Should I connect my inverter to the battery or to the charge controller?

Hi Permies,

I am going to buy the last piece of my solar kit: an AGM battery (12V, 100Ah) (the other elements are: solar panel 100W, a 300W inverter and a 20A charge controller), and I am now a bit confused about where to wire the inverter.

1) According to Renogy, you should NEVER wire the inverter to the charge controller, but to the battery.

2) According to this video it is better to wire the inverter to the charge controller.

What do you recommend?

Thanks

Depending on the intended load, many 12DC/120Vac inverters draw high amps. A small 700W microwave, for example, will easily draw 1000W.  That equates to approx. 77 amps @ 13Vdc.  Because of that, the inverter needs to be connected directly to the battery (including fuse).  The inverter and battery need to be as close to each other as possible, and you'll need a minimum wire gauge size of #4.

Hi Eugene,

thanks for your answer. The intended load is small: just LED lights and phone/laptop charging. I will follow your advise, and connect the inverter directly to the battery.

Greg

A good way to look at battery based system architecture, is that the battery is the hub or mainstreet. It is possible to lug or terminate to the inverter side of the cabling, and sometimes it could be viewed as a set of remote terminals. As long as your combined max loads do not exceed the inverter cable and terminal ratings itl be fine.

Much nicer usually, is a battery distribution panel to place everything or everything, but the inverter on. Rv marine fused terminal blocks are great. Battery terminals get crowded and there is a limit to wanting to stack lugs atop one another. Access to the terminals being out of the battery area is generally a convienience.

Depending on your layout....

https://powerwerx.com/three-hole-panel-mounting-plate

https://www.bluesea.com/products/5025/ST_Blade_Fuse_Block_-_6_Circuits_with_Negative_Bus_and_Cover

Correct me if I am wrong, I thought the charge controller had "Load" terminals, so when the batteries got to the discharge point, the "Load" could be disconnected by the controller to prevent over discharge from occurring?
If you connect the inverter directly to the battery this safety measure is by passed.

This then brings up another question, Charge controllers have an amp rating 10A thru 90A. That rating is usually in reference to the amps coming from the panels but isn't is also the amperage the charge controller load terminals can handle?

in the example above wouldn't a  80 or 90 amp controller work?
Or am I on the wrong track???

Jay

It could if the rating is sufficient. Most inverters, eventn the cheapest, have this feature (low voltage disconnect) already, have not seen one that does not.

I was under the impression that it was mostly a function of the quality of the controller that dictated this. When I first put my little system together it was 1 250w panel and 3 deep cycle batteries. I used a really cheap controller that just connected the panel to the batteries and had places for loads like 12v lights, then the inverter was connected direct to the battery bank. A few years ago I upgraded my system with another 250 w panel and 4 more batteries. I also then purchased a nicer controler that I was told was made to act as the hub, directing energy to the batteries, to loads like lights that could be set on timers/light sensors, and to the inverter. This way, if I wanted to draw from the inverter during the day the controller could direct that energy directly from the panels or from the battery bank depending on the state of the system as a whole. I have found that solar system retailers are often happy to go into excessive detail with you and really down to help you cobble together a system. Obviously they are going to try to sell their wares but once I have made it clear that I am sort of janky and shoestringing this thing together my local guys have been super helpful.

There is no magic to getting power from the panels to go directly to the loads/inverter. If the controller is connected to the battery or any other wiring terminals on the battery line, pv power will flow to everything connected, lights, inverter, all of it.

Its like airing all the tires on your car from one point (controller) if there were a tube connecting all the valve stems (wiring), they will be at the same pressure, even if you pop a hole in one of them (inverter).

Without the battery in line, anytime the load exceeds the available pv power, the load will suffer low voltage and high current and many electronics will just go off if undamaged (LVD protection) by this treatment.

Very possibly, if the controller had a wide margin for current capacity, but most equipment manufacturers aay explicitly, do not do this. The inrush or starting surge of inverters or motors are not good for these as far as i know

I can see some equipment combinations being zorch-o-riffic.

