Kens's diy powerwall build

Ken140tdi

Member
Joined
May 16, 2017
Messages
81
ok so continuing on from my introduction post as I think this is the proper place to expand on what I want to do and learn from you more experienced guys at the same time.

Ok just to quickly go over what I already have.

I have a 6 kw and a 4 kw solar edge system ofsolar Pannels. 6 of the panels are on a solar tracker I built my self and it was well worth it for the 30 to 40% extra power generated from them.
I have also just bought a victron mult 5kw with 5 kWh of byd b boxbattery storage. The idea was to use this as a base battery system and then build my own batteries with the eventual aim of having 30 kWh plus of storage.
I'll try to post some pics or at least links to pics later to show what I've done.


Right. The battery storage I have so far consists of 2 2.5 kWh byd batteries. These have their own bmu or bms for controlling the charge etc. I have them set up so that each battery has its own main cable back to the victron multi. The multi does all the charging by converting ac to dc. It also inverts the dc back to ac when battery power is required.
Mybatteries reached fully charged for the first time today at a voltage of 56.4 volts. I believe the spec on the byd battery means it can produce power to quite a low voltage and the bmu does not switch off the discharge until 47 volts I think from memory but I'll double check this.
So from what I'm told an 18650 is fully charged at 4.2 volts and fully discharged at 3.4 volts. But I have also read that fully charging to 4.2 and fully discharging to 3.4 is bad for the battery and will shorten its life.
Knowing that batteries will self ballence from the highest cell to the lowest I was thinking I could use the byd batteries bmu to monitor the two byd batteries between their charged 56.4 volts and empty 47 volts then rig in my home built batteries. The bmu controlled b box batteries should automatically keep my own built batteries in check as the current should just flow between them from highest to lowest.

Now I know that the b box batteries are fully charged at 56.4 volts so if I build 14 packs of lion cells when the byd batteries are fully charged each of my own cells should be charged to 4.02 volts and when the byd batts are discharged to 47 volts my lion cells would be at :.357 volts which is a little low so I'm pretty sure in the set up of the victron the upper and lower voltage drain and charge cutoffs are configurable so I could raise the cut off drain voltage to 47.6 volts which would stop the std lion cells from being discharged too far.
What do u think guys.
 
Hi Ken,

nice to hear from you about your project with Victron and BYD B-Box.
In other thread Daromer told you all the basics about mixing the chimestries of battery cells.
It is realy a matrix of dependencies you have to watch for: differnt voltage levels, differnt charging curents, differnt charging durations and cylces, and so on. I'm also working in mecahrtonics and am a little bit older than you and have collected some experience in ...
Well, if you want to see your project as a hobby than the energy storage is (the only?) a hobby that reduce your costs - is realy so.
If you want to use your stoarage to work with in your shop, then I would say just extend your BID system with additional racks to max. 80kW and you are fine.
Everithing is possible in electronic - that's true. Till today nobody created (or I do not know) such a box where you can connect lead acid, Li-Ion, Ni or LiFePo together and store/use your energy. May be one day in the very next future some one guy will do it - but me not: no time.

To your idea: how do you want to connect your modules to a U3A1-50P-A module? In parallel? Well, what will be the reaction of the system?
In its firmware it knows the capacity of 2.5kW module and you expand this capacity by "shadowing" it with your banks let say to infinite...
What will happen when the system starts to think it is an error and stops to work? Can you interact with BMU?
You have a CAN-bus there. Can you send commans/parameters to BMU? This are the question I'm asking.


I do not want to discourage you but the questoins should be asked before you start with doing/changing sothething on your system.
You told that the firmware of BID should be updated - I hope for you that the chines guys do it. My experience with them is that after you bouth something from them, they gorget you.
Victron is juts totaly differnt - they never let you fall. You will get support.

Well, there is lot to say, but I will stop.
Wish you lot of succes and stay tune with you...

BigBen
 
Hi Ben.
You have some good points there.
I'm doing it to save money and get more off grid as it were. The biggest point you make about just buying more byd batteries is fine apart from the cost as each pair of 2.5kwh batteries costs 2500 which just means I would not live long enough to recoup the outlay.
Regarding the byd update to the firmware. The whole saler is helping me directly because they are concerned that they have sold a kit which victron day is compatible and the reality is that it's not unless the byd batteries and bmu are newer than march 15th 2017 and that was at the point of production. So I'm sure they will sort it for me.
The next thing you mention is a U3A1-50P-A module. Forgive me for asking but what is this.

