diy solar

diy solar

Battery banks, parasitic loads and low sun hours

When you say "without powering anything, no loads", that leaves a lot open. How do you know there are no loads? Did you physically disconnect the breakers or just assume you didn't have anything running? Even a television that is off will draw power. A battery charger that is plugged in but not charging anything will draw power. Even a desktop computer that is plugged in but turned off will draw power. And that stuff adds up fast.

The only way to be sure no power is being dawn is to either measure it accurately or physically open circuits that feed everything.


Yup.. and we'll help you figure it out.. My money is on unexpected parasitic loads in the house rather than those batteries.. but we'll get there and figure it out.


Your batteries are lithium phosphate.. a safe type of lithium. One of the minor drawbacks is that the discharge voltage curve is very flat. It is easy to use a multimeter to determine a full battery from an empty one, but everything in between from about 10% to 90% charge all seems to be within a very small voltage range. For this reason, multimeters don't do a very good job at determining state of charge on phosphate batteries so you kind of have to diagnose their capacity by looking at fully charged and then fully discharged voltages and determining how much juice you sucked from them, or used to charge them, between those two extremes.


That is normal. The bigger concern are the individual cell level voltage differences. If your battery has 16 cells in series and the voltage of the highest cell minus the lowest cell is more than a certain amount, then you have issues. Most lithium is kept to within 0.020 volts (20mV) and some folks like to see even less than that. I use lithium ion cells (not as safe but have more power), and I keep mine within 20mV, although during charging they sometimes do get as far out as 50mV.


Balancing is about cell level voltages not so much module level.

Still waiting on Fortress to respond.
When I say there are no loads, I am literally going to the panel and turning off the solar breakers and turning the power from our PUD back on. That's how I know the only loads are the batteries and the inverters.

This whole line of questioning started when I was struggling to keep the SOC high enough to get us through the night after several days of poor uptake. When that happened, I turned the solar breakers off and went back to PUD while the batteries recharged. It was taking several days to bring the battery bank back up so I started documenting daily uptake and using SOC at two points each day to determine how much was getting to the bank. At that point, I was seeing missing kWHs that were too big to make sense. This was when I made my first attempt at getting the equipment's parasitic load information. It's been Fortress causing the problem in these specific calculations because, at first I was given no information at all and was told it was a communication issue between the Sol-Arks and Fortress. They sent new cables, everything was changed around and updated and we all assumed it would fix things. When it didn't I contacted them again to get clarification on their battery's energy usage. Now you can see the information I'm actually getting. I assumed it was correct so I was asking, in this forum post, what others were seeing in system parasitic load so I'd have some idea if what I was told was even normal and, if so, what options would make more sense for our system.
 
When I say there are no loads, I am literally going to the panel and turning off the solar breakers and turning the power from our PUD back on. That's how I know the only loads are the batteries and the inverters.

This whole line of questioning started when I was struggling to keep the SOC high enough to get us through the night after several days of poor uptake. When that happened, I turned the solar breakers off and went back to PUD while the batteries recharged. It was taking several days to bring the battery bank back up so I started documenting daily uptake and using SOC at two points each day to determine how much was getting to the bank. At that point, I was seeing missing kWHs that were too big to make sense. This was when I made my first attempt at getting the equipment's parasitic load information. It's been Fortress causing the problem in these specific calculations because, at first I was given no information at all and was told it was a communication issue between the Sol-Arks and Fortress. They sent new cables, everything was changed around and updated and we all assumed it would fix things. When it didn't I contacted them again to get clarification on their battery's energy usage. Now you can see the information I'm actually getting. I assumed it was correct so I was asking, in this forum post, what others were seeing in system parasitic load so I'd have some idea if what I was told was even normal and, if so, what options would make more sense for our system.
I was just contacted (email) again by Fortress. A new guy replied and stated:

"It is important to point out that when a battery is turned on, it's not only the BMS that consumes energy, but also the contactors and the LED lights as well. Combined energy use can easily reach 30 watts."

His statement tells me he's probably another "application engineer" because he's wrong. He's claiming that the LED lights are even part of the consideration? ROFLMAO. Do your batteries illuminate your entire home with their LED's? Because if not, and if they're just tiny little red or green indicator lights, then their power consumption is so ridiculously insignificant that most engineering calculations wouldn't even factor them in. We're talking about a single AAA alkaline battery running them for several days.

And on the subject of the contactor inside the module using up that much energy, that's a load of garbage as well. I use REC BMS's and we have to hook up external contactors.. so we use the 48 volt Tyco Industrial Kilovac contactors rated for a whopping 600 amps, and they only pull 1.74 watts to hold them closed.. That's not even 2 watts to hold a massive relay closed.

