Solexx Powerwall Project

Solexx X

Member
Joined
Nov 16, 2017
Messages
69
I've just started my powerwall project. So far I'm stripping the Molicel 2200mah cells from the modem packs and am experimenting with packs of 160P. I intend to build 14 packs to start so I will have to buy 340+more cells. I intend to use batrium as a BMS and already have charge controllers and a 48 volt 12Kv inverter that is split phase. I have purchased a pallet of Canadian Solar 265 watt panels and can fit 24 of these on my flat garage roof. I'm also working on a wood gas reactor to power a 20kw ICE generator for battery charging. Another related project is a 1kw PMA 48 volt wind generator that is hooked up to a bank of 8 T-105 batteries with a dump load into my hydronic heating system. My plan is to reduce my energy footprint as low as possible.

Any help or suggestions is appreciated. I'll try to post some photos as I progress.

Cheers
Pete
 
Welcome!

12Kv or kW?

If 12kW inverter you need to design your battery pack for up to 2x that. As said you would need MAASSIVE busbars to cope with 300+A :=)
Just have that in mind when doing the design that the battery pack is rather small to that massive inverter. I got 100+kWh of storage to my 15kW system and adding up soon to be 200.
 
JasonMorris said:
Youre going to need some thick bus bars if your doing 160p

I don't intend to go anywhere near 12kW output for now. I may even wire the battery blocks as 80p X 2 with two longmons per block and then rewire them to 160p when I build the second 14s160pbattery.Can't I just use a current limiting breaker/fuse on the output until I double the size?


wim said:
Big plans.... love it !!! :)

Welcome

Thanks!


daromer said:
Welcome!

12Kv or kW?

If 12kW inverter you need to design your battery pack for up to 2x that. As said you would need MAASSIVE busbars to cope with 300+A :=)
Just have that in mind when doing the design that the battery pack is rather small to that massive inverter. I got 100+kWh of storage to my 15kW system and adding up soon to be 200.

12kW, My intention is above 30 kW when completed. I'll have to limit the current output until I getenough batteries for the second 14s160p. My daily energy use is down to about 20kwh and I can store energy credit through net metering in the spring,summer,fall.
The large inverter can't be grid tied so that would require another grid tie inverter and an approved transfer switch to use the grid tie option.
 
SolexxX said:
JasonMorris said:
Youre going to need some thick bus bars if your doing 160p

I don't intend to go anywhere near 12kW output for now. I may even wire the battery blocks as 80p X 2 with two longmons per block and then rewire them to 160p when I build the second 14s160pbattery.Can't I just use a current limiting breaker/fuse on the output until I double the size?


wim said:
Big plans.... love it !!! :)

Welcome

Thanks!


daromer said:
Welcome!

12Kv or kW?

If 12kW inverter you need to design your battery pack for up to 2x that. As said you would need MAASSIVE busbars to cope with 300+A :=)
Just have that in mind when doing the design that the battery pack is rather small to that massive inverter. I got 100+kWh of storage to my 15kW system and adding up soon to be 200.

12kW, My intention is above 30 kW when completed. I'll have to limit the current output until I getenough batteries for the second 14s160p. My daily energy use is down to about 20kwh and I can store energy credit through net metering in the spring,summer,fall.
The large inverter can't be grid tied so that would require another grid tie inverter and an approved transfer switch to use the grid tie option.




12 kW is a big inverter already. You arethinking of 12kW instantaneous use? ie not confusing kWhs (kW used over a time period)vs kW (power used at a giveninstant of time)?

So just thinking some numbers here (don't know where abouts your site isso this might vary a lot based on longitude, etc):
If you've got 24x 265W panels on a flat roof your input per day might be approx 70% (approx flat roof derating)x 265W x 80% (temp derating) x 24 panels = roughly approx 3.5kW per hour of good sun. Say 6 hours = 21.3 kWhrs per day generated on a good day.
If you're using anything like 12kW, the batteries would last about 2 hours......
 
