My Open Source Solar BMS and Digital MPPT

electrodacus

New member
Joined
Dec 10, 2018
Messages
14
I'm new here and this is my first post.
Over the last few year I designed an open source Solar BMS and had them funded successfully trough Kickstarter currently I'm on the third generation of Solar BMS and running my forth Kickstarter. While I read the rules of the forum not sure if is OK for me to link to my Kickstarter page so I will not do that.

Solar BMS is an all in one BMS + solar PV charger + advanced energy monitor.

The main reason I started designing this about 7 years ago was that I was moving offgrid and needed some energy storage solution for my house. I looked at Lead Acid that is even today used by most offgrid people and found that storing energy in those was more expensive than I will have liked so looked in to Lithium as a possibility and that was the time that A123 started advertising their LiFePO4 that just looked more than perfect for what I needed.
Since there was no BMS for LiFePO4 I decided to do designsomething myself and started looking for best solution. Just at the same time Intersil announced the ISL94203and that was exactly what I needed and started designing something based on that even if only the spec was available at that time and I was the first one to order sampled when they where available.
I was thinking it will be a relatively simple design and just build a unit for my own use but as I started working I realized it was way more work than estimated :) and as I shared the idea and prototype on my youtube channel other people become interested so I was going to build a few for them also so that was when I did the first Kickstarter.
From there it evolved to new more advanced levels and on last Kickstarter I added a completely new product the DMPPT450 again taking way more time than initially estimated but it was worth the trouble as now it fully heats my house (DMPPT may be completely different from what you imagine so read the user manual for more details).

I will be happy to answer any SBMS or battery related questions.

I think LiFePO4 is far superior for energy storage than LiCoO2 or NMC that are designed for high energy density to the detriment of cycle life and that is why I selected LiFePO4 for my own use.
Typical LiCoO2 and NMC are good for 500cyles at 100% DOD (of course a bit more if you limit the DOD) and while they are great for both portable electronics and EV's they are to expensive for typical stationary energy storage applications.
When I say expensive I refer to cost amortization overthe usable life of the battery and not the initial equitation price.
Typical LiFePO4 are good for 3000 to 6000 cycles at same 100% DOD so an order of magnitude better and more than adequate for energy storage applications.

There are two main mechanisms for battery degradation one is the known cycle life and the other less known but even more important is the calendar aging.
Typical calendar aging is in the region of 0.5% to 2% capacity loss per year dependent of the quality of the additives added to batteries and manufacturing quality but also very dependent on cell temperature. As a general rule a 10 degree Celsius increase in cell temperature will double the rate of calendar aging degradation.
I did a precise capacity test when I installed my 8s10p A123 battery about 4.8kWh usable and I repeated the same test after a year of full time use in my offgrid house.

So over 8544h (that is about 356 days) the house used a total of 1122.349kWh + 373.944kWh = 1496.293kWh that is an average of 4.2kWh/day or 126kWh/month
Out of all that 806.714kWh went trough battery so in average day 2.27kWh went out of the battery and that is about 47% of the battery capacity that is around 4.8kWh

The result in capacity loss was just 0.9% and large part of this is due to normal aging maybe 0.6% and just a smaller part maybe 0.3% is due to cycling. The reason that calendar aging was so low has to do with the quality of A123 cells but also low internal impedance and low charge/discharge rate in this application 0.25C max charge and 0.5C max discharge thus battery temperature was never more than 2 to 3C above ambient and ambient inside my house where battery are is +18C to +26C all year.

I have a lot to say about this subject but I will leave it at that for the first post :)
 
I remember your first Kickstarter, and to this day I regret not going it.

I will be participating on this one!
 
CrimpDaddy said:
I remember your first Kickstarter, and to this day I regret not going it.

I will be participating on this one!

Thanks. Let me know if you have any questions. There are 3 models now the current SBMS120 and SBMS40 and there is a new low cost SBMS0 that as the name imply is just BMS + energy monitor it has no internal charger. But there are more details on the Kickstarter page.
 
Hey, welcome bud -
 
Hey Buddy!!!! Welcome (fellow Canuck Here) When I first started to look into BMS I saw your Kickstarter. but Was not ready to get into it that deep. YET!!!!
 
Thanks Mike and hbpowerwall,
I see you both have large capacity batteries 10 to 15x the capacity of my battery. What is that huge battery capacity used for ? Offgrid or grid tie ?
I will need to have a 160kWh for heating but that will be the way more inexpensive thermal storage for now I only have 97kWh but not quite enough for heating.
jdeadman you just posted before me :) I see you also have a fairly large battery capacity still a bit more moderate.
I also noticed you all used recycled cells how did you got all those 16850 ? seems like a lot of work.
 
electrodacus said:
Thanks Mike and hbpowerwall,
I see you both have large capacity batteries 10 to 15x the capacity of my battery. What is that huge battery capacity used for ? Offgrid or grid tie ?
I will need to have a 160kWh for heating but that will be the way more inexpensive thermal storage for now I only have 97kWh but not quite enough for heating.
jdeadman you just posted before me :) I see you also have a fairly large battery capacity still a bit more moderate.
I also noticed you all used recycled cells how did you got all those 16850 ? seems like a lot of work.