I'm still conversing with them.. at this point, I think I've cleared Customer Service Guy #1, and now just communicated with Customer Service Guy #2. I'm hoping I eventually reach someone with some knowledge.

In the mean time, you should 1) Charge those batteries up to 100% using the grid power if you need, 2) Totally Disconnect them from any thing, including the inverter. Take the power cables off if you need. 3) Use a multimeter to record the voltage at the battery terminals and write it down. 4) Repeat step 3 every 12 hours. 5) Continue repeating the voltage measurements until the battery modules are telling you they are low.

If your batteries are really drawing 30 watts for each module, they should run out of juice within 7 days (which is ridiculous at best). I suspect that after two or three days, the voltage drop you record will be so small it will make you wonder if the modules are even on. Mine would go from 99% SOC to 98% after two days. By the end of the first week, my battery bank might be at 95 to 96%, if even that low It would probably take 2 or 3 months or more to drain my battery just because the BMS and contactor are active.

I can tell you this, if we verify these modules are using a wopping 30 watts just to remain active, I'm going to be bad mouthing them everywhere. That's a serious design flaw.
 
Got a question for you..
What is the ambient temperature of the room the battery modules are in? If it's really cold then this isn't going to help, but if the room is warm, like any common room in your home, then I want to know if the battery module feels warm if you put your hand on it.

Battery should be turned on for several hours before this test. Room should be relatively warm like a living room. Does the case feel warm or does it feel like any other object in the house?
 
This is one issue most people don't think about when they go off grid. I didn't plan for it but I guess I got lucky. My system has low draw, about 30w in night mode.

I use multiple inverters and most only come on when needed. My well pump for example is on an AIMS LF 24v 4KW. It draws about 92w idle. Its a pig but its got some balls. It only comes on when the pump is needed.
 
Got a question for you..
What is the ambient temperature of the room the battery modules are in? If it's really cold then this isn't going to help, but if the room is warm, like any common room in your home, then I want to know if the battery module feels warm if you put your hand on it.

Battery should be turned on for several hours before this test. Room should be relatively warm like a living room. Does the case feel warm or does it feel like any other object in the house?
The batteries are in our uninsulated shop, unfortunately. They weigh a whooping 108 lbs so I can't move them. They have R19 insulation packed around them because it's been cold this year. This is our first few months with the system up and running. The lowest the temp has been in the shop in the area of the inverters has been 38F. I live in SW Washington and this year the temps have been down to 16 at night a few times. Whenever there is potential for below 32F outside, we put propane heaters on to be sure we don't drop too cold.

This was another issue I had with them last week which is when I met Mr. Honey who you probably talked to. My batteries are laid out from left (by our shop/garage door) to the right which is in the middle of the shop wall well away from any doors or drafts. I'm putting a pic up for you to see. The top two pieces of insulation were removed for the picture as I was working on the BMS warning light with Fortress. They sit up on a bit of a shelf which is supported by 2x4's for extra strength. Each 5 battery section has 4 pieces of insulation one on the back, one on the front and two to close in the top allowing only gaps for the battery cables. The temperature, according to the thermometer on the wall next to our Sol-Arks (closest to the garage door) said 38F at it's lowest overnight. Anyway, I had shut the system down because of low power. Their stated storage temp is 20F. When I brought the batteries back online, I had 4 of my newer batteries in the middle of the line up (from the left, it was batteries 6-9) showing a BMS warning light (flashing red instead of the normal green). I worried and at the time didn't know how to see what the alarm was so I called Fortress and was walked through how to see the alarm. It was for a warning that the temperature was getting too low. Funny that it was low in the middle when it had been wrapped in insulation all night and it wasn't that cold in the shop! That's how I met Mr. Honey who informed me that I should just let the system stay on so it could keep itself warm. And this is when I asked about their parasitic load.

When we read the specs on the batteries, it says operating temperature min was 32F and storage was 20F. We don't normally have very low temps so we hadn't been worried until these guys told us they needed to be in a warm environment. I was like, "WTF?!"

Nothing is as advertised. And their stated "efficiency" is a fraud if they eat 30 watts.

I will get them back up to full and start my testing on Saturday when my husband is available to work with me.

Again, thank you so much for all of your help!
 

Attachments

  • Battery Bank.jpg
    Battery Bank.jpg
    272.8 KB · Views: 47
If you have those battery modules wrapped up so well, then they probably aren't drawing the 30 watts each that Fortress claims.

