diy solar

diy solar

Planning 1000W/3kWh off grid system

DavidR

New Member
Joined
Jul 12, 2020
Messages
8
Hi, I’ve just joined this forum

I’m planning an off grid system to run a fridge and freezer in my shed. The shed has grid power but off grid will save me about $500 per year. Max power 600W (when both compressors are on at the same time). Energy budget 3 - 4 kWh per day. I imagine that 2.5 kWh storage should be sufficient to run fridge and freezer overnight. 1000W panels should be enough for sunny days but I may consider 2000W for cloudy days. I will most likely need an ATS so that fridge and freezer still have power when batteries low. I currently have plenty of panels and an old 2000W inverter (not sine wave)
I have not decided whether to go with an all in one system or seperate components. Options I am considering:
1. Victron 150/35 mppt, victron battery protect, ATS, existing 2000 W inverter.
2. Victron 1600/24 Easysolar
3. mpp solar 3000/24

Any advice here, or other options, would be welcome

Energy storage: 2 x 100ah LiFePO4 connected in series. I have not selected batteries yet, still looking for something economical. If anyone knows of inexpensive lithium batteries in Australia please let me know
 
Are the fridge and freezer mains voltage based compressors or are they inverter types? If they are mains voltage they will run off that modified sine wave inverter but the compressors will run hotter than usual. The combined starting current of both would likely cause the inverter to trip off due to overload.

Inverter based fridges effectively decouple the compressor motor from the mains and have greatly reduced start current, ie they gently ramp up to the desired power level rather than take a big slurp at turn on so are much more inverter (as in the thing that takes DC in and puts AC out) friendly.

The Victron battery protect should not be connected between the inverter and the battery. They aren't designed to support the very high loads that the inverter will present at initial power up if the inverter is dead connected to the battery, and will likely be damaged. The Victron MPPT's controlled output could be used to do low battery voltage control of the inverter instead. It may not be straight forward depending on how the inverter is turn on / off.

For an easy, inexpensive slap on the wall and you are done approach I don't think you can go past the MPP Solar unit. The Victron will obviously do the job too and as much as I like Victron equipment the MPP Solar will likely work well enough for you. If you do decide on MPP consider buying from the local authorised seller rather than direct from Taiwan (or where ever) as you'll get better support, including warranty support.

You might want to take a look at the Amptron lifepo4 battery. I have no direct experience with it but the specs seem reasonable for the price. Remember that the battery has to be able to supply the peak draw of the appliances, even if its only for a second or two. If you try to pull more current from them the BMS may well step in and drop the load so pay attention to the short term peak rating of what ever battery you settle on. Non-inverter compressors might draw 3 x their operating power during startup.

Series connected lifepo4 batteries may need occasional attention, ie bring them to fully charged separately occasionally. This is because the BMS will step in and disconnect if they get too far out of step with each other permitting voltage across one to rise too high during charging. Lead acid typically doesn't have a BMS so no disconnection there, but if they drift out of balance one battery will get overcharged - not a problem for a flooded open cell battery, but a problem for a sealed battery.
 
Last edited:
Are the fridge and freezer mains voltage based compressors or are they inverter types? If they are mains voltage they will run off that modified sine wave inverter but the compressors will run hotter than usual. The combined starting current of both would likely cause the inverter to trip off due to overload.

Inverter based fridges effectively decouple the compressor motor from the mains and have greatly reduced start current, ie they gently ramp up to the desired power level rather than take a big slurp at turn on so are much more inverter (as in the thing that takes DC in and puts AC out) friendly.

The Victron battery protect should not be connected between the inverter and the battery. They aren't designed to support the very high loads that the inverter will present at initial power up if the inverter is dead connected to the battery, and will likely be damaged. The Victron MPPT's controlled output could be used to do low battery voltage control of the inverter instead. It may not be straight forward depending on how the inverter is turn on / off.

For an easy, inexpensive slap on the wall and you are done approach I don't think you can go past the MPP Solar unit. The Victron will obviously do the job too and as much as I like Victron equipment the MPP Solar will likely work well enough for you. If you do decide on MPP consider buying from the local authorised seller rather than direct from Taiwan (or where ever) as you'll get better support, including warranty support.

