Newbie - general advice for off-grid cabin?

[OffGridMike]

I am new to the forum and new to solar. I recently purchased an off-grid 2400 ft2 log home in SW Washington state. It has an existing system from 2009 and I am looking to update/upgrade. I have pretty good southern exposure and typically only use the home on weekends (twice per month). The entire system is in a small non-insulated shed and the roof has enough space for 6 panels. You can see the shed in the right side of the pic below. I am looking for any feedback. I have a few questions I am thinking about at the end of the post and would welcome any suggestions or comments. Is there anything I may not be thinking about that you would recommend I look into? Thanks for everyone's comments in advance.

My current System
Xantrex XW6048 hybrid inverter/charger
Xantrex C40 DC controller
Xantrex XW System Control Panel
Xantrex XW Automatic Generator Start
Schneider Electric Insight Home
8x, Duracell SLI6V370S flooded lead acid batteries
4x, Moser Baer MBPV CAAP 215w panels
Kohler 12RES propane generator

Proposed new components
EG4 6000XP Off-Grid Inverter
3x, EG4 LifePower4, 48V 100AH server rack lithium batteries (with 6 slot rack)
6x, SunPower 430W Mono solar panel with RIGO rapid shutdown module

Based on what I understand, this will increase both my battery capacity and solar power generation, and replace everything except my Kohler generator.

Questions:

1. Is the generator functionality sufficient in the 6000XP that I do not need a separate AGS unit?
2. Do I need some sort of control panel or is the 6000XP enough to monitor inputs, outputs, batteries, and the overall system?
3. The generator is 15 years old and has a lot of hours on it (partly because the existing batteries were not maintained well and don't hold much charge). Do I need the EG4 chargeverter or some other charge controller?
4. Can I use my existing panels on the north side of my shed roof? If I run them in a separate string, can I connect them to the inverter and use the additional power they might generate?
5. What happens if the batteries are full and the sun is shining, and I am not using my cabin for a few weeks? Do I need anything to discharge power?
6. Do I need to worry about a heat source in the shed for the winter?

 

[400bird]

Why throw away good old inverters for new import inverters?

I'd just buy new batteries and panels.
Maybe a new charge controller if the C40 doesn't work. Also, the C40 isn't MPPT, so there's room for improvement there.

 

[400bird]

I see now it renumbered your list when I responded. I don't have answers for the first two questions, so let's try my counting skills.

3. The generator is 15 years old and has a lot of hours on it (partly because the existing batteries were not maintained well and don't hold much charge). Do I need the EG4 chargeverter or some other charge controller?

Only one way to know for sure, try it out.
But if the generator doesn't hold a stable idle, plan on it not cooperating well.

1. 4.Can I use my existing panels on the north side of my shed roof? If I run them in a separate string, can I connect them to the inverter and use the additional power they might generate?

Sure, separate input to the inverter/charge controller.

1. 5.What happens if the batteries are full and the sun is shining, and I am not using my cabin for a few weeks? Do I need anything to discharge power?

Nope, the charge controller just stops accepting power from the panels.

1. 6. Do I need to worry about a heat source in the shed for the winter?

The batteries have a minimum temp of about freezing. So, probably yes.

 

[OffGridForGood]

Questions:

1. Is the generator functionality sufficient in the 6000XP that I do not need a separate AGS unit?
2. Do I need some sort of control panel or is the 6000XP enough to monitor inputs, outputs, batteries, and the overall system?
3. The generator is 15 years old and has a lot of hours on it (partly because the existing batteries were not maintained well and don't hold much charge). Do I need the EG4 chargeverter or some other charge controller?
4. Can I use my existing panels on the north side of my shed roof? If I run them in a separate string, can I connect them to the inverter and use the additional power they might generate?
5. What happens if the batteries are full and the sun is shining, and I am not using my cabin for a few weeks? Do I need anything to discharge power?
6. Do I need to worry about a heat source in the shed for the winter?

