buck/boost regulator set to 12v out. It can make 12v at 5A from voltages ranging from 5v to over 24v. It's output buss seems to be clean and stable and not bothered by the pulsing 12-20v battery buss.
There are higher current rated buck boost on eBay.
For my use 5A is good for $5 as I'm only drawing .5A full time and every 3 min, 2.5A total for 10 seconds.
There is a waste of stored power with all this  m but that'll be reduced in a future design once it proves the concept. Probably more solar panels. I have a free source of damaged but fully functional panels that a swap outs, and a 40ft container roof to install them on.
I'll post as it progresses, my success and final solutions. Got to think positive ha ha.
Thanks again.

Hi, I'm new to this forum. I'm trying to sort out a inexpensive pwm 30A controller issue, my first time using solar. My setup will be 2 150w 18v panels in parallel charging 2 batteries in parallel to start. The batteries are AcDelco DC31's.
This setup will be used at a remote RC field to recharge lipo batteries via 12V smart chargers. The duty cycle of discharge vs charge time will be weekends used, not used weekdays pretty much.

It will also power a 12V link trasmitter that reports wx conditions and battery voltage. 2 A on transmitt and .5A idle when the suns up, .1 amp when the suns down at night.

As demand gets more, more solar panels will be added as well as an upgraded controller and added batteries. For now this is a prof of consent project to generate interest and donations.

So my issue, my bench setup. I'm ising a 12v@18A sealed battery. Not having the solar panels to use for a bench test I've tried usung a 20A 20v power supply or 2 3cell lipo packs in series. The condition I get is both the battery charge output and the 12v user port, both climb to 15-6v and then pulse up to 20v.
This in turn make a very dirty and too high of a 12v DC source across the sealed battery.
The controller led's showing panel, charge, and output ate all lit up.

In doing a search I'm reading that one should never use a substitute for the panel input. I also read where one should never connect power to the panel input without first having the 12v battery connected. Sadly, I've done both now. Accepting that I may have damaged the controller, I've ordered another one and brand under $20 to try again. This time by the book!
What would help is info on these pwm controllers as to what the 12v buss output should be like. I would expect it to not pulse to 20v or test at 15-16? Especially the managed output 12v port?
14.2v range yes, but not 15-16v and dirty pulsing 20v spikes?

If I've presented my setup and outcome ok, I'm hoping someone can tell me this isn't normal and I've simply trashed the controller with my bench testing? That or I'll have to spend above $100 for a better option.
Low voltage load cutoff is going to be a separate device and it will kill via a 120A 12v relay, any external container  RC battery charging by users. I will also have the option of controlling this relay from anywhere via my smart phone once a clean 12-14v battery buss is solved. The link transmitter can also receive commands. More on that once this is solved.
Thanks