At the moment as the byd bmu is not working I have just left it unplugged from the communications ports and set the victron to use its own internal battery monitoring system set for lipo4 and manually configured the charged and discharged voltages to what byd specs are.
Each byd battery has its own doc lights that are indipendant of the control box so I can see the state of charge manually.
I was planning to make a 14s battery with 40 cells in each pack each with its own tesla style 5amp blow fuse. If I use a 14s set up when the byd batteries are fully charged at 56.4 volts the home made individual packs cells should all be at 4.02 volts which I think is just under fully charged. I also saw somewhere that if you don't fill an 18650 cell to the brim or completely flatten it you will make it last longer.
I also found yesterday that when the byd batteries are drained to the std shut off of 10% they are at 49 volts so on my home made cells that should be 3.5 volts.
So I was hoping to use the byd management system to manage the byd batteries and then because my home made batteries will be conected to the byd ones the byd system will kind of indirectly manage my batteries too.
If I then found and bought some sort of separate monitoring system that would tell me the voltages of my individual packs I would then be able to keep a good check on them.

Hopfully my ideas are good but I'm hoping you guys here will either say that's fine or no u need to do it this way.
 
Its worth noting that combining different types is not an issue if... :)

For instance a hybrid inverter have a central connection where all is hooked up. solar, battery, grid, load. Its just a matter of raising or sinking the voltage and it will start to flow in either direction. Think of it as a water pond with a couple of pipes attached... :)

So basically you could do the same on batteries. ie have an junction point where all send or take energy from. bi directional dc/dc :)
Then you have 1 unit per battery type that controls charging/discharging to this point based on set of values.


Of some reason i could not move/copy my post here from the other one (Need to check that later)

But here is my original post regarding mixing chems:
http://secondlifestorage.com/t-Hi-from-south-west-England?pid=5533#pid5533

Perhaps should compile some kind of faq or just separate thread about it later on
 
Ken140tdi said:
The next thing you mention is a U3A1-50P-A module. Forgive me for asking but what is this.

Hi Ken,

there is nothing to forgive ;)

So which system do you have? Is it this one?


image_tghrlj.jpg


And this is that U3A1-50P-A module:


image_ewasnk.jpg


Pls confirm or make some pics.

If you have that module than this has an active electronics inside and the output voltage is in the range you mentioned.
To access the cells individually you have to open the module. And maybe you will lost in that way the warranty - who knows.
And there is no sence to go inside the module - maybe only to take pictures ;-)
The modules are daisy-chained with a CAN-bus and you can not influence the BMU by yourself from outside. For you it is indeed a "black-box".
But... in the cabinete you have a plate for additional connectors of U3A1-50P-A boxes - plus and minus. This connectors can be used to connect yours batteries - as an assembly of 14s40p.
May be you first try/verify this plan with used laptop cells and after you can buy yourself new cells.
It is enough to start first with 14s10p and see what happens.


BigBen
 
Hi guys. Thanks for the help on this.
Ben the bmu that is inside the b box cabinet is the same unit that I have it's just a wall mount system and not in the cabinet so instead of daisy chaining them to a bus bar in the cabinet you just link the dc terminals in a daisy chain and then connect the communication from the bmu. The batteries and bmu are exactly the same as is fitted in the b box cabinet.
As far as I can tell with my limited knowledge if you were to connect another battery to the dc terminals of the byd battery the central bmu would not know the difference as it's charging and discharging parameters are done on voltage So it would just take longer to charge as there would be more capacity. (I HOPE LOL )
 
Daromer. I've re read you post from yesterday and now understand what you mean. My only question at this point is the upper and lower voltages of the lifepo4 batteries.

Byd give a spec on their byd box batteries of fully charged at 56.4 volts and the cut off for the discharge is 47 volts so on that basis is I build 14s batteries I should be with in the normal working levels of the 18650 cells
 
Above voltage must be 16s. That do work with lion but as i said. It wont hardly use the lion capacity unless you drag ur life to the end :)

I also added general top voltages. You need to stick both soc curves on top of each othet and then ypu will see how many % of the other chem will be used. For instance if u use 40% in middle of life you will only use 5% of the lion....
 
Could u explain further please as I'm clearly getting something wrong here. My calculations are this
Byd battery full charge is 56.4v. 18650 in a 14s CFL would be 4.02 v on each cell or pack so not quite fully charged.
Byd battery flat is supposed to be 47v. And that should put a 14s pack of 18650 cells at 3.357volts so depending on cell specs about right for almost flat.

I can't work out why your saying I need a 16s battery.
 
Ken140tdi said:
Could u explain further please as I'm clearly getting something wrong here. My calculations are this
Byd battery full charge is 56.4v. 18650 in a 14s CFL would be 4.02 v on each cell or pack so not quite fully charged.
Byd battery flat is supposed to be 47v. And that should put a 14s pack of 18650 cells at 3.357volts so depending on cell specs about right for almost flat.