I would not try to rely on the inverter lvd for this either unless its adjustable. Monitoring of the system will ensure proper operation where low-low disconnect voltages are built in. Serious inverters have programable adjustable setpoints for disconnect and re-connect.

I'm new at this. My 650 W panels only produce 28A at 12 volts which corresponds to approx 325 watts at 12 Volts. It seems that my batteries are telling the controller to cut back on power from solar panels. I'm running a 5000 BTU AC from 4000w (8000 peak) pure sine wave inveter, but when I run AC controller is still shwoing only 28 Amps from solar. If I run the inverter from the controller will it bring in more power from solar panels? If I add another 325 wats into the system will the controller let me use it?

OK I have jumped in, got a mid-range 48V charge controller with the PV/Battery/DC out as six wire-ins on the bottom. The charge controller (in not terrible english) has settings to manage the DC output but they are quite simple- either off/auto(on with sensed battery V)/or timed off (like off at 10PM on at 6AM).  There is an undervoltage off in another setting. This seems like it is reasonable to attach the inverter right to the DC load. The whole system nominally will support 60A but I have limited it to 30A because I am using 10g wire and 40A fuses. 48Vx30A is basically 1500W so I should be able to quite safely draw 1000W.

The "battery" is my electric UTV, so the 48V is a requirement of the system, I probably would have gone 24V otherwise but this seems resilient since I can use the same equipment in the house if I need to.

I just want to see if this makes sense to people who have used the cheapo charge controllers. It hasn't burned up pushing 1000W from the panels to the UTV.

Atanacio Luna wrote:I'm new at this. My 650 W panels only produce 28A at 12 volts which corresponds to approx 325 watts at 12 Volts. It seems that my batteries are telling the controller to cut back on power from solar panels. I'm running a 5000 BTU AC from 4000w (8000 peak) pure sine wave inveter, but when I run AC controller is still shwoing only 28 Amps from solar. If I run the inverter from the controller will it bring in more power from solar panels? If I add another 325 wats into the system will the controller let me use it?

Many charge controllers won't go as high amperage as the numbers say they will because the charge profile wants to give only a certain percentage of c20 for instance during bulk absorption to keep the batteries healthy. Mine has settings for this stuff. You can change it in the profile sometimes but a very dumb controller may not subtract the load loss, and almost certainly won't if the batteries are separately wired to the inverter. How would the charge controller know?

I wouldn’t consider a charge controller that isn’t fully adjustable. The advanced menu is accessible when connected to a computer with Victron controllers. Ive watched videos but have not done it myself. I can adjust the setpoints and parameters of my Outback and Midnite Solar charge controllers without a computer. A computer is required to update the firmware on all advanced charge controllers that I know of.

Thanks, TJ Jefferson. That is exactly what I need to do. I'll have to read the instructions, but I think it allows that in the menus. I was afraid to go into it and blow something up, but I'm gaining confidence and your comment gives me some good sense that it's what I need to do. Thanks again.

I would think that connecting the inverter directly to the batteries would make it hard for the charge controller to know what the batteries are really doing.  Wouldn't the load on the inverter changing over time as the load on the inverter changes make the charge controller a less effective charger?

Both the inverter and charge controller are directly connected to the batteries with low resistance copper, brass, and stainless steel hardware. The electrons follow the path of least resistance from highest potential to lowest potential. A clean way to wire it is using cables to bus bars that connect everything together. They also make a neat place to bolt a shunt, fuse, etc.
I’d suggest following the directions written by engineers who designed the units.  Keeping the inverter cables as direct and short as possible to the batteries minimizes the inductance and resistance. As few connections as possible reduces resistance and failure points. Potential hot spots.
Some charge controllers such as Morningstar have voltage sense wires to the battery for fine tuning the charging voltages. It’s not required to use the sense wire when installing it. But I hook them up just in case it in helps. Some other advanced charge controllers such as Midnite and Outback have voltmeter calibration adjustment so the voltage can be adjusted using a calibrated digital multimeter. Most people never get around to it. Ive not calibrated mine.
 

