diy solar

diy solar

Why does Will Prowse suggest not using load output ports on the EPEVER MPPT Solar Charge Controller?

GGameBoy

New Member
Joined
Feb 22, 2021
Messages
63
I have watched many of Will Prowse videos on YouTube and he never recommends using the load ports on the solar charge controllers and in fact he suggests against it but leaves no explanation why, At least that I have seen.

My understanding of load ports on solar charge controllers is that the power comes directly from the solar panels to the load and thus potentially saves battery cycles and increasing battery life. Is this correct? If so I can’t see why he does not recommend using the ports.

I know on this 40 amp mppt solar charge controller the load ports r only 20 amps at 12 volts but that’s still up to 240 watts per hour that is not being cycled by the batteries.
 
I have watched many of Will Prowse videos on YouTube and he never recommends using the load ports on the solar charge controllers and in fact he suggests against it but leaves no explanation why, At least that I have seen.

My understanding of load ports on solar charge controllers is that the power comes directly from the solar panels to the load and thus potentially saves battery cycles and increasing battery life. Is this correct? If so I can’t see why he does not recommend using the ports.

I know on this 40 amp mppt solar charge controller the load ports r only 20 amps at 12 volts but that’s still up to 240 watts per hour that is not being cycled by the batteries.

Your understanding is incorrect. Load ports are battery voltage. They vary with the battery. If there is solar available, then the device's load on the battery, and the subsequent drop in voltage will be compensated for by the solar. This behavior is IDENTICAL if the load is attached AT the battery or AT the load ports.

Think of load ports as a different place to connect to the battery that may have additional programmable functions.

A load port has the following advantages vs. direct connect to battery:
  1. May offer a timer function to allow the load to be turned on and off on a schedule.
  2. May offer low and high voltage disconnect to protect the device from high voltage or the battery from excessive discharge through the load port.
The biggest disadvantage is they are generally limited to a much lower current.
 
I only use the load port on my EPever to program on/off low current LED outside lights. 1-3a X 2 it also gives a dusk to dawn feature if prefered.
My other CC is a Victron 100/30 and there are no load ports, so I use an inexpensive small timer connected direct to the battery for lights on that CC.
 
Your understanding is incorrect. Load ports are battery voltage. They vary with the battery. If there is solar available, then the device's load on the battery, and the subsequent drop in voltage will be compensated for by the solar. This behavior is IDENTICAL if the load is attached AT the battery or AT the load ports.

Think of load ports as a different place to connect to the battery that may have additional programmable functions.

A load port has the following advantages vs. direct connect to battery:
  1. May offer a timer function to allow the load to be turned on and off on a schedule.
  2. May offer low and high voltage disconnect to protect the device from high voltage or the battery from excessive discharge through the load port.
The biggest disadvantage is they are generally limited to a much lower current.
Thank you! That explains a lot.
 
The ‘load’ output terminals can be connected to a NO or NC 12V relay to utilize the ‘sunset’ or low voltage cutout features of certain charge controllers.
That could be a handy way of doing a number of switching tasks automatically.

I think the primary reason to not use the load outputs is the low 20A capacity still needs fuse panel to be safe. A decent fuse panel could handle / distribute 40, 60, or 100A...
A second reason is why would you want to put 20A of load that’s potentially creating heat in the charge controller?
A third obvious reason is with an inverter the amp load is huge. A 1200W inverter running a 600W coffeemaker for example is at least 50A. That needs fat cables- big gauge 250A cables - (I used 2/0) hooked to the battery, not 10ga wires from a charge controller.
 
The ‘load’ output terminals can be connected to a NO or NC 12V relay to utilize the ‘sunset’ or low voltage cutout features of certain charge controllers.
That could be a handy way of doing a number of switching tasks automatically.
I do this as a crude way to energize an opportunity circuit. LOAD -> [timer long enough that the Vabs is likely to be achieved] -> relay
 
The only reason for the load port is so the s.o.c. can “see” both the input and output of the battery, well you can program things to turn and off at certain times, etc. but for most uses connecting directly to the battery gives less loss of running through the controller again. On the tracers the s.o.c. display will not be able to accurately measure the load current and will allways say the battery is at 100% no matter the true state of charge. This is probably true of most inexpensive controllers of any type. The truly good controllers all have remote shunts at the battery to see the true current going into and coming from the battery.....this will not be found on the inexpensive controllers.
 