I can't work out why your saying I need a 16s battery.

Hi Ken,

I would also suggest to start with 14s. Yes, your edges will be 4.02 and 3.375 volts. Depending on the capacity of your battery your voltage delta will be even between two volts and this delta can be placed between even less that one volt.
One example from my real situation: Vmin is about 3,7xx and Vmax is about 3,9xx.
It seems there is almost a balance between charged and used energy. In that range the cells can do more than some hunderts of cycles.
Just start to experiment and see what happens :)

Can you take some pictures of your system?

BigBen
 
Im saying that the LiFe pack that you are talking about must be 16s to get thos voltages.

16s LiFePo4 almost matches 14s LiCo (laptop cells in general)
 
Here are a few photos of my solar inverters and my byd batteries. It's all in the crawl space under my house and because it's underground it states quite cool. As you can see the byd batteries are mounted one above the other. Each battery has its own separate main dc cable connecting to the victron multi. I'm then going to add batteries in pairs so add one battery to each along the wall so I'll have two strings.
image_zmahxl.jpg

image_mvclve.jpg

image_anejza.jpg


Could some one please check my calculations and tell me if I've got this right.
Below is the spec of the 18650 I'm thinking of using.


Batteries Brand: LG
Batteries coding: 18650 MG1
Nominal capacity: 2850 MAH
Batteries Specifications: 65 * 18.4 mm
Batteries resistance: 50 milliohms or less
Batteries Voltage: 3.6V
Test current: 0.2C
Charging cut-off voltage: 4.2V
Discharge cut-off voltage: 2.75V
Maximum discharge current: 10A

So if I want to make a 10 kWh battery is the capacity the total of just one pack or the total of all 14 packs.
I thought 2850mah x 40 cells is114,000mah. Which I thought was 10 kw but I'm thinking this is only 1 kw.
So basically I'm asking how many mah there is in 1 kWh
 
If I am understanding you correctly, you want to use build 14 packs of 40 cells @ 2.85 aH each cell?

This would equate to 3.7 nominal voltage a pack * 2.85 aH nominal capacity each cell * 40 cells giving each pack 473.6 wH * 14 packs in serial 7.1 total KWh.

I could be wrong, but this is what I believe.
 
Thanks for the info Robert.
Ok so my calculations are completely off lol.
I think in light of this and the info I have learned over the last few days I would be better to build two separate 5 kWh batteries so that I keep it all ballenced. This would slow me to addd the same amount of extra capacity to each of the byd batteries.

So my next question is what is the formulae to work this out.
Could you show me how it's worked out please.
 
Im not sure what you want to work out?

Each cell has 3.6*2.85= 10.26Wh. So if you got 40 cells in each bank and 14 of them in the string you have
10.54*40*14 = 5.74kWh (And if we make this a more useable number its more likely 4kWh of useable energy)

Robert must have missed a number somewhere?

Above batteries are normal LiIon batteries ie same as in laptop cells.
 
Thanks daromer. That's it. I wanted to know how to work out how many kWh my battery would be so what you have posted is what I needed to know.
:)
 
Right. I understand how to work this out now.
So it's calculated on the nominal voltage. And I get the same answer if I count each pack as 1 cell. So one pack has 114 wh. And the nominal voltage of the complete battery will be 50.4.
50.4 x 114 = 5.745 kWh.
So I need pretty much double the number of cells to make my two 5kwh batteries.
 
daromer said:
Im not sure what you want to work out?

Each cell has 3.6*2.85= 10.26Wh. So if you got 40 cells in each bank and 14 of them in the string you have
10.54*40*14 = 5.74kWh (And if we make this a more useable number its more likely 4kWh of useable energy)

Robert must have missed a number somewhere?

Above batteries are normal LiIon batteries ie same as in laptop cells.

It looks like I stand corrected, 3.6v is the nominal voltage.... Not sure were I got 3.7v, just too many different voltages listed on too many chargers.

note to self: Place yellow sticky to remind myself :)
 
daromer said:
Im not sure what you want to work out?

Each cell has 3.6*2.85= 10.26Wh. So if you got 40 cells in each bank and 14 of them in the string you have
10.54*40*14 = 5.74kWh (And if we make this a more useable number its more likely 4kWh of useable energy)

Robert must have missed a number somewhere?

Above batteries are normal LiIon batteries ie same as in laptop cells.

It looks like I stand corrected, 3.6v is the nominal voltage.... Not sure were I got 3.7v, just too many different voltages listed on too many chargers.

note to self: Place yellow sticky to remind myself :)
 
On a side note how do you guys get hold of so many used laptop packs to tear down and get all your cells.
I've been asking around a few computer shops and some friends that do lap top repairs but not had much luck
 
Back
Top