Adding solar to an old on grid homestead

Homesteading & Country Living Forum

Help Support Homesteading & Country Living Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Well the first day of using the new control system is over (it's sunset here right now), although we have had an overcast day I was able to utilize 2kWh's and the battery bank is still at 12.6 Volts. According to my Kill-a-watt meters the maximum load that I had on the inverters was just under 3kW, I think the average load was closer to 1000 watts. The most in-put that I saw from the solar charge controllers to the battery bank today was about 700 watts, so the system is doing what I wanted, it is allowing me to get some usable power even when the arrays can't provide enough power to meet all my needs... I would call that a win.

The other nice thing is that because I am removing the load before the inverter reaches that low voltage warning, I don't get those annoying low voltage alarms. This control system should help me maintain my batteries for a longer time to....
Furnaces and AC systems have delay to prevent them from switching too often.

Have given any thought to preventing your system from repeatedly switching under conditions of marginal sun ?

Just thinking

Ben
 
Furnaces and AC systems have delay to prevent them from switching too often.

Have given any thought to preventing your system from repeatedly switching under conditions of marginal sun ?

Just thinking

Ben
There are 2 factors that delay the switching, first when the load is switched the battery voltage is above 12.8Vs and it stays on until it drops below 12.2Vs, depending on the load and the amount of power being provided by the solar panels it can take quite a while >20 minutes before it switches back over to grid. During the charge phase the solar cells have to bring the battery bank from about 12.1 volts up to above 12.8 volts, which can take quite a while in itself. Plus the photo cells need several minutes of exposure before they switch off the grid. So in poor solar generating conditions there are less than 2 cycles of switching events per hour. The controllers I have can be set to a lower low voltage setting 11.4 V, this would extend to period of inverter power but then I might hear that low voltage alarm beeping on the inverter(I don't want that). I might be able to add a rheostat to the DC control circuit to lower the observed voltage at the controller, then I might be able to not switch off grid until the battery bank was above something like 13.2 Vs and then not switch back onto the grid until the battery bank dropped to 11.8 Volts, this would give me much longer on and off cycles and work the batteries a little harder. But, it would require a little more playing with circuitry. I have found some DC coil operated AC power switches rated at 10 or 30 amps, if I could find one rated at 15 amps it would fit the build targets.

This is just one more experiment.

Oh, I did check how much grid power I used today for those circuits (4.4 kWh), so today these circuits used a total of 6.4kWh, 31% of that was supplied by the solar array on a partly cloudy day.
 
Today we are having a real snow day, the solar panels are covered and producing less that 10 watts. The battery bank is at 12.4 volts, but if the wind comes up there is a high likelihood of losing grid power all together. So I decided to manually over-ride the inverter on relay and put a small charger on the battery bank, this way I can push the battery bank up to 100+% of capacity and be ready for anything that comes my way. If the grid were to go down under these conditions it will still switch to the inverter/battery bank automatically, but I would want to off-load some of the less critical demand to extend my coverage.

It occurred to me that all of those grid tied house top systems are only putting out less than 1% of normal and the grid demand is up due to the low temperatures and "green" electric heating. I wonder where the "extra" power is coming from to cover all the demands? I'll bet that every natural gas turbine across the country is fired up right now to make up for the drop in "green" energy production. The nice thing about having ground level solar arrays is that I can sweep them off when the storm passes, if they were on the roof top I would have to wait a couple of days to have my backup power back up....
 
Today we are having a real snow day, the solar panels are covered and producing less that 10 watts. The battery bank is at 12.4 volts, but if the wind comes up there is a high likelihood of losing grid power all together. So I decided to manually over-ride the inverter on relay and put a small charger on the battery bank, this way I can push the battery bank up to 100+% of capacity and be ready for anything that comes my way. If the grid were to go down under these conditions it will still switch to the inverter/battery bank automatically, but I would want to off-load some of the less critical demand to extend my coverage.