You might want to take a look at the Amptron lifepo4 battery. I have no direct experience with it but the specs seem reasonable for the price. Remember that the battery has to be able to supply the peak draw of the appliances, even if its only for a second or two. If you try to pull more current from them the BMS may well step in and drop the load so pay attention to the short term peak rating of what ever battery you settle on. Non-inverter compressors might draw 3 x their operating power during startup.

Series connected lifepo4 batteries may need occasional attention, ie bring them to fully charged separately occasionally. This is because the BMS will step in and disconnect if they get too far out of step with each other permitting voltage across one to rise too high during charging. Lead acid typically doesn't have a BMS so no disconnection there, but if they drift out of balance one battery will get overcharged - not a problem for a flooded open cell battery, but a problem for a sealed battery.

Thanks so much for the detailed reply. There are a lot more issues than I thought. The fridge and freezer are not inverter style. Thanks for the info on lifepo4 batteries. I did suspect I would have to check the balance of the batteries from time to time ( I have plenty of experience with Lipoly batteries for RC planes)

I like the idea of wiring up all the separate components but it seems that will not be practical or economical. And I have not been able to find an ATS available in Australia. I am now leaning towards the mpp solar. I have spoken with the Australian seller and he advised against the 2.4kW 24V system for 24/7 operation. Rather he recommended to 3kW 24V. He said it is much heavier duty with large heat sinks and no fan issues.

However, I cannot afford this larger system as well as lifepo4 batteries. I do have 2 x 100ah agm batteries (in my boat) and originally thought of just adding 2 more agm but decided this would be a waste of money.

So, my new plan is to use the 800W 12V mpp solar (probably just for a year or 2 and then upgrade) and use the 2 existing agm batteries while I source lifepo4 batteries. This means there is no rush to buy the lifepo4 batteries and I can hunt around for a good deal. Obviously this system won't be large enough for fridge and freezer so I'll just keep the freezer on grid power and use the solar system for the fridge and also for a 140W pond pump. The pond pump will only be on for 2 - 3 hours per day (daytime). This plan gives me plenty of time to learn about off grid solar before I make the decision to go with the larger mpp solar or Victron easy solar, or something else. Does this sound like a reasonable plan?
 
Do you have a MPP Solar unit / MPPT charger already or are the panels currently not being used at all?
 
A little outside the box thinking, with a view to minimising money spent - you have the batteries and inverter, all that is lacking is a charger and the ATS.

Instead of an ATS that would simply rely on the inverter running the battery flat and turning off (or going into its own low DC cutoff) you could use two relays (or a double pole relay, either way two separate circuits one for active one for neutral) and have them switched by the load output terminals of a charge controller. Then the charge controller could change your fridge/freezer over to mains should the battery run low. The idle draw of the inverter should be low enough that it won't drain the battery before the sun comes up, if you set the charge controller's disconnect correctly.

You could even simulate this before committing to the SCC by setting up the relays but instead of having the SCC controlling them, a 12VDC plugpack from a modem could be used to control the relays. Set everything up, plugpack on so load is connected to inverter, turn off, load goes to mains, wait 20 seconds, turn plugpack off, load goes back to inverter. That way you can see if your existing inverter would be happy getting the load dropped onto it live (ie, compressor is running off mains, then is dumped straight onto the inverter's output). It may not be, may go bang, let smoke out etc hence the test.

Of course anything DIY like this is not legal in Australia and you'd have to know what you are doing to do it safely.

You can buy a 40 amp (suitable for 1kW of panels into a 24v battery) Renogy (OEM SRNE if you want to get it a little cheaper) charge controller for approx $150.

Can you post a link to the 800W MPP you are looking at?
 
Last edited:
A little outside the box thinking, with a view to minimising money spent - you have the batteries and inverter, all that is lacking is a charger and the ATS.

Instead of an ATS that would simply rely on the inverter running the battery flat and turning off (or going into its own low DC cutoff) you could use two relays (or a double pole relay, either way two separate circuits one for active one for neutral) and have them switched by the load output terminals of a charge controller. Then the charge controller could change your fridge/freezer over to mains should the battery run low. The idle draw of the inverter should be low enough that it won't drain the battery before the sun comes up, if you set the charge controller's disconnect correctly.

You could even simulate this before committing to the SCC by setting up the relays but instead of having the SCC controlling them, a 12VDC plugpack from a modem could be used to control the relays. Set everything up, plugpack on so load is connected to inverter, turn off, load goes to mains, wait 20 seconds, turn plugpack off, load goes back to inverter. That way you can see if your existing inverter would be happy getting the load dropped onto it live (ie, compressor is running off mains, then is dumped straight onto the inverter's output). It may not be, may go bang, let smoke out etc hence the test.