12kW continous inverter can easily push 24kW shorter times! Yes you can limit and you NEED to limit that system alot. Question is do you really want to limit by a fuse that shuts it all down just because of that? On some inverters you can limit Battery usage. Can yours? The MPP solar I got I can set software limits on battery usage when it goes to grid instead. If you can do that it would be a lot better.

Limiting with fuses could cause issues in long run and most systems you dont want the fuse to become a limiter as such unless really needed. You can potentially add limiting system on high loads that for instance if you go above 6kW from the battery the system will disable high current loads untill it goes down again. This to save the battery and not shutting it all down when you hit the max.

Just a tip!

Love to see more images of the setup when you get somewhere :)
 
electrotech said:
SolexxX said:
JasonMorris said:
Youre going to need some thick bus bars if your doing 160p

I don't intend to go anywhere near 12kW output for now. I may even wire the battery blocks as 80p X 2 with two longmons per block and then rewire them to 160p when I build the second 14s160pbattery.Can't I just use a current limiting breaker/fuse on the output until I double the size?


wim said:
Big plans.... love it !!! :)

Welcome

Thanks!


daromer said:
Welcome!

12Kv or kW?

If 12kW inverter you need to design your battery pack for up to 2x that. As said you would need MAASSIVE busbars to cope with 300+A :=)
Just have that in mind when doing the design that the battery pack is rather small to that massive inverter. I got 100+kWh of storage to my 15kW system and adding up soon to be 200.

12kW, My intention is above 30 kW when completed. I'll have to limit the current output until I getenough batteries for the second 14s160p. My daily energy use is down to about 20kwh and I can store energy credit through net metering in the spring,summer,fall.
The large inverter can't be grid tied so that would require another grid tie inverter and an approved transfer switch to use the grid tie option.




12 kW is a big inverter already. You arethinking of 12kW instantaneous use? ie not confusing kWhs (kW used over a time period)vs kW (power used at a giveninstant of time)?

So just thinking some numbers here (don't know where abouts your site isso this might vary a lot based on longitude, etc):
If you've got 24x 265W panels on a flat roof your input per day might be approx 70% (approx flat roof derating)x 265W x 80% (temp derating) x 24 panels = roughly approx 3.5kW per hour of good sun. Say 6 hours = 21.3 kWhrs per day generated on a good day.
If you're using anything like 12kW, the batteries would last about 2 hours......




The roof is flat but the racks will be tilted at 35 degrees winter, spring and fall and 12 degrees summer. It's 4 rows of6 panels. I'm located at 42 North so winter sun is limited. Ideal would be about 60 degree tilt in winter, 40 degree in spring/fall and 12 deg in summer where I live. The 35 degrees is a compromise to prevent shading of the rows behindin December.
 
I just bought some more battery packs from alarmhookup. I now can build a powerwall that is 80p 14s 3p with 3360 cells and with the new cells willaverage about 2300-2400mah. I'm about half way through breaking down the modem packs and cleaning the cells. It looks like i will need 42 longmons to do this for my inverter. I was going to build 160p packs but the longmons can't balance a pack that size according to batrium website info.

Alarmhookup gave me a good discount (30%) so i paid 105USD for the most recent buy of the 2600mah LG cell packs. Since he charges 20$ per box for shipping, it's still quite expensive by the time I bring them into Canada. That said, the quality of these cells seems to be good and I believe that they are new old stock.

Any help with configuration is appreciated. I haven't soldered up the packs/fuses yet since I have many cells to test but I will buy a shi*load of Opus to test the cells. I also bought 100 charging chips to speed up the process. I have a 40 amp 12 volt power supply that I will use to power the Opus and will use a computer power supply for the chargers.I have to go to India in a couple days for work (don't ask) for a week but will be working on this full time in a couple of weeks until its complete.