It occurred to me that all of those grid tied house top systems are only putting out less than 1% of normal and the grid demand is up due to the low temperatures and "green" electric heating. I wonder where the "extra" power is coming from to cover all the demands? I'll bet that every natural gas turbine across the country is fired up right now to make up for the drop in "green" energy production. The nice thing about having ground level solar arrays is that I can sweep them off when the storm passes, if they were on the roof top I would have to wait a couple of days to have my backup power back up....
This is why I built our solar array at ground level, thing is, it seems that even though there may be a layer of snow on the panels that if the sky clears the nest day there is heat generated from the panels producing power that causes the snow to melt and slides off, also, for whatever reason, the panels never seem to accumulate any dust and I have not cleaned them since installing them in 2018.
 
My control strategy while unorthodox has been very effective at maintaining my battery bank. The average battery voltage has been staying at about 12.4 Volts with excursion ranging from 11.9 to 13.2 volts. I have been testing relay circuits to eliminate the light and photovoltaic elements, but so far I have not been satisfied with the replay performance in early testing.

I am seeing a much higher utilization of the power I am generating (I'm using multiple kill-a-watt meters to track performance), so that's a good result.

Maybe I'll take some pictures and schematics down to @Bacpacker next month, let his "smart guys" figure an elegant solution... :)
 
My control strategy while unorthodox has been very effective at maintaining my battery bank. The average battery voltage has been staying at about 12.4 Volts with excursion ranging from 11.9 to 13.2 volts. I have been testing relay circuits to eliminate the light and photovoltaic elements, but so far I have not been satisfied with the replay performance in early testing.

I am seeing a much higher utilization of the power I am generating (I'm using multiple kill-a-watt meters to track performance), so that's a good result.

Maybe I'll take some pictures and schematics down to @Bacpacker next month, let his "smart guys" figure an elegant solution... :)
Bring them on. But I doubt we'll improve on what you've already proven to work.
 
I have been gathering DC controlled AC relays, project boxes, and breakers so last night I started building my control box. I had a little trouble getting the relays to trigger properly until I realized that I had wired them in series and they needed the full 12Vs to operate, quick fix. Set it up and tested it last night (no load) and it worked well. I was wanting to add the ability to provide some grid-supplied charging if the battery bank gets too low and I found/ordered an adjustable low alarm control relay that might do the job. That relay might have other uses and I may spend some time playing with it to see what's possible. Anyway I may end up re-building the system (might just make a new control box) to provide the expanded capabilities.

With the new system working, I think I can adjust the low voltage inverter load cutoff (my new box) to just above the Inverter low alarm voltage which would allow me to use a little more power out of the battery bank (without hearing any beeping :) ). I think I will wait until I have a full day where I can babysit the system before I switch everything over in earnest.

But things are getting much closer to what I had originally envisioned. A Solar Powered emergency power back-up that automatically provides surplus power to supplement my grid demand. An integrated system that also provides battery monitoring and protection to ensure that the battery bank is always ready to meet my emergency needs.
 
Back in January I had developed a plan to increase my utilization, but I was never able to fully execute it. I had most of the parts but just couldn't visualize how to make it all work together. I ended up opening up an extra control module to figure out how to control the trigger's observed voltage while still providing full voltage to the relay switch. I found that the control circuit shared a ground with the load circuit, but only the hot was switched. I decided that I would provide the load an independent ground and added a switched clipping circuit to the control circuit's ground leg, this should allow me to provide full voltage to the relay while making the controller think that the battery bank voltage is lower than it actually is.

So today after work I decided to go back to building a mock-up of my control system, the current one has just too much "bounce" for my taste. The algorithm was that the loads would come on the inverter once the battery bank exceeded 12.8 volts and then shed at 12.1, the control unit would not reset until it reached 12.8, but the load will draw the battery voltage down but then bounce back 0.4 volts as soon a the load is shed so it's really only covering about .4 volts of operational range. The control module has fixed voltages where is will shed loads, 12.1, 11.4, & 10.8 but my inverter alarm goes off at 11.8 volts (so I don't like hearing beep beep beep so I don't want to go to 11.4).

Anyway, I found a low battery alarm module and rigged a circuit that would "trick" the control unit into thinking that the battery voltage is lower than it really is. So now I flipped a switch to make the control module want to shed loads at 11.4 volts and I can dial in the low module alarm module to trigger at 11.85 volts, at which point it makes the control module "think" that the battery bank is at 11.2 volts and shed the load until the battery bank is back up to 12.8 volts. During this time the circuits reset themselves for the next cycle. The result is that I can double the amount of power that I pull out of the battery bank during each duty cycle.