Of course anything DIY like this is not legal in Australia and you'd have to know what you are doing to do it safely.

You can buy a 40 amp (suitable for 1kW of panels into a 24v battery) Renogy (OEM SRNE if you want to get it a little cheaper) charge controller for approx $150.

Can you post a link to the 800W MPP you are looking at?

Thanks for trying get the cost down for me but this sounds a bit more involved. I'm not sure how to wire this system and don't follow how to do the test. I'm willing to learn but I might be better off with an all in one.
The 800W is actually 1000W. I thought the 12V mpp solar inverters were 800W after watching one of Will's Youtube videos.
I spoke with Daniel at mppsolar (Australia) and he strongly recommends the 3kW 24V inverter.

New plan: Start with the 3kW 24V mpp solar and my 2 existing 100ah batteries. Set the low voltage cut off so that I don't use more than 50% Dod and let the grid take over when batteries get low. Purchase the lifepo4 batteries later. Do I need an RCD and 10A circuit breaker?
 
Australia's GPO circuits are 15 amps, 10 amps max draw from the socket. At 3000 watts the inverter would do about 13 amps continuous, but there is a surge rating there too which should let it pop a 15 amp breaker / fuse if you used one that size and a short developed. Strictly speaking if you had both the fridge and feezer starting at the same time on the circuit they shouldn't draw enough to pop the 10 amp breaker but you might be sailing close to the wind there.

RCD would be a good idea, but you need to get the earthing right for it to work. Since this system will be able to draw from your mains supply you should talk to an electrician about this as getting it wrong will compromise the operation of any existing RCDs and possibly allow fault currents to flow in places you don't want them to.
 
I'm doing something similar in USA, but are you sure your fridge and freezer consume $500/yr of electricity? I'm planning on powering 9 and 7 cubic ft. chest freezers, and I'm not mainly doing it to save on electricity as combined I estimate they consume no more than $65/yr (US$) of electricity. I'm mainly doing it to beef up my emergency power capability. If you are comfortable DIY building a LiFePO4 battery, that may be another way to go to get more battery capacity at a reasonable cost. I looked at the MPP all-in-one units, and the inverter on those pulls ~ 2A even when there is no load. This will eat into your battery capacity a lot at night. I decided to go with a Giandel 24v 2kw pure sine inverter because it is supposed to pull less than 1A under no load. I have decided to buy two dirt cheap PWM 30A charge controllers to start out as my panels will be the right voltage to charge a 24v system and MPPT won't have much efficiency improvement over them. - I may regret this, but two together are less than $30. If they don't work out I can always upgrade to something better. If they do work out, I can afford to buy a couple of spares just in case one or both die. I've also been thinking about contingency in case of several days of cloudy weather and batteries run low. I don't know whether this will work, but I was thinking of getting an adjustable DC power supply, set it for ~24.5 volts (make sure it is protected by a beefy diode) and connect to my battery. I'm thinking if battery voltage drops below that threshold, then current will flow from the mains to supplement the battery charge. It seems like a simple solution. My main worry is that if the inverter fails or fails to start the freezer compressors (both freezers cycle at exactly the same time?), then I don't want to lose the contents of my freezer. I'll want to put in a temperature alarm and check on them frequently.
 
I'm doing something similar in USA, but are you sure your fridge and freezer consume $500/yr of electricity? I'm planning on powering 9 and 7 cubic ft. chest freezers, and I'm not mainly doing it to save on electricity as combined I estimate they consume no more than $65/yr (US$) of electricity. I'm mainly doing it to beef up my emergency power capability. If you are comfortable DIY building a LiFePO4 battery, that may be another way to go to get more battery capacity at a reasonable cost. I looked at the MPP all-in-one units, and the inverter on those pulls ~ 2A even when there is no load. This will eat into your battery capacity a lot at night. I decided to go with a Giandel 24v 2kw pure sine inverter because it is supposed to pull less than 1A under no load. I have decided to buy two dirt cheap PWM 30A charge controllers to start out as my panels will be the right voltage to charge a 24v system and MPPT won't have much efficiency improvement over them. - I may regret this, but two together are less than $30. If they don't work out I can always upgrade to something better. If they do work out, I can afford to buy a couple of spares just in case one or both die. I've also been thinking about contingency in case of several days of cloudy weather and batteries run low. I don't know whether this will work, but I was thinking of getting an adjustable DC power supply, set it for ~24.5 volts (make sure it is protected by a beefy diode) and connect to my battery. I'm thinking if battery voltage drops below that threshold, then current will flow from the mains to supplement the battery charge. It seems like a simple solution. My main worry is that if the inverter fails or fails to start the freezer compressors (both freezers cycle at exactly the same time?), then I don't want to lose the contents of my freezer. I'll want to put in a temperature alarm and check on them frequently.