I agree with @400bird if the Xantrex is working fine, I would just keep it, unless expansion beyond 6kW is in the plan?
A few more points to add to his:
I like Solar Assistant for monitoring - some members report they like Home Assistant - or both together - remote monitoring capable with both
The Chargeverter is a good piece of equipment - but keeping the Xantrex and current generator = you know they work. I wouldn't plan on changing out the generator unless you see issues. A dual fuel generator may be a nice upgrade.
Yes you can use the existing PV as a sep string, an upgrade would be to relocate them to a better solar orientation.
If the batteries are full, and sun shinning, nothing bad happens, other than that feeling of lost potential, especially if you see days of rain in the forcast.
LFP batteries need to be above freezing temps to be charged. Close to freezing they may have reduced charge rate requirements. Some type of heating may be needed - depends on your local climate, but Insulating the shed would be a low cost first step. Many locations benefit from mini-split combined heating/cooling equipment since your LFP also have issues if too high temps. Generally batteries like temps people like.

 

[OffGridMike]

Why throw away good old inverters for new import inverters?

I'd just buy new batteries and panels.
Maybe a new charge controller if the C40 doesn't work. Also, the C40 isn't MPPT, so there's room for improvement there.

Being a novice and being about 90 minutes from the nearest town, I was thinking I would opt for something that is clearly compatible and communicates easily between each other. I think I need to upgrade the charge controller if I go to Lithium. Then I need to get something that converts the EG4 communication to Xanbus... Then the charge controller communicates to the Inverter...

Am I thinking about that incorrectly? I was going for simple (and the cost is about the same).

I wouldn't throw the inverters away, but probably sell them or donate them to a neighbor.

 

[OffGridForGood]

I realize this is all new for you, some thoughts on the overall system:
A few PV panels in the Pacific NW are unlikely to produce what you need - search PVWatts online, plug in your location and PV and it will tell you the monthly average solar you can expect - note PV Watts often over-estimate solar, especially if you have any shade on those panels during the day/year due to tall trees etc. I plugged in Cougar WA, 1kW PV tilted 46 degrees, due South and see Nov Dec and Jan as 40 kWh per month ie nearly nothing.

We usually start a design with an audit of the loads you are likely to run on those weekends, then calculate the PV to supply those, and the batteries needed to store. The super small existing 800W PV array could only work if the user does short visits with long periods away to allow the solar to catch up again, and the generator ran for poor weather periods.
Three 5kWh server racks are not really a lot of storage either, depends if you have propane appliances for some loads or not. ie back to that energy audit to know peak power, and average energy use. Consider if the loads are different during various seasons - more during summer works well with longer days, but not in Winter.

 

[MichaelK]

This is sort-of like suggesting you want to get rid of a somewhat older Mercedes to purchase a new Hyundai. The quality of the Xantrex is far greater than the EG4.

Your biggest single problem is the classic "too large a battery for available solar". Your Duracells would be most happy with a charge rate of 1/8th of C, which is 370Ah/8 = 46A. With an MPPT controller, at theoretical best, you would only get 17A, and hobbled with a PWM controller, most likely only 10A. So, I'm surprised your batteries actually lasted this long.

To get an honest 46A, I'd de-rate panels to 85%, so what you should have had is (46A X 50Vcharging)/85% = 2705W of panels, more than three times what you have now. For my own 48V system, with which I started with Trojan L-16 batteries of ~ the same size, I went with 4500W of panels. That is literally 5 times more than what was implemented at your site.

I can understand that you can only fit so many panels on one roof. I think the alternative you have to look at is ground mounts to add capacity. From day one at my own place, I designed my own system around ground mounts that can be rotated East to West over the course of the day. This has allowed me to keep my batteries charged even in the rain. Years have gone by before I ran a generator.

So, here is my advice. Replace the batteries, charge controller, and add panels. Don't replace the inverter. Install thousands of watts of additional solar. Don't buy any more of those little 12V panels. Go with large, high-voltage residential panels. For my own system, I've got 3-4 high-voltage grid-tie style panels wired in series to make 120VDC. This allows me to run ~8A of current at 120VDC with no detectable voltage drop. Shop for large panels on Craigslist, or FaceBook Marketplace. Get a 200V controller like a Midnight200 or Epever Tracer 8420.

 

[Bluedog225]

Welcome. Do some research on the company you are planning on buying from. The forum has many threads on vendor quality and customer service.

And roof mount, though appealing, has many drawbacks.

 

[OffGridMike]

Thank you all for your excellent feedback. I am truly appreciative of the time you took to write back.