The key functions of charge controllers are:
Protecting the battery from overcharging by limiting the charging voltage
Protecting the battery from deep and/or unwanted discharging. The charge controller automatically disconnects the loads...
Preventing the reverse current through PV modules at night
Providing information about battery state of charge.

I believe the two items, inverter and charge / controller are separate units and should be dealt with separately.
So the inverter is connected directly to the battery.Then the battery has no weak links along its path.

John,

Thanks for clarifying the specifics of the charge controller for the rest of us.

I am in the early planning stage of building a "battery generator" (really a deep cycle battery connected to an inverter) and this is all very helpful information.  I say I am in the early phase and by that I mean I am still designing and considering components I need.

I did recently build a little 12 volt "battery box" that will provide USB power and 12 volt electricity.  The main idea is I can recharge phones and such during a power outage.  Also, this build was a baby step in building its big brother.

Overall, this is a great thread for all of us considering building such a device.

Eric

I have always connected my charge controllers directly to the batteries in all the small systems I've made.  I have a larger system that was professionally designed and installed.  On that the charge controllers and the inverter each has it's own breaker switch allowing me to either separate the battery from the inverter, cutting of the flow of power out to my house while still letting the batteries get charged from the panels through the charge controllers, or I can shut off the charge controllers (for whatever reasons I've never found, maintenance maybe) so the batteries aren't being charged while still allowing the inverter to draw from the bank to power the house.  Given this I have to say they have the inverter designed to be wired directly into the battery bank.

Quote: “ The charge controller automatically disconnects the loads... ”. The load is the inverter. So, the inverter has to be connected to the load terminals if the charge controller is going to control the load (inverter). the charge controller cannot turn the inverter (load) on and off if inverter is connected to battery instead of to the charge controller load terminals.  Connecting the inverter (load) directly to battery, bypasses the charge controller, thus the charge controller cannot control the load(inverter).  

Ok,  I saw this post resurface recently and I thought I would comment. I am (still) planning my build around a YouTube video that nicely describes the build process.  The video clearly shows the inverter connecting to the battery.

I will update this post with the video link shortly.

Edit:  The video link is HERE:   https://www.youtube.com/watch?v=a6KqZ80EEc4

This particular video is part 4 of a series and focuses on showing the wiring.  If you are really interested, the whole series is really worth a look.  It is very instructive and it giving me most of my information about how to build my own unit.

Eric

Like charging a laptop while charging with a low capacity charger both keep the battery charged and the display on and CPU running. You do not see any flickering in performance. The BMS in the Li-ion battery pack uses the excess power of the charger to charge and if to low the BMS starts discharging the battery just a little bit to compensate. The specs on solar and BMS says that it is not possible to charge and discharge at the same time. That said it would mean a lamp connected to the controller load leads would basically stop the charging process to the BMS and the battery. At night, Okay but at daytime when the sun is shining on the solar panels, what then? Same as with the laptop, the excess is used to charge the battery through the BMS, all power from the solar panel is going strait to the load. It is a process of filling up a bucket and it does not matter from which it is filled. The BMS takes always care of dangerous discharge levels, it's what it does best. The solar controller distributes the power accordingly, it's what it does best. The load (Lights) leads should be rated to draw current at what the solar controller is rated at for solar current input, basically pass through.

The vid in your first post is exactly correct. he covers when you would and when you wouldn't connect to the load output of the controller.

in your case you're not going to pull 300W from the load output because it's a 20A controller and 20A x 12V =240W so that's the most the inverter will draw. Additionally what the inverter gives out will be somewhere between 120W & 240W depending on its efficiency.

A square wave (sometimes called modified sine wave) will be closer to 120W.
A pure sine wave will be closer to 240W.

Inverters to battery directly, not charge controller. Stable power, avoids overload.