30 watts each x 10 = 300 watts. That 300 watts would have warmed those batteries up quite well in that small space. This is now another reason that I think that 30 watt figure is a typo or misprint.
 
The batteries are in our uninsulated shop, unfortunately. They weigh a whooping 108 lbs so I can't move them. They have R19 insulation packed around them because it's been cold this year. This is our first few months with the system up and running. The lowest the temp has been in the shop in the area of the inverters has been 38F. I live in SW Washington and this year the temps have been down to 16 at night a few times. Whenever there is potential for below 32F outside, we put propane heaters on to be sure we don't drop too cold.

This was another issue I had with them last week which is when I met Mr. Honey who you probably talked to. My batteries are laid out from left (by our shop/garage door) to the right which is in the middle of the shop wall well away from any doors or drafts. I'm putting a pic up for you to see. The top two pieces of insulation were removed for the picture as I was working on the BMS warning light with Fortress. They sit up on a bit of a shelf which is supported by 2x4's for extra strength. Each 5 battery section has 4 pieces of insulation one on the back, one on the front and two to close in the top allowing only gaps for the battery cables. The temperature, according to the thermometer on the wall next to our Sol-Arks (closest to the garage door) said 38F at it's lowest overnight. Anyway, I had shut the system down because of low power. Their stated storage temp is 20F. When I brought the batteries back online, I had 4 of my newer batteries in the middle of the line up (from the left, it was batteries 6-9) showing a BMS warning light (flashing red instead of the normal green). I worried and at the time didn't know how to see what the alarm was so I called Fortress and was walked through how to see the alarm. It was for a warning that the temperature was getting too low. Funny that it was low in the middle when it had been wrapped in insulation all night and it wasn't that cold in the shop! That's how I met Mr. Honey who informed me that I should just let the system stay on so it could keep itself warm. And this is when I asked about their parasitic load.

When we read the specs on the batteries, it says operating temperature min was 32F and storage was 20F. We don't normally have very low temps so we hadn't been worried until these guys told us they needed to be in a warm environment. I was like, "WTF?!"

Nothing is as advertised. And their stated "efficiency" is a fraud if they eat 30 watts.

I will get them back up to full and start my testing on Saturday when my husband is available to work with me.

Again, thank you so much for all of your help!
As I keep digging into this, I'm beginning to get the idea that the problem isn't the batteries.. Of course, without your testing results being available yet, and without any data, I'm relying on a combination of experience, technical know, and social media information.

Social Media
If those modules really had such a high self-consumption, I would think that you would not be the only one complaining about it. Loosing 14% of your power reserves to self consumption would surely show up on someone's radar. Folks who operate off-grid are pretty intensive when it comes to knowing where their energy is going, and I have to believe that someone, more likely everyone, would be raising alarm bells.

Experience
Secondly, any storage device that loses 14% of its own energy reserves PER DAY, wouldn't make it off the engineering table.

Technical Know
I'm guessing you have some other problem, or combination of problems. All battery technologies are based on chemistry, and as such, all batteries lose capacity when they are cold. Lead acid is absolutely horrible and can lose 50% of its charge capacity at 30 degreesF. Lithium is significantly better, but not magically so. Cold temperatures in the low 30's can zap the capacity of a lithium battery by 10% to 15%. So keep that in mind.. It would be a good idea to maybe tape a surface mount thermometer to the case of one module to monitor them. This way, you know if that case is down in the 30 degree range, you can expect a 10% drop in storage capacity. On the upside, cold operation just above freezing will extend the battery life.

My next concern is based on you.. As you have indicated that this kind of stuff isn't your wheel house, I'm going to assume, based on my experience with others in similar situations, that your comprehension of energy draw is probably not up to the requirements needed to manage an off grid system. To make a long story short, I'm going to postulate that your household appliances (refrigerators, heaters, phone chargers, televisions, etc), probably draw a lot more energy than you think they do. You said you ran a 2.5 ton heater? Is that a heat pump? Because that is a MAJOR HUGE load.. gigantic energy hog. Heat pumps are just air conditioners running in reverse and they are the axis of evil for anyone living off grid.

Among my other thoughts are
1) your charge and discharge profile settings in your inverters. You need to make sure your inverter is set to charge your batteries to full voltage. Fortress power recommends a charge voltage of 54.4 volts. A little math says their modules are set up as 16 series units and 54.4 volts divided by 16 = 3.4 volts. That's just a hair on the low side for most folks running phosphate cells. Phosphate cells can be charged to 3.65 with most folks going to 3.6. Kind of odd that Fortress is limiting it to 3.4. I would make sure the settings in your Sol-Arks are set for 54.4 volts and I would make sure to verify that voltage with a multimeter when your cells reach 100%.