I added voltage gauges at critical points in the circuits I could dial in and monitor everything this afternoon, it all works as designed on my trial control box. This weekend I will move it over to the actual control box and let it start controlling the AC loads... :) I have 3 10 amp load circuits and with this arrangement I can actually off load them at different voltages. So it should be a much more stable arrangement.
 
Last edited:
So I checked A-zon to see when I will get the second low alarm module to finish my build and I learned that it will not be delivered for another 2 weeks. You get used to just in time delivery and then it doesn't arrive in time for what you want :(
 
Okay, I just checked again and I saw a digital control module that I can rig to replace 2 parts of my system, the price is lower and it may simplify things a bit (assuming I can program the thing). I will provide an update after I try one out this weekend. I would really like to get this thing nailed down so I can direct my attention to my next project...
 
Our son installed an instant propane water heater on his trailer, he says it's very efficient, the only problem is that it's mounted outside but he's researched and found a compact propane instant water heater that will go where the original propane water is mounted, the main problem with the instant units is their rather costly. As to solar, it's been mentioned to eliminate connecting any equipment running on 240 VAC, the Magnum Energy inverter/charger we have is 120/240 VAC split phase which I have decided up the load evenly but if I needed to I could run a deep well water pump from the inverter, thing is I'd only use it to fill a storage tank during a good sun day when the batteries would have a good chance to recover their charge. As to batteries, our first set of 20 six volt golf cart batteries we bought in 2013 provided five banks of four batteries each for 1,025 amp hours, up until mid 2019 these batteries were charged by the inverter/charger only, in 2019 I completed installing the solar array, eight 195 watt panels and connected them to the batteries through an MPPT solar controller, after using that type of controller, it's the only type I'd ever use because you can run on higher solar voltages, even on cloudy, rainy and sometimes snowy days, the batteries will get a charge. The first set of batteries took a hell of a beating, first due to frequent grid power outages and very large discharge and recharge amperages and after over 9 years of this the batteries were needing to be topped off with distilled water every month and so in 2023 it just happened that Costco had enough GC2- 6 volt golf cart batteries come in with the same manufacture dates that we ended up getting twenty four 210 AH batteries, adding one more bank to our system raising the total AH to 1260 as it turns out it seems like the solar system loved the extra bank of batteries and it took a year before I had to add distilled water to top off the batteries and even then it didn't take as much water as it did with the old batteries, sometimes it was up to 6 gallons, the new batteries only took about 3 to 4 gallons. Dealing with deep cycle lead-acid batteries is not always fun, so if you go that route design the battery banks to have easy access, where ours are are under a counter where the Inverter/charger is mounted and it always seems that my butt muscle get really sore the next day after servicing all the batteries, being as that I'm 81 doesn't add to the joy of that task. There are automatic battery filler systems available using small tubing and automatic shut off filler caps, it seems that they are only available on the internet, they are also spendy, so I just went with the old fashioned 1/2 gallon automatic hand filler. I also bought our own water distiller, distilled water is generally pretty low cost but it's good to have on hand.
 
I'll try a play by play.
Here is my DC test bed: power supply, high/low voltage alarm module, voltage load control module, and DC powered AC relay.
PXL_20240309_161059567.jpg

Circuit on "Normal" 13.5 Volts:
PXL_20240309_161127654.jpg


Voltage Low Normal all greens, 11.7 volts:
PXL_20240309_161406563.jpg

Low Voltage condition Trip light amber, Low Voltage disconnect red closed, relay off: Note the differences in the voltages shown: 11.6 vs 10.9

PXL_20240309_161500100.jpg

When the relay on the right is in the "off" condition the AC/Grid power comes on and the automatic transfer switch bypasses the inverter output.
I will have 3 of these circuits, 2 for loads and 1 for charging, because I can dial in the voltage on the beige module, I can have the AC/Grid charger come on just as the inverter low voltage alarm is about to go off, I can also control how much charge from the grid I take before it disconnects.

I'm bummed that I couldn't show you this in a video...
 