Interesting to see someone doing something similar. Maybe we can compare notes as we progress. I’m pretty sure that our power is more expensive in Australia - it actually varies a lot from state to state and even different locations. People in the capitol cities usually can get a cheaper rate since there are several energy companies. I’m in a regional area with only one energy company. We pay .27 per kWh. I already have a 5kW system (installed commercially) that exports to the grid at 44 cents per kWh (this was a state government incentive to promote solar power). Any power that I use during the day is power that I can’t export so power I use during the day actually costs me 44c per kWh.
The main reason that I’m going with the all in one is for the ATS. Interesting idea to use a dc power supply. I’m not sure if that will work and I don’t know how much power it will draw. I know that my dc power supply runs 2 fans that stay on even if I’m not using any dc power. I guess they will all vary but I’d check that first. Probably a good idea to ask others on this forum if it will work.
Where did you find data on the mpp inverter. I couldn’t find the idle power consumption on their website. It’s probably different for different models also.
 
Where did you find data on the mpp inverter. I couldn’t find the idle power consumption on their website. It’s probably different for different models also.
This is from Will Prowse's page. He lists high idle current as a con. In one of his videos he talks about it consuming 40-45 watts at idle. I'm expecting my Giandel inverter that I have on order will probably consume 20-25 watts at idle. https://www.mobile-solarpower.com/all-in-one-solar-power-packages.html

I don't have a grid tie system yet, but am seriously thinking about it as we get a tax credit of 26% if I do it by the end of 2020. My grid power cost is only around US $0.12 per kWh, so break even here is about 8-10 years on a grid tie system.
 
I can only speak for my particular model Giandel inverter, but the specs for it state idle draw is 1 amp but actual measurement shows it is less.
 
I'm planning on powering 9 and 7 cubic ft. chest freezers, and I'm not mainly doing it to save on electricity as combined I estimate they consume no more than $65/yr (US$) of electricity.
$65 / 0.12 = ~ 541kWh. I know a chest freezer can be modified to act as a fridge and run around 270 Wh / day but a freezer at that rate too? That'd be impressive.
 
$65 / 0.12 = ~ 541kWh. I know a chest freezer can be modified to act as a fridge and run around 270 Wh / day but a freezer at that rate too? That'd be impressive.
My cost estimate might be off some. I just remember the 7 cu ft. model had an energyguide sticker that said $27/year. I'm not sure what electricity cost was figured into that sticker. I did use a Kill-a-watt meter on it once and found that it used about 1kw/day in the summer in my hot garage. (at least I think I measured it during the summer last year. I should probably go back and measure again since it is hot again. I'm planning on putting 1kW of panels up and a 5kWh DIY LiFePO4 battery. If that's not big enough for both freezers, I'll just run one for now.
 
This is from Will Prowse's page. He lists high idle current as a con. In one of his videos he talks about it consuming 40-45 watts at idle. I'm expecting my Giandel inverter that I have on order will probably consume 20-25 watts at idle. https://www.mobile-solarpower.com/all-in-one-solar-power-packages.html

I don't have a grid tie system yet, but am seriously thinking about it as we get a tax credit of 26% if I do it by the end of 2020. My grid power cost is only around US $0.12 per kWh, so break even here is about 8-10 years on a grid tie system.

Giandel looks pretty good, really good value.

I have revised my plan:
Victron 150/35 mppt and Victron low voltage cut out. I think I’ll use my existing 2 x 100ah agm batteries and existing 2000W inverter. This will get the system up and running for under $500. I can then check operation before completing system.
If everything works out ok I’ll probably go with the Enerdrive 24v 2000W inverter. It’s expensive but very good quality and comes with automatic transfer switch and an rcd. The LiFePO4 battery I’m looking at is the Amptron 24v 100ah ($AU1859). One of these initially then a second one next year.
 
Back
Top