2) Shut down one of those Sol Arcs and leave it off for now.

3) Purchase a kill-o-watt meter and check every single thing plugged into your home. I was stunned when I found my fridge pulled a whopping 5kW or more. Cleaning out the coils reduced it to 3.5kW, which was still more than I expected. A kill-o-watt meter will help you identify the energy hogs in your home.

4) Fortress power advertises their modules as 5.4kW, then the specs say 5.374kW (close enough), but their charge voltage of 3.4 per cell does not take advantage of that capacity. I might have to inquire about that being included in the capacity spec or not.

5) And lastly, one unfortunate truth you're not going to like.. and I suspect this is the real problem.. I live in Michigan and winters here are solar horror shows. Our solar array is 12kW in size and we use a 25kWh lithium battery. Between late November to the end of January, our solar system barely makes enough daily power to run the fridge for that day, let alone the computers, lights, television, or a freaking heat pump. There's just no getting around it.. when the days get so short and the angle of the sun gets so low, I'm lucky to see my 12kW solar system making 2kW of power. And while we can generate up to 80kWh in a single day in the spring, I'm lucky to see 5 to 7 kWh's a day during the December-January solar nightmare months.
I have included December's solar generation graph for you so you can see what a 12kW system does in Michigan.. it's kind of pathetic.

You might need to add more solar, or you might have to suck some energy from the grid for those months. One solution we use is to heat with wood, which cuts down our energy use in a huge way.

Something else we did that surprised me.. (and as an engineer a bit embarrassing). In 2021, we replaced all the 30 year old windows in our home with the new energy efficient ones. And WOW.. the difference is staggering.
I bought a thermal rifle scope some time back and I remember looking at my home from a distance.. the outside walls of the house were gray, (that means cold in a thermal scope), while the windows were all bright red and orange, which indicates a heat source. I just thought "hey, its a window, its going to bleed heat, that's probably normal" Then we replaced the windows and I took the scope out to look at them again.. the windows are now gray too.. We were losing crazy amounts of heat through them, and they were heating up the house in the summer forcing more cooling. I have most certainly noticed the difference in our energy monitoring.. it is significant. Just something to think about.


1677824391856.png
 
Last edited:
My next concern is based on you.. As you have indicated that this kind of stuff isn't your wheel house, I'm going to assume, based on my experience with others in similar situations, that your comprehension of energy draw is probably not up to the requirements needed to manage an off grid system. To make a long story short, I'm going to postulate that your household appliances (refrigerators, heaters, phone chargers, televisions, etc), probably draw a lot more energy than you think they do. You said you ran a 2.5 ton heater? Is that a heat pump? Because that is a MAJOR HUGE load.. gigantic energy hog. Heat pumps are just air conditioners running in reverse and they are the axis of evil for anyone living off grid.
Again, the reason the questions came up at all is because, with the breaker for the solar off (meaning not powering loads) and the PUD handling the loads for everything (shop and house), I was seeing loss that didn't make sense. This was after insulating the batteries and adding propane heaters when below 32F outside. This is when the questions really started. When I'm low on battery capacity, I do still have the ability to put everything back on grid and often have this winter.

I've only run my house twice (less than 24 hours) and those results were actually shockingly good. I saw some good weather and had several days of sun in the forecast so I decided to test the ability of the system. Most often this system has only been used to power the uninsulated shop which has a 1/2 HP water pump, three freezers, one refrigerator, and an electric fence. Because the refrigeration is in a colder environment, they're not needing to run as much. They are all grouped into a room where the water pump is. The heat from running the refrigeration in that room easily keeps the water above freezing but not nearly as warm as having them in our heated/insulated house. I have thermometers in that room and next to the Sol-Arks that record the min/max temperature and I've been checking these daily and resetting the min/max so I get a better idea of when I need to add heat based on the forecast to keep everything above freezing.

I do realize our heat pump is a HUGE load. My husband has beat that into my head. I will say that I'm contemplating adding a soft start to the heat pump because, during good weather I'm wasting a ton of energy if I don't run the house off the system but we're still a ways off from doing that as we're still trying to figure out what's going on with the rest of the system. We also have some wire upgrades that need to happen between the shop and the house.

In the summer, I am hoping to run the house in it's entirety as often as possible so I can take advantage of all of the PV we can harvest. This is really why I added the 2nd Sol-Ark because, with just one, we were tripping the system with what seemed like pretty small loads and, frankly, I got greedy seeing all that PV coming in with nowhere for it to go. Initially, this system was half the size (and investment) intended only for the SHTF critical water/refrigeration.