Here is my combined AC/DC test bed with 3 circuits, relay lights are green, no AC at the outlets.
PXL_20240309_133743374.jpg

When the voltage is low enough the AC turns on to bypass the inverter: note the far outlet is still off, it's the charging circuit:
PXL_20240309_133724292.jpg

Here I have turned on the Charging circuit: and you can see the small charger in the background showing 12.1 Volts:
PXL_20240309_134048649.MP.jpg
 
Hi @Neb, The DC Voltage Monitor module has a time delay adjustment and the DC Low Voltage Disconnect has a short built in delay. Adding the extra module allows me to fine tune the low voltage events lower (Closer to the inverter alarm of 11.5 Volts) and the high voltage events (resets at 12.7 Volts).

The old system (DC Low Voltage Disconnect alone) only had fixed reconnect (12.1 or 12.7 Volts) and disconnect voltages (10.0, 10.7, 11.4, or 12.1 volts) making the safe usable voltage range only 0.6 volts, now I can adjust that to about 1.1 volts, greatly expanding my usable operating window. Using 3 control circuits I can shed loads incrementally to make it a softer transition as the battery bank approaches the low voltage limit (that I have chosen).

So in theory, as the voltages of the battery bank change so can the loads on the inverter.
Here is my updated Control Algorithm:

>12.7 volts all the loads re-connect, total load 3300 watts (max) - Normal Service ~ 1500 watts +/- 500 watts
12.7-11.9 volts - maximum load (up to 3300 watts)- Normal Service ~ 1500 watts +/- 500 watts
11.85-11.65 volts - Load reduced to (1650 watts Max)- Normal Service ~ 750 watts +/- 300 watts - Even if the voltage comes back up this Load will not re-connect until battery voltage is >12.7 volts. So the Solar can charge the system back up while supplying partial loads, even in winter.
<11.65 volts - All loads removed from the inverter (Battery still sees 1 watt draw, measured)
<11.55 volts - the AC charging circuit comes on and remains on until battery bank reaches 12. 6 volts, Inverter Loads remain disconnected during this period.

Voltage from the Solar charge controllers is required to bring the system back on line >12.7 volts.
The control system is designed so that if the AC power goes out (Grid down) all loads are transferred back onto the Inverter regardless of battery voltage. This is a result of my using the DC Voltage circuits to control the AC supply to the Loads using single circuit automatic transfer switches.

Note every DC and AC line has it's own fuse or breaker limiting current to my Maximum Values, My Inverter is a 4000 watt unit so I am only using 40% of my capacity. Total DC current draw (my Max) is 275 AMPs, max current draw at each battery is 23 AMPs. Typical in-use observed values are 125 AMPs at the pack and 11 AMPs per battery. The battery bank usually hoovers at about 12.3 volts +/- 0.7 volts. The batteries are 105 AMP hour units, so I have a max of about 9 hours usable energy stored in the pack.

Most of my loads are daytime events, so the Solar is charging most of the time, ~ 1500 watts winter, 2500 watts summer.
On cloudy days, once the battery hits the low voltage limit all loads are shed and the AC charger comes on to maintain the batteries at about 12.6 volts.

Simple.
;)
 
Last edited:
After work yesterday I did some work in the greenhouse/battery room, I added an induction current monitor on the charging lines and finished installing my new control arrangement. I only had 2 of the battery monitor alarm modules, but I put them to good use. I finished just as the sun was starting to set (but before the solar panels stopped producing), it was nice to see how much power was going in and coming out at one place (at the same time). When it was all working I had 300 watts going in and 500 watts coming out, the battery bank was at 12.7 volts. At lunch time yesterday I noticed that my battery bank was up at 13.5 volts so there was definitely some charging going on.

I still have some DC wiring clean up and then I can move on to my next project.

I plan to add a wooden shelf under my large solar array, it will be a cool weather crop growing table, I figure the shade and indirect light will work fine to allow me to keep growing spinach long after the stuff in full sun has bolted. :) When you got no space you have to think 3 dimensional. I plan to use the ebb-flow system I developed for the greenhouse for watering, it uses simple seed starter trays, bulk head fittings, and 1/2" pipe. I may add a gutter at the bottom of the solar array to catch rain water to fill the water tank that will be below the system. For a water tank I am using the big black storage boxes from HD (the ones with the yellow lids). I'm actually thinking of going to the largest version for several outdoor container growing solutions.