My tracer seems to know state of charge fairly closely??????
(I don’t use the load output either, and I’m pretty sure load isn’t factored in the mppt programming)
 
I'll just toss this out there for consideration. My understanding is that the load out terminals on my charge controller (Renogy Rover 40A MPPT) are a regulated 12 volt output. Voltage sensitive devices connected directly to the battery could be damaged when the battery voltage increases during equalization, etc.
 
I have sold off all my offshore controllers and replaced them with Midnight and Morningstar, they were problematic. I had 4 early EPSolar AN1530 and 1 AN1040, could be wrong on exact model number, load outputs were very high EMI/RFI, very much worse than the MPPT front end. Too many fails and shutdowns. Failed to fully charge charge true solar lead acid cells, not settable. The load output is after the MPPT front end, its basically switched battery power but radio noise completly blocked the FM radio spectra.....to me that is not tolerable....gotta have sometunes..

Note these were EPSolar before they reorganized into EPever bankruptcy, I hear, didnt care to follow that.
 
My tracer seems to know state of charge fairly closely??????
(I don’t use the load output either, and I’m pretty sure load isn’t factored in the mppt programming)

if you change the LV disconnect voltage (eg 12.2v) it will show the battery as being a lot lower in it's state of charge reading

The LV reconnect has to be lower than the Boost reconnect
 
Last edited:
snoobler has defined it exactly, that is what i found with mine, My system outgrew the EP Solar units, I now have two 94 amp Midnight Classic 150’s and two Midnight Kid’s,
 
I only use the load port on my EPever to program on/off low current LED outside lights. 1-3a X 2 it also gives a dusk to dawn feature if prefered.
I use the load ports for something similar. At my cabin I disconnect the batteries to everything but a small SCC that just keeps the batteries topped up while I'm gone.
When I get to the camp after dark, I have some small (3× 4w led) light strips that come on when the door to the utility room is opened so I can see to turn everything else on.
The load port is the only port on the array that has power available when everything is secured.
 
Load ports are battery voltage. They vary with the battery. If there is solar available, then the device's load on the battery, and the subsequent drop in voltage will be compensated for by the solar. This behavior is IDENTICAL if the load is attached AT the battery or AT the load ports.
Forgive me citing this old post, but I have a question that I've been asking for some time (a little of it here, mostly to Google searching). However this above strikes me as not fully correct. This is how I see it:
  1. Load ports may be battery voltage (they may also be from the panel as the poster indeed goes on to suggest)
  2. the behaviour is not in my experience identical comparing if the load is attached to the battery or to the load port. I have definitely seen this in the actual measurements I've taken on my setups. Perhaps this is controller dependent.
I've noted that a "shunt" is often used here (on this forum and perhaps in this "domain") to simply monitor the current in and out of the battery. Its been a while since I studied electronics at college (electronics engineering, but did my specialisation in computer systems) but what is called a Shunt here is also known to me as a Ballast Resistor.
1645215374769.png
I propose that inside a Solar Charge Controller (SCC) {perhaps especially an MPPT type} which has a load uses a Shunt internally to ramp up the panel to compensate for the load in as much as the panel can (or to the limit of its rated capacity).

I believe that this is actually a more ideal way of handling minor loads (loads being a demand for electricity) on the system because then the SCC can continue charging the battery in a more "ideal way" to fully groom it to go into the night. If it goes into the night without this situation then it is going to lead to the battery perhaps not ever reaching a float level (thinking Lead Acid here, not Lithium chemistries) which will lead to battery plate sulfation.

It seems to me that without a shunt to know whats going on a in terms of load, a SCC can not compensate for that change in power demand by simply looking at the voltage drop of the battery (or if there is please do share that with me).