I really appreciate the analogy @MichaelK posed. It makes a lot of sense to me. I also hear loud and clear that the capacity of my panels is a huge opportunity. Let me ask a few different questions.

• I assumed I want to move from FLA to lithium batteries if the price was similar. the cost of replacing 370ah of FLA is more than getting 200ah of server rack lithium. It seemed like a no-brainer to me but do you agree now is the time to convert? I was thinking about a 50% increase in storage (300ah at 95+% usable instead of 370 at 50% usable).
• As far as I can tell, the XW6048 and the C40 DC controller are not designed to support lithium batteries and I need to replace the controller at a minimum. Am I wrong about that? Should I stick with a Schneider Electric Conext MPPT 60 150 for compatibility issues? Or am I overcomplicating the compatibility issues? I want as close to plug-and-play as possible.
• I agree that my existing panels are totally insufficient. That is why I was thinking about going to 6x 430w panels. It seems like 10 of those (or similar sized panels) would be in the range others are suggesting in their responses and able to be supported by a new charge controller. Does this sound like enough of an improvement? Should be 5x the existing solar potential.

 

[OffGridForGood]

I agree that my existing panels are totally insufficient. That is why I was thinking about going to 6x 430w panels. It seems like 10 of those (or similar sized panels) would be in the range others are suggesting in their responses and able to be supported by a new charge controller. Does this sound like enough of an improvement? Should be 5x the existing solar potential.

I will take this part, the other guys will reply on the first two questions I expect.
PVWATTS.NREL.GOV calculator, Cougar WA, 6x430w(2.58kW) due south, tilted 46 degrees from horizontal: Monthly total energy
Jan 106 kWh/mth
Feb 164
Mar 187
Apr 211
May 257
Jun 260
Jul 325
Aug 337
Sep 278
Oct 213
Nov 125
Dec 111

Total for year = 2,574kWh, Annual Average per day = 7kWh per day
Worst month January Average per day = 3.4kWh/day
Best month August Average per day = 10.9kWh/day
Ratio best to worst = 3.2 : 1


So the loads, per day are ?????
are the daily loads more during summer - A/C?
Say you plan 6 days per month on site, and base design on January solar collection (worst month)
111kWh collection x 80% (account for real world conditions) = 88.8kWh
you will do two weekends say 6 days total per month, 88.8/6days = 14.8kWh available per day.
However during January you can expect only 3.4kWh per day on average, and may need 40kWh ESS (or more) to store the sunny days and ride out the overcast ones.
Most of the year you would cruise along with solar to spare.
Some information on the likely loads per day would help.

 

[OffGridMike]

@OffGridForGood , Thank you for your response. I checked out the website you recommended and it was very helpful.

I should have provided some information about load history from the beginning. I don't think I use very much but that is partly because I am constantly worried about the generator coming on. My Insight Home web portal shows an average idle daily use of about 2.5 kWh per day. On the weekends I am there, the use appears to be about 7.5-8kWh. The peak I saw last year (first full year owning the cabin) was 15 kWh and that was Thanksgiving with lots of people and activity. I do not have electric heating or cooling. Aside from my electric fridge and Starlink (w/ heated dish), the largest power draws are coffee maker, microwave, and fans in the summer. Lights are all LED and we are really good about turning them off when we leave a room.
Based on the seasonal data from the calculator, the 6 panes might be just barely enough. I should find a way to get 10 panels installed and it should be plenty of solar in the summer and a comfortable margin to maintain the daily draw throughout the winter unless I get a lot of use. And I could probably go without using the generator except in peak times in the winter.

 

[Q-Dog]

If you only ever spend 2 days at a time here, you have 6 days or more for your batteries to charge back up after a weekend of use. I don't think you need to go crazy on the solar.

I would do the battery upgrade and see how things go with the rest of the system.

 

[MichaelK]

If you only ever spend 2 days at a time here, you have 6 days or more for your batteries to charge back up after a weekend of use. I don't think you need to go crazy on the solar.

I would do the battery upgrade and see how things go with the rest of the system.

The problem as I see it is that the OP is located in the Pacific Northwest, which means gloomy, cloudy weather 8 months of the year. Designing the system to charge over 6 days means that in marginal weather, the system can not charge itself at all.