We do have the ability to heat our house with wood. We have a Blaze King that's only a few years old. It's not awesome but will keep us from freezing. We also have a wood cookstove we're trying to figure out how to incorporate. There's just not an ideal place to install it inside the house as it sits and I don't want to waste wood heating it outside of our living space. Our wood supply (especially as we age) will be harder and harder to restock should we not be able to just call someone to deliver it. We have timber on our property but it's not easily accessible due to terrain. So I'm conserving what we do have pretty carefully.


1) your charge and discharge profile settings in your inverters. You need to make sure your inverter is set to charge your batteries to full voltage. Fortress power recommends a charge voltage of 54.4 volts. A little math says their modules are set up as 16 series units and 54.4 volts divided by 16 = 3.4 volts. That's just a hair on the low side for most folks running phosphate cells. Phosphate cells can be charged to 3.65 with most folks going to 3.6. Kind of odd that Fortress is limiting it to 3.4. I would make sure the settings in your Sol-Arks are set for 54.4 volts and I would make sure to verify that voltage with a multimeter when your cells reach 100%.
The parameters were recently reset by Fortress when I had them online fixing the communication issues between the Sol-Arks and the batteries. I will double check these settings again as it does appear their tech support is not all that trustworthy, apparently.

2) Shut down one of those Sol Arcs and leave it off for now.
I have a 4kWH PV array attached to the slave Sol-Ark so at least that one is "paid for" in solar uptake most days. But, if we get skinny on power and cannot go back to grid reliance, the slave will be shut down when not harvesting PV.

3) Purchase a kill-o-watt meter and check every single thing plugged into your home. I was stunned when I found my fridge pulled a whopping 5kW or more. Cleaning out the coils reduced it to 3.5kW, which was still more than I expected. A kill-o-watt meter will help you identify the energy hogs in your home.
I agree this is a "must have" moving forward.

4) Fortress power advertises their modules as 5.4kW, then the specs say 5.374kW (close enough), but their charge voltage of 3.4 per cell does not take advantage of that capacity. I might have to inquire about that being included in the capacity spec or not.
I did notice the small difference in the kW but the other issue is interesting. I'd love to hear their response if you do make that inquiry.

5) And lastly, one unfortunate truth you're not going to like.. and I suspect this is the real problem.. I live in Michigan and winters here are solar horror shows. Our solar array is 12kW in size and we use a 25kWh lithium battery. Between late November to the end of January, our solar system barely makes enough daily power to run the fridge for that day, let alone the computers, lights, television, or a freaking heat pump. There's just no getting around it.. when the days get so short and the angle of the sun gets so low, I'm lucky to see my 12kW solar system making 2kW of power. And while we can generate up to 80kWh in a single day in the spring, I'm lucky to see 5 to 7 kWh's a day during the December-January solar nightmare months.
I have included December's solar generation graph for you so you can see what a 12kW system does in Michigan.. it's kind of pathetic.
I knew we were going to have extremely low uptake in the winter. I'd found a website that allowed me to put in some good parameters that approximated how we had to install our PV. The website told me I'd have less than a single sun hour on average during those months. That's why I wasn't thinking that I could power my whole home during those times. It's not even a pipe dream. And, sadly, our uptake has often been far less than that single sun hour. My fantasy of living grid free was over long before we made the investment in the system that we did.

If we end up with a long term grid down, I have a lot of supplies and redundancy to get us through. I'll be canning my little heart out with what's left of my propane, shutting down the excess refrigeration and I'll be making our water pump and UV light the priority from our solar if that happens and it's winter. In summer, I need the AC to run to keep my temperatures down where I store my canned stuff. Sadly, we have no basement or root cellar which is truly problematic for prepping and a big reason why I decided to double our solar investment.

Something else we did that surprised me.. (and as an engineer a bit embarrassing). In 2021, we replaced all the 30 year old windows in our home with the new energy efficient ones. And WOW.. the difference is staggering.
This house had all new windows installed before we bought it. I'm doubting how well that installation was because I can feel the cold from the window behind me at night. Our front doors are old and leak a good amount of heat as well but I couldn't get my husband to bite off on replacing them. They're big double doors so the cost would run us thousands. If SHTF, I'm literally going to board those doors up because they're a security risk too. lol That's why I bought myself a big ___ self-defense shot gun (which I've never used, don't know how to load and am, frankly, scared of)! ?
 
Again, the reason the questions came up at all is because, with the breaker for the solar off (meaning not powering loads) and the PUD handling the loads for everything (shop and house), I was seeing loss that didn't make sense. This was after insulating the batteries and adding propane heaters when below 32F outside. This is when the questions really started. When I'm low on battery capacity, I do still have the ability to put everything back on grid and often have this winter.