I see many projects here with evolved ideas that will make the powerwall doable for a non-electrician like me. I'm a power engineer but not an electrician. I'm an old guy so please go easy on me. If you have constructive criticism I welcome it.
 
Hi Powerwall builders,

I have been reading the FAQ on the Batrium site and believe that my 160p X 14s configuration may not balance in a reasonable time frame.

I was looking at Mikes new 300p pack design and was unsure if it would be possible to keep it balanced.
I rewrote this post after posting on Mikes post so I could ask some questions here.

This link "https://www.batrium.com/pages/cell-monit...iderations" from the Batrium FAQsays the ideal for a longmon is 200 Amp/hr. or less so they would be able to balance in a week.

I would appreciate some advice from the builders that are running 80p 14s battery systems as I don't want to find out later that my packs won't balance.

Please take a look at my plan below and comment if you see problems


My new plan is to build the 160p packs electrically split into 2 X80p packs each with their own positive, negative and longmon. Just really 2X80p packs with the holders stacked. I wanted larger blocks so they are more compact. The downside is that I will have to swap out a 160 cell block (2 X80p) if I get a bad 80p cell. The other obvious downside is that I will need 2 longmons instead of one per 160 cells. Another problem is that at present I have 3400 cells so I will have to get another 1100 cells to complete the 28 X 160 blocks. I will also thenneed 56 longmons and that will add a bunch of cost.

I'm looking at the starter pack with shunt, longmons and expansion boardplus the extra 42 longmons.
My question are:
What is the maximum number of cells (Xp) that you would use per lomgmon and why?
Is 80p 14s 4p going to cause any problems that I am not aware of?
Do you think 2 X 8 AWG copper wire (wound) is big enough for my cell busbars?
Does anyone have experience with theMorningstar TS-MPPT-60 TriStar MPPT Solar Panel Charge Controllerfor charging 14s 18650s?
I have 2 of them and 6400 watts of panels that will be dedicated to charging. No grid tie at this time. Just an ATS
Any help insight is appreciated.
 
Solexx X said:
I have been reading the FAQ on the Batrium site and believe that my 160p X 14s configuration may not balance in a reasonable time frame.

1) I'm gonna guess you mean you have 14 packs of 160p connected in series. This is 14s160p
2) What do you mean by reasonable amount of time? Minutes, hours, days? You can increase the effectiveness of balancing by adding more longmons to a pack. So you could have 28 longmons (2 per pack) and it would work just fine.
Also, if you are needing to keep your packs in balance that constantly where you need to burn off 100's of watts with every charge, there's something wrong with 1 or more of your packs. After the initial balance, the work needed to keep them in balance should be minimal.


Solexx X said:
My new plan is to build the 160p packs electrically split into 2 X80p packs each with their own positive, negative and longmon. Just really 2X80p packs with the holders stacked. I wanted larger blocks so they are more compact. The downside is that I will have to swap out a 160 cell block (2 X80p) if I get a bad 80p cell. The other obvious downside is that I will need 2 longmons instead of one per 160 cells. Another problem is that at present I have 3400 cells so I will have to get another 1100 cells to complete the 28 X 160 blocks. I will also thenneed 56 longmons and that will add a bunch of cost.

You could use 2 longmons, yes. But electrically you could use just 1. Every pack that is connected in parallel could all use the same longmon. Again, stated above if your packs are balanced, then continued balancing would be at a minimum.

Solexx X said:
My question are:
What is the maximum number of cells (Xp) that you would use per lomgmon and why?
Is 80p 14s 4p going to cause any problems that I am not aware of?
Do you think 2 X 8 AWG copper wire (wound) is big enough for my cell busbars?

A) The number is basically unlimited. The greater number of cells comes a harder time balancing. However, if the packs are balanced, then this shouldn't be a big problem. (ad nauseam from repetition, I know)

B) It shouldn't cause any issues overall. Electrically, you should be fine. There are several people who do this. However, if you have multiple chemistries, then each chemistry needs to have it's own BMS. Don't mix LiFe banks with LiIon banks. They have different characteristics.