It is a lot of work. Many users on this forum have built large batteries from second hand cells. Check out the cell counter. Some people use new cells, others have managed to source cheap scrap laptop and ebike batteries. Some are even lucky enough to have sourcednew old stock laptop batteries.

Many users have youtube channels detailing their builds. Just to list a few:

hbpowerwall
DIY Tech and Repairs
Mike's Lithium Solar Channel
 
Hey Thanks for Check it out. Yea I'm getting there in size. Right now only in the building Phase. I currently have 5Kwh Tested and Soldered. The rest is still loose cells.

I started like many here getting a couple boxes of cells from work or friends then the bug hit and I needed more. started trolling around looking for deals and landed 2 boxes of unopened cells that were about 75% good. I got up to about 500 Cells that were tested and just resting to check for Self Discharging cells.

Then I stumbled upon a FB group member looking to get rid of his stash of over 2000 cells mostly processed and just needed to be tested. Really cheap so I could not pass it up. He was loaced just south of the border from me here so took a drive one weekend.

Right now I have about 15Kwh tested and above 2000mAh and a pile from 1600-2000 as wells as a box of 1k-1599 So I have lots to play with.
 
The reason I ask why you have such large capacity batteries is that I can not imagine how they will be useful.
Also regarding recycled cells the problem is that this small 18650 where not designed for stationary energy storage but for energy density.

Most of you have used cells in the 2000mAh range but when new those cells where around 2400 to 2600mAh that means they are very close to the end of usable life so if new they where good for 500cycles and if now in average they are just good for another 100 cycles the amount of energy you will be able to store in it before completelyunusable may be 3.7V x 2Ah x 100cycles = 740Wh so say is even more to be a round 1kWh per cell then the value of that cell is best case whatever you pay for the grid energy say $0.2/kWh minus the cost of energy going in to the cell say 3 to 5 cent if is from PV solar so each cell value may be 15 cent.
Now you will likely need to get that cell form somewhere maybe a laptop pack then test the cell that is a full charge discharge cycle maybe to see the capacity and maybe internal impedance and then you will need to put that in a pack so some sort of tab welding.
I know you may do this as a hobby but you will need to do all the above for like 100 cells to get back $15/hour and that will be impossible as you can not get the cell make a full test and build in to a pack in just 36 seconds per cell (just impossible).
This means there is absolutely no economic value and you do this just as a hobby only.

For me it needs to be an economic aspect so thus I use LiFePO4 as they will last for way longer.
Like my small 4.8kWh was around $2000 and I take out of them in average around 2.3kWh/day x 365 days x 15 years (may last 20 years but I will use 15years for this calculation) = 12592kWh over usable life so
$2000 / 12592kWh = $0.16/kWh so still not cost effective to use a grid storage in most places in the world but as good as it gets for offgrid where it is needed.

Thermal storage for example can have a cost amortization as low as $0.01/kWh but is only good for heating applications.
 
electrodacus said:
Most of you have used cells in the 2000mAh range but when new those cells where around 2400 to 2600mAh that means they are very close to the end of usable life so if new they where good for 500cycles and if now in average they are just good for another 100 cycles

Yes, this is true, mostly. This is assuming we are using the cells from 4.2V to 2.7V full range. However, most are using the cells from 4.0V to 3.2V, which drastically increases their cycle life. Hence the reason for so many cells in the packs.
Daromer has been running at least 1 string for almost 4 years (if I remember correctly) now. There's others who have ran theirs for 2 or more years as well.

We have to remember that the life expectancy of the cells is dependent on how hard we charge/discharge them, as well as what the working range in voltage is. These two factors helps us to get far more life out of the cells than what the datasheet states.

However, you do have a valid point. This is one reason why I am going to be building my packs with ease of replacement and monitoring so I know which cells need to be replaced. And this will be for new or used cells.
 
here lifepo4 18650 from scap..(only 100 cartridge 5P ready of 1000)

i ll take a look of Your BMS because i like ti have items that do only One job.

great prpject!
 
Welcome.
 
Korishan said:
Yes, this is true, mostly. This is assuming we are using the cells from 4.2V to 2.7V full range. However, most are using the cells from 4.0V to 3.2V, which drastically increases their cycle life. Hence the reason for so many cells in the packs.
Daromer has been running at least 1 string for almost 4 years (if I remember correctly) now. There's others who have ran theirs for 2 or more years as well.

We have to remember that the life expectancy of the cells is dependent on how hard we charge/discharge them, as well as what the working range in voltage is. These two factors helps us to get far more life out of the cells than what the datasheet states.

However, you do have a valid point. This is one reason why I am going to be building my packs with ease of replacement and monitoring so I know which cells need to be replaced. And this will be for new or used cells.