I've only run my house twice (less than 24 hours) and those results were actually shockingly good. I saw some good weather and had several days of sun in the forecast so I decided to test the ability of the system. Most often this system has only been used to power the uninsulated shop which has a 1/2 HP water pump, three freezers, one refrigerator, and an electric fence. Because the refrigeration is in a colder environment, they're not needing to run as much. They are all grouped into a room where the water pump is. The heat from running the refrigeration in that room easily keeps the water above freezing but not nearly as warm as having them in our heated/insulated house. I have thermometers in that room and next to the Sol-Arks that record the min/max temperature and I've been checking these daily and resetting the min/max so I get a better idea of when I need to add heat based on the forecast to keep everything above freezing.

I do realize our heat pump is a HUGE load. My husband has beat that into my head. I will say that I'm contemplating adding a soft start to the heat pump because, during good weather I'm wasting a ton of energy if I don't run the house off the system but we're still a ways off from doing that as we're still trying to figure out what's going on with the rest of the system. We also have some wire upgrades that need to happen between the shop and the house.

In the summer, I am hoping to run the house in it's entirety as often as possible so I can take advantage of all of the PV we can harvest. This is really why I added the 2nd Sol-Ark because, with just one, we were tripping the system with what seemed like pretty small loads and, frankly, I got greedy seeing all that PV coming in with nowhere for it to go. Initially, this system was half the size (and investment) intended only for the SHTF critical water/refrigeration.

We do have the ability to heat our house with wood. We have a Blaze King that's only a few years old. It's not awesome but will keep us from freezing. We also have a wood cookstove we're trying to figure out how to incorporate. There's just not an ideal place to install it inside the house as it sits and I don't want to waste wood heating it outside of our living space. Our wood supply (especially as we age) will be harder and harder to restock should we not be able to just call someone to deliver it. We have timber on our property but it's not easily accessible due to terrain. So I'm conserving what we do have pretty carefully.



The parameters were recently reset by Fortress when I had them online fixing the communication issues between the Sol-Arks and the batteries. I will double check these settings again as it does appear their tech support is not all that trustworthy, apparently.


I have a 4kWH PV array attached to the slave Sol-Ark so at least that one is "paid for" in solar uptake most days. But, if we get skinny on power and cannot go back to grid reliance, the slave will be shut down when not harvesting PV.


I agree this is a "must have" moving forward.


I did notice the small difference in the kW but the other issue is interesting. I'd love to hear their response if you do make that inquiry.


I knew we were going to have extremely low uptake in the winter. I'd found a website that allowed me to put in some good parameters that approximated how we had to install our PV. The website told me I'd have less than a single sun hour on average during those months. That's why I wasn't thinking that I could power my whole home during those times. It's not even a pipe dream. And, sadly, our uptake has often been far less than that single sun hour. My fantasy of living grid free was over long before we made the investment in the system that we did.

If we end up with a long term grid down, I have a lot of supplies and redundancy to get us through. I'll be canning my little heart out with what's left of my propane, shutting down the excess refrigeration and I'll be making our water pump and UV light the priority from our solar if that happens and it's winter. In summer, I need the AC to run to keep my temperatures down where I store my canned stuff. Sadly, we have no basement or root cellar which is truly problematic for prepping and a big reason why I decided to double our solar investment.


This house had all new windows installed before we bought it. I'm doubting how well that installation was because I can feel the cold from the window behind me at night. Our front doors are old and leak a good amount of heat as well but I couldn't get my husband to bite off on replacing them. They're big double doors so the cost would run us thousands. If SHTF, I'm literally going to board those doors up because they're a security risk too. lol That's why I bought myself a big ___ self-defense shot gun (which I've never used, don't know how to load and am, frankly, scared of)! ?
All good information.
I realize it's going to take a while, but I think that in the absence of a verifiable and trusted answer from Fortress, we're going to have to wait until you do that test on a battery module.

One other thing I'm wondering about. When this conversation started, your descriptions centered around State of Charge (SOC) and I'm wondering if the information the Sol Arks were giving you was inaccurate. I say this because it happens to my system all the time. My Sunny Islands have settings to turn off based on a low battery State Of Charge.. and while the setting can be changed, it is based on SOC and not actual battery voltage. That's always been a head scratcher for me because SOC is almost never accurate. Voltage however, absent the hysteresis issues under heavy loads or aggressive charging, is always spot on.