I did add 2 solar led lights to the garden area, they are motion sensor lights and give just enough light to walk around at night.
 
Update:
Today it's overcast with some rain, my solar is providing about 200 watts, the load being supplied from my system is 350 watts, so I am draining about 150 watts out of my battery bank. The reason why the load is so low is once my control circuit noticed that the battery bank (under load) had dropped below the set point part of the load was shed to extend the life of the battery bank. After running all night with no input, the system has gone through 2 load shedding events where 750 watts were removed each time. Now the system is at the minimum output mode and it will remain at that level until the minimum voltage set point is reached, then all loads will be removed. But even then the battery bank will be above 11.5 volts, so I could pull some more power out of it if the Grid were to go down.

The greenhouse battery bank has about 16,000 watt-hours of storage so under these conditions I'm still in pretty good shape.
I have an additional 16,000 watt-hours of indoor storage for emergency appliances, this is only utilized after the grid is down.

FYI, yes I do wear a belt and suspenders :)

Additional update, at 11:00 the solar charging has now exceeded the low load demand wattage and the battery bank is being recharged. It is still overcast, so it's not charging at full capacity but the system is now working as planned.
 
Last edited:
Looking back, it's been almost 3 years since I started playing with solar, back then I had 200 watts of solar and 2400 watt-hours of storage. I have learned a lot and am now able to use my solar power on a daily basis. The lead-acid battery bank is over 2 years old and it is still doing a good job, the LiFePO4s are just over a year old and have very few cycles on them (they are indoors for emergency use). I have 2500 watts of solar installed, but I can see that in a SHTF event I would really want at least 5000 watts (twice that would be nice) to cover just a few small loads ~1500 watts average load.

I have also come to realize the biggest problem for the green power grid. Solar is un-reliable from a base load perspective, about 1 day in 3 here is a poor solar power day. To make up for that the grids relying on solar power need to have natural gas turbines generating capacity equivalent to their installed solar capacity (for night time or overcast days), or they need sufficient energy storage to cover the load for 48 hours. So for a 1 MW solar system to be reliable they need 1 MW of solar and 1 MW of natural gas turbine, or they need 8 MWs of solar with 48 MW-hours of energy storage (The 8 MWs of solar is to allow the energy storage to be charged in a typical 6 hour high solar power day). This is why electric rates will need to be going up to achieve the green power goals.
 
I have also come to realize the biggest problem for the green power grid. Solar is un-reliable from a base load perspective, about 1 day in 3 here is a poor solar power day. To make up for that the grids relying on solar power need to have natural gas turbines generating capacity equivalent to their installed solar capacity (for night time or overcast days), or they need sufficient energy storage to cover the load for 48 hours. So for a 1 MW solar system to be reliable they need 1 MW of solar and 1 MW of natural gas turbine, or they need 8 MWs of solar with 48 MW-hours of energy storage (The 8 MWs of solar is to allow the energy storage to be charged in a typical 6 hour high solar power day). This is why electric rates will need to be going up to achieve the green power goals.
Thank you for being honest.
I see that every week down here where there are zero solar panels.
People get caught up in the hype and think if they just get some panels, they can 'cut the cord' and have free power for life :D.
If you live in the desert with 12-months of scorching sun per year, maybe so...
The vast majority of normal places, no. Solar doesn't work nights either.:(
(Y'all hurl your sticks&stones at me, not him. I prefer the label "denier"🤪)
 
Thank you for being honest.
I see that every week down here where there are zero solar panels.
People get caught up in the hype and think if they just get some panels, they can 'cut the cord' and have free power for life :D.
If you live in the desert with 12-months of scorching sun per year, maybe so...
The vast majority of normal places, no. Solar doesn't work nights either.:(
(Y'all hurl your sticks&stones at me, not him. I prefer the label "denier"🤪)
During winter we get a lot of fog, snow and cloudy days. Unless the solar panels are covered in snow, we still get some power out of the panels, not much but some. My solution was to add more solar panels. I also have a backup diesel generator with an auto start. When my battery bank drops below 23.9 volts the generator automatically starts. I have it programed to run for 4 hours, or to Float, whichever comes first. We have seamless power with this system. Now that we're getting more sun and longer days, the generator seldom needs to start. Solar may not be perfect, or work for everyone, but in our situation and location it works out very well. I'm very happy with it. Of course at our location, if we wanted electric, there wasn't any other option than solar. We're too far from grid power.
 