When you say cycle life you always need to mention the depth of discharge but for those type of cells the rule is close to being proportional so say DOD is 50% then you will have 2x the cycles at that DOD but then total energy stored over the life time will be the same if you do 100 cycles at 100% DOD or 200 cycles at 50% DOD or 2000 cycles at 5% DOD

If you barely use the battery then they will last even 10 years the degradation over that period in this case will be mostly related to calendar degradation so cost amortization may be even higher.

This high amortization cost for high energy density cells is also the reason why an EV like Tesla will be more expensive than an equivalent ICE
If you look at the warranty of model 3 long range that has the new cells (supposedly better).
Model 3 has a 192000km or 8 year warranty whatever comes first and is based on a 80kWh high energy density NMC
That model 3 has a range of 500km on a full charge (almost 100% DOD excluding the small top and bottom reserve)so 192000km / 500km = 384 cycles
Now the old cells if you look at Panasonic spec sheet is capable of 500 cycles at 100% DOD and by that time it will only have 70% of original capacity.
The warranty also claims 70% of original capacity in 8 years or 192000km thus it means they took in consideration the worst case that is 384 cycles about 90% DOD as there is some 5% top and bottom limits. People will also not drive like that they will likely charge to 80% and almost never get close to 0% SOC so they should be fine with around 15% or so degradation from cycling and the other 15% degradation for aging assuming around 2% per year calendar degradation that will be typical for EV due to higher battery operating temperature.
That 80kWh battery still cost around $16000 (that excludes the BMS and other stuff around) thus you pay that extra compared to an ICE car and when you consider that extra cost it will be $16000 / 192000km = $0.083/km or put in another way $8.3/100km and that is more than what my inefficient 6 cylinder ICE will require in gasoline cost. That is just the battery amortization cost not counting the electricity to charge the EV. This will be about the same for all EV's Tesla was just an example.
Put in a different way $16000 / 70kWh x 384 cycles = $16000 / 26880kWh = $0.59/kWh and that will be the cost amortization for that battery over the life time of the battery.

If they could offer a better warranty they will do so but they can not as those are the limitations of the specific battery technology. That is more than adequate for the average driver so is a good choice of battery for the specific application but as you seen in my earlier comment LiFePO4 can do much better at just 0.16cent/kWh so LiFePO4 is much better for stationary energy storage.
LiFePO4 will not be so great for an EV as is about 2x lower energy density so you will have a car that costs the same but has half the battery capacity and half the range so people will not by this. BYD uses LiFePO4 in their buses and taxis since those type of vehicles will need to put much more km than a consumer personal car.

For me cost amortization over the life time of the product is very important and I calculate that for anything.

That is how my small energy efficient65m^2 (700sqft) housecost a total of $42 / month in therms of cost amortization for both electricity and PV electricheating about half of that amount for each and it is the most cost effective solution even when compared to natural gas.
 
Eletrodacus: Same as I do. Its always about total energy or cost over time that is essential and not if they will do x or y cycles or they have x or y capacity but over time capacity or cycles :)

Community is still a bit young and many still stare alot over cycles or voltage ranges or capacity in cells but forget about how much it will yield over set amount of time.
 
Electrodacus. Maybe i am blind but i dont find a detailed description of tthe bms0 somewhere... can you pls give me a hint!

Thx
Karl
 
electrodacus said:
That is how my small energy efficient65m^2 (700sqft) housecost a total of $42 / month in therms of cost amortization for both electricity and PV electricheating about half of that amount for each and it is the most cost effective solution even when compared to natural gas.

Slightly off topic...
I linked to another forum where electrodacus is talking about his house here, where Dan is alsotalking about going off-grid in sunny Wales.
 
For me it's about using what is going to be discarded. I ran my entire Camp all summer on 3Kwh of Storage with only 200W of Solar but Since I was only out there 4 days out of 7 it was all I needed. this was my test bed. 7s56p and it worked flawlessly. This bank also ran 80%DOD All last winter before taking it out to camp. so 8 Months of 80%dod and 4 months of 30% dod and the cells were just finished being re-tested and all but 9 cells were the same capacity or better than originally tested.

So this leads me to believe that as long as you are using good testing cells that are fairly close to their rated capacity a low dod Wall should last for a long time.

Now this is just a hobby right now for me. but in the next 10 years or so I plan of buying some land and building a retirement home which will be offgrid. I have been researching what batteries I want to use and since it's so far off I have time to play now.

Just a side note the 8 months I had my camp wall running at home I only ran my fridge, freezer and Internet in a time shifting mode. where during Peak pricing It ran off my wall and during low cost power I charged that bank. No solar at all. I used what I had gotten for free for everything other an the busbars, solder and battery holders. ( I think I spent ~100$ on that and test equipment) and I was saving about $20/month just doing that. That savings was the difference from last years bill to this years bill where the prices did not change much.
 
Back
Top