When I first ran my system, we lost lights and power because the inverters thought the battery only had 10% left, and yet my cells were still at 3.55 volts, which means there was still another 25% to 30% left in the battery. That equates to another 6 hours of running for us.

By the way, I almost never pay attention to any information coming from my inverters anymore. The only screen that matters is the data produced by the battery's BMS software that shows actual battery pack voltage and cell level voltages. The SOC meter on the inverter is pretty much useless because it is not reliable.

We'll figure out your problem...
 
An "application engineer" is a fancy term to make an uneducated person feel important in their job.

That's almost as bad as a "Sales Engineer". I know someone that had that title and he often tried to make people think he was a real engineer.

Kudos to @MurphyGuy for doing Yeoman's work in this thread. (y)

The vendor's suggestion to leave the batteries online/connected to keep them warm has some merit. But only if the charge/discharge is significant. A trickle of energy in/out isn't going to raise the temperature of the typical LiFePO4 cell. My DIY LiFePO4 batteries (4s2p EVE 280Ah) will chug along at the same temperature when I have small draws. Say no more than 300 watts. When my Multiplus kicks in with a 120 amp charge the batteries do increase in temperature. But I'm seeing an increase of maybe 1°F to 2°F, not 10°F.

My batteries have a warming system that keeps them between 35°F and 45°F. I'm careful to not put a lot of charge current into the batteries when they're at the bottom of that range. The closer you get to 32°F the lower the charge current should be.
 
That's almost as bad as a "Sales Engineer". I know someone that had that title and he often tried to make people think he was a real engineer.

Kudos to @MurphyGuy for doing Yeoman's work in this thread. (y)

The vendor's suggestion to leave the batteries online/connected to keep them warm has some merit. But only if the charge/discharge is significant. A trickle of energy in/out isn't going to raise the temperature of the typical LiFePO4 cell. My DIY LiFePO4 batteries (4s2p EVE 280Ah) will chug along at the same temperature when I have small draws. Say no more than 300 watts. When my Multiplus kicks in with a 120 amp charge the batteries do increase in temperature. But I'm seeing an increase of maybe 1°F to 2°F, not 10°F.

My batteries have a warming system that keeps them between 35°F and 45°F. I'm careful to not put a lot of charge current into the batteries when they're at the bottom of that range. The closer you get to 32°F the lower the charge current should be.
Yeah, that's about what I see too.. I pump 140 amps into a 25kWh lithium cells from a Chevy BOLT EV, and the temperature might rise 2°F if I do it for a couple hours. It is rare that I draw more than 100 amps out, and if I do, its only for a minute.. so no warming action to speak of.

This 30 watt quote from Fortress has me baffled. Tried reasoning with them and it isn't working.
 
All good information.
I realize it's going to take a while, but I think that in the absence of a verifiable and trusted answer from Fortress, we're going to have to wait until you do that test on a battery module.

One other thing I'm wondering about. When this conversation started, your descriptions centered around State of Charge (SOC) and I'm wondering if the information the Sol Arks were giving you was inaccurate. I say this because it happens to my system all the time. My Sunny Islands have settings to turn off based on a low battery State Of Charge.. and while the setting can be changed, it is based on SOC and not actual battery voltage. That's always been a head scratcher for me because SOC is almost never accurate. Voltage however, absent the hysteresis issues under heavy loads or aggressive charging, is always spot on.

When I first ran my system, we lost lights and power because the inverters thought the battery only had 10% left, and yet my cells were still at 3.55 volts, which means there was still another 25% to 30% left in the battery. That equates to another 6 hours of running for us.

By the way, I almost never pay attention to any information coming from my inverters anymore. The only screen that matters is the data produced by the battery's BMS software that shows actual battery pack voltage and cell level voltages. The SOC meter on the inverter is pretty much useless because it is not reliable.

We'll figure out your problem...
So, you are spot on about the SOC being a problem. When I first contacted Fortress about the substantial loss of power I was seeing based on uptake and beginning and ending SOC, they said there was a recent change that rendered the communication protocol between the Sol-Arks and the batteries useless. Sadly, they did not send out a communication. I had to find this out the hard way by calling. They sent new cables and did some work online resetting things for me and we went on our merry way. I hadn't been even attempting my original analysis that caught the SOC problem since the protocol was changed. It was when we had the BMS warning lights come on I asked about self-consumption when I talked to them and was appalled by their 30 watt response when I did the math. That's when I came back here and asked these questions to see if this was normal.

Again our goal is to survive in a long term, grid down situation. So having a huge draw down like that, assuming we will also not be able to purchase gas for the generator (that is ridiculously undersized and problematic for our system), that huge loss of power won't even keep us in water in the winter. Our water is gravity fed to the property but still has to have a pump to move it through the treatment system.
 