Today is the first sunny day since getting my control system 75% completed (I'm missing a component for fine tuning 1 load circuit), and it has been a real eye opener.

First Last night, the loads dropped off 1 by 1 as the battery voltage dropped, but after each load was removed the measured battery voltage recovered slightly, so there was a significant period of time between events. By midnight they system had shed all it's loads and the measured voltage was just above the inverter low voltage alarm value.

Then this morning as the sun came up I noticed that as the voltage came up one of the loads came on, causing the measured voltage to drop, this delayed the time before the next load came on. As one of the arrays got full sun there was sufficient voltage to bring the second load on, the observed load at this point was 750 watts and the charging was at 800 watts. Finally around 13:00 all of the solar panels were seeing full sun and the final load circuit came on, the measured load was 1600 watts and the charging was showing 1900 watts (one array is still shaded), battery bank voltage is 12.9 volts and climbing...

I am measuring the voltage at 3 location, on the charge leg fuse to the buss bars, across the center battery leg across the buss bar, and across the inverter terminals, the control circuits are based on the voltage across the inverter. There is a small voltage difference at each location with the inverter showing the lowest voltage. I have matching induction current pickups on the charge leg and on the inverter leg of the system, so I can measure the DC voltage, current, and wattage at each point.

The voltage difference changes with the load due to the resistance in the wires between the buss bars and the inverter. The slightest difference in "Load On" voltage set points for each the loads control circuits has a big impact on when the loads come on. The first load came on almost as soon as the solar panels started generating (~50 watts) more power than the base load(~3 watts), it was almost an hour later before the second load came on, and the third load was close to an hour after that.

The one thing that I am seeing is that I could add just a little more load to the system at each stage of the control strategy.
 
I'm looking around some sites here for off emergency options. Would appreciate your opinion on this for indoors. I have a generator for the pump and to boost the freezers. I have a small solar charger for phones, torches etc. This would be to plug in the wifi, because for some stupid reason the pack doesn't take batteries and low power stuff.
https://offyourgrid.ie/product/ecof...MI0L2psfePhQMVWpJQBh1nlQs6EAAYASAAEgLRK_D_BwE
 
Modern life off grid is much different than most think. I say that as I sit here in the bathtub soaking while the clothes are washing in the machine while I'm on the tablet. Such a hard life, not! LMAO
It is however a bright sunny day and the batteries are full up the water is also Heating
 
@Magpie, I looked at the link you provided and it has a little of everything, an 800 watt inverter, 220 watt solar input (charge controller), 61 Ah-LiFePO4 battery (calculated). The system has enough energy storage to run flat-out (800 watts) for 1 hour. Then you could re-charge it using the AC from your generator in just over an hour. The cost is 649 pounds or $817 (just looked it up today)...

It has all the parts in one unit which is good and bad in my book, it's a nice clean package and the price is not totally bonkers.

So if you made your own peace meal, what would it cost? : (prices thanks to Amazon)
100 Ah - 12 V liFePO4 battery - $250
1000 watt Inverter - $250
30 Amp Charge Controller - $40
2 GA Wire - 3'/1meter - $40
35 AMP (420 Watt) Smart Charger- $150
Kit I forgot (10%) $75

Total $805 Us/ 640 Pounds, plus VAT (sales tax)
So from a value proposition it's a wash, as long as nothing breaks...

I have something similar in the Lou here: Only difference I went with 4 LiFePO4-100AH batteries (8 times the system in your link); So the number of batteries and cables went up by 4. It did cost me $1650/ 1310 pounds to setup, But it gives 4800 watt hours of stand by power.
PXL_20230825_193909249.jpg


Everyone has to make decisions based on their comfort level and knowledge is power. I'm sure you will figure out what is best for your situation.
 

Latest posts

Back
Top