C) 2x8awg is fine. Most people use this size that use the twisted copper wire method.
 
Korishan said:
SolexxX said:
I have been reading the FAQ on the Batrium site and believe that my 160p X 14s configuration may not balance in a reasonable time frame.

1) I'm gonna guess you mean you have 14 packs of 160p connected in series. This is 14s160p
2) What do you mean by reasonable amount of time? Minutes, hours, days? You can increase the effectiveness of balancing by adding more longmons to a pack. So you could have 28 longmons (2 per pack) and it would work just fine.
Also, if you are needing to keep your packs in balance that constantly where you need to burn off 100's of watts with every charge, there's something wrong with 1 or more of your packs. After the initial balance, the work needed to keep them in balance should be minimal.


SolexxX said:
My new plan is to build the 160p packs electrically split into 2 X80p packs each with their own positive, negative and longmon. Just really 2X80p packs with the holders stacked. I wanted larger blocks so they are more compact. The downside is that I will have to swap out a 160 cell block (2 X80p) if I get a bad 80p cell. The other obvious downside is that I will need 2 longmons instead of one per 160 cells. Another problem is that at present I have 3400 cells so I will have to get another 1100 cells to complete the 28 X 160 blocks. I will also thenneed 56 longmons and that will add a bunch of cost.

You could use 2 longmons, yes. But electrically you could use just 1. Every pack that is connected in parallel could all use the same longmon. Again, stated above if your packs are balanced, then continued balancing would be at a minimum.

SolexxX said:
My question are:
What is the maximum number of cells (Xp) that you would use per lomgmon and why?
Is 80p 14s 4p going to cause any problems that I am not aware of?
Do you think 2 X 8 AWG copper wire (wound) is big enough for my cell busbars?

A) The number is basically unlimited. The greater number of cells comes a harder time balancing. However, if the packs are balanced, then this shouldn't be a big problem. (ad nauseam from repetition, I know)

B) It shouldn't cause any issues overall. Electrically, you should be fine. There are several people who do this. However, if you have multiple chemistries, then each chemistry needs to have it's own BMS. Don't mix LiFe banks with LiIon banks. They have different characteristics.

C) 2x8awg is fine. Most people use this size that use the twisted copper wire method.

Thanks for the reply. I'm using recycled 18650s that are a mixture of cells from 2000mah to 2700mah so the use of a longmon on each 80p pack will allow me to balance and monitor each pack. Did you get a chance to look at the Batrium FAQ that addresses the pack size issue?I want to be able to swap out a twin-pack (2X80p packs in one holder not electrically connected directly to each other)with a spare twin-pack easily if a pack fails so I don't have to shut down a string of 14s. It seems that almost everyone on youtube who is using the 80p/100p and 14s format has had a bad pack for one reason or another and although the extra longmons add significantly to the price I may go that way since I want to use a larger inverter. The larger number of cells total (4480) should keep the individual battery amps below 1amp. I want to finish this project by summer and end up with a reliable 30kWh+ of storage that I don't have to babysit. I have lots of other retirement projects and thankfully most of the battery mistakes have already been uncoveredin this Forum so I can copy the successes of others.
Cheers,
Pete
 
I want to be able to swap out a twin-pack (2X80p packs in one holder not electrically connected directly to each other) with a spare twin-pack easily if a pack fails so I don't have to shut down a string of 14s.
Any setup that uses parallel packs that are connected in series will be put offline when 1 of the packs is pulled out of the string. The only way around it is to have a "fly pack" as some people refer to it. You connect it to the string before you disconnect the problem child. Then you have 14s+1 per string and 14+1 longmons for that string.

The only way to not effect a string is to have packs that are system voltage and then parallel all the packs. But this does add a heap of complexity and monitoring issues.

I still don't fully understand the reason for the split-parallel pack in your design. Perhaps a diagram could better facilitate the transference of ideas.
 
Back
Top