So, you are spot on about the SOC being a problem. When I first contacted Fortress about the substantial loss of power I was seeing based on uptake and beginning and ending SOC, they said there was a recent change that rendered the communication protocol between the Sol-Arks and the batteries useless. Sadly, they did not send out a communication. I had to find this out the hard way by calling. They sent new cables and did some work online resetting things for me and we went on our merry way. I hadn't been even attempting my original analysis that caught the SOC problem since the protocol was changed. It was when we had the BMS warning lights come on I asked about self-consumption when I talked to them and was appalled by their 30 watt response when I did the math. That's when I came back here and asked these questions to see if this was normal.

Again our goal is to survive in a long term, grid down situation. So having a huge draw down like that, assuming we will also not be able to purchase gas for the generator (that is ridiculously undersized and problematic for our system), that huge loss of power won't even keep us in water in the winter. Our water is gravity fed to the property but still has to have a pump to move it through the treatment system.
I am 95% sure I know why they quote that 30 watt figure. When a BMS is active, the little processor that monitors all the cells only draws about 30 to 50 mW of power. Lets be generous and call it 250mW (1/4 of a watt). But when a cell gets out of balance, the processor turns on a bleeder circuit, which is basically a resistor. For all practical purposes, an old fashioned incandescent light bulb and a resistor are the same thing. Current from the cell that has a higher voltage than the rest will bleed through that resistor, which causes the resistor to warm up, just like a light bulb, and causes that cell with the higher voltage to lose some and bleed down in line with the next highest cell.

Each cell in the battery has its own resistor and it is possible, albeit unlikely or realistic, that if the BMS were to turn on a bunch of the bleeder resistors, the total energy being used by the battery could add up to 30 watts. Well, that's probably a big stretch because even being generous I can't see it going beyond 15 watts, but you get the idea.

So in short, I think that 30 watt quote is "The highest possible self consumption figure the battery could ever pull". That is accounting for the unrealistic scenario of many cells being bleed down, the power to run the led lights, and the power to keep the main relay energized.

It still doesn't add up, but if we are generous and stretch the bounds of reality, we can get 15 to 20 watts accounted for.

I'm anxious to know what your test results are.
 
I am 95% sure I know why they quote that 30 watt figure. When a BMS is active, the little processor that monitors all the cells only draws about 30 to 50 mW of power. Lets be generous and call it 250mW (1/4 of a watt). But when a cell gets out of balance, the processor turns on a bleeder circuit, which is basically a resistor. For all practical purposes, an old fashioned incandescent light bulb and a resistor are the same thing. Current from the cell that has a higher voltage than the rest will bleed through that resistor, which causes the resistor to warm up, just like a light bulb, and causes that cell with the higher voltage to lose some and bleed down in line with the next highest cell.

Each cell in the battery has its own resistor and it is possible, albeit unlikely or realistic, that if the BMS were to turn on a bunch of the bleeder resistors, the total energy being used by the battery could add up to 30 watts. Well, that's probably a big stretch because even being generous I can't see it going beyond 15 watts, but you get the idea.

So in short, I think that 30 watt quote is "The highest possible self consumption figure the battery could ever pull". That is accounting for the unrealistic scenario of many cells being bleed down, the power to run the led lights, and the power to keep the main relay energized.

It still doesn't add up, but if we are generous and stretch the bounds of reality, we can get 15 to 20 watts accounted for.

I'm anxious to know what your test results are.
This might irritate you but when I turned the Sol-Arks back on to get the batteries back to 100% I noted a 5% drop in SOC (until hubby gets back that's all I have). If that SOC was correct (beginning and ending) that would mean I dropped around 2.7 kwh in about 23 hours. I too look forward to confirming these numbers.

Would you say the figure showing harvested PV is likely accurate on a Sol-Ark? Until our conversation, I thought all of the Sol-Ark reported data was accurate. I'm pretty sad that it's not. I'm beginning to dislike my choice in inverter as well.
 
My sol- ark 12 is pretty accurate with all the data it reports except SOC, which it calculates. It does as good a job at that as my eg4 batteries, but that’s not saying much.
I agree wholeheartedly with Murphy, you’re going to have to do some measuring and testing to see what’s going on.
I’ve learned to take soc reports with a grain of salt.
 
All 4 of my devices that report SOC differ from each other, the longer between 100% charged day's the worse it gets. Simple voltage is by far the best metric dispite what Will says ?. You just have to remember you are dealing with millivolts here not 1 decimal place.
 
Back
Top