Batteries that take 6 days to charge are going to sulfinate faster than if fully charged immediately. The OP would just be setting himself up for a second battery bank failure. I'm further South, at 35 degrees North, and what I'm seeing on wet rainy days is ~ 0.5 sunhours per day, while raining. Doing the math, to supply 2500Wh of idle power under all conditions, you'd need 2500Wh/0.5sh = 5000W of panels.

That is not so horrible. Right now today, I have 5500W online, making power. My batteries stay fully charged even on rainy days. The OP just needs to find a local venue where panel prices are not objectionable. Just two months ago I purchased another four 250W residential panels for 40$ each. That's 160$ cash for 1000W of panels. Start looking for the best locations on the ground to position ground mounts. My new V5.0 design can handle as many as six large residential panels on one rotating array.

In the short-term, shut off the inverter to reduce the idle drain to a minimum. Replace the old charge controller next. You don't need to go with the very high-priced Conext. My XW+6848 gets along just fine with a Midnight200 charge controller. Maybe even a Tracer 6420AN. Then start shopping for panels. You might be able to still utilize the older 215W panels if you wire them all in series, then parallel them with more panel strings of the same voltage. We need to know the specs of those panels to facilitate this.

 

[Q-Dog]

The problem as I see it is that the OP is located in the Pacific Northwest, which means gloomy, cloudy weather 8 months of the year. Designing the system to charge over 6 days means that in marginal weather, the system can not charge itself at all.

Batteries that take 6 days to charge are going to sulfinate faster than if fully charged immediately. The OP would just be setting himself up for a second battery bank failure. I'm further South, at 35 degrees North, and what I'm seeing on wet rainy days is ~ 0.5 sunhours per day, while raining. Doing the math, to supply 2500Wh of idle power under all conditions, you'd need 2500Wh/0.5sh = 5000W of panels.

That is not so horrible. Right now today, I have 5500W online, making power. My batteries stay fully charged even on rainy days. The OP just needs to find a local venue where panel prices are not objectionable. Just two months ago I purchased another four 250W residential panels for 40$ each. That's 160$ cash for 1000W of panels. Start looking for the best locations on the ground to position ground mounts. My new V5.0 design can handle as many as six large residential panels on one rotating array.

In the short-term, shut off the inverter to reduce the idle drain to a minimum. Replace the old charge controller next. You don't need to go with the very high-priced Conext. My XW+6848 gets along just fine with a Midnight200 charge controller. Maybe even a Tracer 6420AN. Then start shopping for panels. You might be able to still utilize the older 215W panels if you wire them all in series, then parallel them with more panel strings of the same voltage. We need to know the specs of those panels to facilitate this.

LiFePO4 doesn't sulfate and doesn't care how slow you charge them.

 

[OffGridForGood]

and doesn't care how slow you charge them.

this part is not quite true (plating issues), but I get your point.
I think the OP could set up the system for remote starting the generator - only if the forcast is poor solar, and they have a trip to site planned. Remote start and recharge the batteries a few days prior to the trip, arrive to fully charged ESS and no noise.

 

[400bird]

Alternatively, they could put the inverter into standby when away. That drops the inverter energy use significantly. Single digit watts, if memory serves.

 

[OffGridMike]

You guys are all awesome. I am learning so much as you debate my situation. I cannot thank you enough.

 

[OffGridForGood]

I believe we have some agreement - at least between the guys commenting on this thread:
keep the inverter
get new SCC
move the original panles
add some more panels
replace the batteries with LFP (I would suggest based on your 7-8kWh/day onsite use, 25kWh ESS)
keep the Genset - will not run as often now
keep remote monitoring
be able to remote start the genset if you need it for an upcoming trip to site.
I really like Solar Assistant personally, others like Home Assistant (or both).

I think the only other item to consider, is if the LFP need heating while you're away?
Build them into an insulated cubby and add a 48volt silicone heating pad on stat.

 

[mvonw]

This is sort-of like suggesting you want to get rid of a somewhat older Mercedes to purchase a new Hyundai. The quality of the Xantrex is far greater than the EG4.

I would bought a Xantrex system for my off grid system, and I have a Sol Ark (which has been flawless), if I knew what I know now. I would not even consider a lower tier product.

 

[MichaelK]

LiFePO4 doesn't sulfate and doesn't care how slow you charge them.

The batteries have not yet been purchased, so this will be a wait and see problem.