As an aside it was mentioned somewhere here that it is not ideal for the SCC to actually manage this because "why put another source of heat in the SCC". However MOSFET switches are very effective, Let me cite a reasonable source (here) :

Another approach to switching DC loads is by employing a transistor. Modern MOSFETs (Metal Oxide Semiconductor Field Effect Transistors) can switch very fast, don't suffer arcing problems and require negligible energy to switch states. They are also a fraction of the size of a relay. As an example, an Infineon BSC030P03NS3GAUMA1 MOSFET is only 5mm wide, 6mm long and 1mm high yet it can switch 100A and would only burn 23W doing so, it barely even needs a heatsink!

So if a SCC has a MOSFET in it to do the load switching of a mere 20A it strikes me as a good design feature. Just working with that 20A limit (no reason why it can't step up) a 48V system will give a load of 980W (at its limit) which in an RV or smaller off grid situation is nothing to be sneezed at. Sure it won't power a fan heater (2000W) but will amply power a domestic fridge of reasonable size (I've measured my mates two door fridge freezer and it seems to peak at 600W and consume about 1kWh over 24 hours).

So unless I'm misunderstanding things it seems we should be using shunts in our systems to feedback into the SCC the actual demand so that they can (panel capacity and weather permitting) suck that load current out of the panel and keep the battery on its charging trajectory by stepping up the amps its providing. Indeed this is something I think I see in the more complex arrangements that Victron offers in products like this.

Anyway, if @snoobler isn't active anymore, perhaps someone else has thoughts on this, because while this seems right to me the devil is often in the details I didn't know.

Thanks
 
However this above strikes me as not fully correct. This is how I see it:
  1. Load ports may be battery voltage (they may also be from the panel as the poster indeed goes on to suggest)
  2. the behaviour is not in my experience identical comparing if the load is attached to the battery or to the load port. I have definitely seen this in the actual measurements I've taken on my setups. Perhaps this is controller dependent.
You are confusing yourself with the idea that electricity is spatially volume oriented. All the draws on a battery or inputs to a battery (bank) are buffered by the battery capacity and influenced by the portion of the circuit they are located. Whether the panel (s) contribute or not is irrelevant to the weighting/buffering of the battery. Electricity doesn’t pour in or out like water from a jug b
 
So unless I'm misunderstanding things it seems we should be using shunts in our systems to feedback into the SCC the actual demand so that they can (panel capacity and weather permitting) suck that load current out of the panel and keep the battery on its charging trajectory by stepping up the amps its providing. Indeed this is something I think I see in the more complex arrangements
Quality SCC’s do have shunts at the battery , even quality inverters have them as well, All my Midnights have them as well as my Magnasine Inverter/charger
 
This discussion appears to touch on my major concern of the following situation:

A “high” voltage panel is feeding an MPPT controller. The controller is managing that high voltage and up-converting the amperage flowing into the battery. I assume while everything is working OK, the voltage supplied to the LOAD OUT terminals will be at or near battery voltage. But what happens if the controller is disconnected from the battery…intentionally, or via blown fuse, or controller failure. Will the panel voltage be supplied to the loads? Or the (edit: battery voltage…not panel voltage)?

In my case, newly-acquired 52 Voc 250w panel, nominal 12v system, proposed Victron 100/20 MPPT (not implemented yet). Today I have a switch that directs the controller output to one-of-two battery “banks,” and another switch that connects-disconnects the solar panel to the controller. But my current system is lower voltage (22v Voc 160w panel, Morningstar 15a PWM Prostar). Flipping the output switch between the 2 battery banks may, or may not, reset the controller. If I do it fast enough, nothing really happens, except the loads and solar are connected to the different battery bank. If I “hesitate” while flipping the switch, the controller goes through its restart sequence (<5-seconds). I try to discipline myself to flip off the solar connection before flipping the “battery” switch, but I have been known to omit this step.

So my concern comes from 2 directions: I am now dealing with a panel voltage that could more easily damage my loads (or batteries), and I don’t know that the Victron is as robust, or as protective, as the Morningstar might be. Thoughts?
 
Last edited:
Back
Top