BMS Suggestions for 16s3p 280ah CATL

[xTratrestrial]

Long awaited, 50x CATL 280ah will be hitting my door within the next month. I still need to make a BMS order and am hoping to get some good suggestions on the best way to manage these in 16s3p. I have several questions that I'm still unsure of after digging through quite a few threads.

I've pretty much landed on using 3x Overkill 16s 100a for each of the series groups. One of the primary things I'm still unsure of is the importance of individual parallel group balancing. Should I be using a 4th BMS for this, or perhaps a balancer, or do I not need to worry too much about it? I'm aware that voltage indifferences should balance themselves across parallel groups since a lagging pack would offer less resistance, but what about longevity? What if one pack, at some point, for whatever reason, degrades faster and there becomes a slight capacity mismatch? I would think this could be noticed easier and managed better if there were another BMS/balancer on the 3p groups.

Another thing I'm trying to consider is future growth. If I were to add to the bank later on, if I wanted add used cells/different chemistry or simply add on a 4th parallel group with 150ah cells instead of 280, can this be done without negative repercussion if there is a BMS/balancer managing the parallel groups?

Finally, if anyone has suggestions to toss in regarding compression, I'd love hear opinions here as well. I haven't began to dig into that topic yet, but I do know that some folks are using boxes or other structures while others are simply using Kapton tape on these cells. Reading shows that CATL typically have thinner walls than other manufacturers, so I feel like leaning to the side of some sort of tight fitting box my be prudent, but I'd really like to hear from others with experience.

 

[mrzed001]

Many use the JK Heltec 200A Active BMS. A bit more expensive but it is a BMS and an active balancer too.
If you do the initial top balance correctly the parallel strings will work fine.
Not a good idea to add different chemistry or different Ah LiFePO4 string to the battery.
If you put a 150Ah string parallel to 280Ah string(s) then 150Ah will go empty faster and then the 280Ah strings will charge it until they go empty too. Messy.
And when you charge the battery the 150 Ah string will be full fast and then you float this cells until 280Ah strings are full too. Even more messy.

If there is a case that is ready with compression ... I am interested too

 

[Steve_S]

Couple of things to clarify.
You want to build 3, 48V/16S Battery Packs to put into a Paralleled Battery Bank.

Paralleling Cells within a Battery Pack can be done but ONLY with Properly Matched & Batched Cells, NOT with Bulk/Commodity cells. Most vendors say they are Matched, THEY ARE NOT, no Handheld tester like a YR1035 is capable of such. Most vendors sell Voltage & IR Matched at storage voltage only.

Paralleled packs will split/share Load & Charge capacity between packs.
Each Battery is "Independent" unless a Distributed BMS system is used (very $$$$ not needed for ESS).
Each Battery Pack should be setup to be able to operate as "The Last Soldier Standing". Meaning as a Standalone capable of accepting 100% Charge & Discharge capacity of the system. This is for Failsafe / Fail-Over purposes to ensure overall reliability.
Each Battery Pack to Common DC Bus should use the exact same length & gauge if Battery Cable (both Pos & Neg).
Each Battery Pack should be Fused to the Max Capacity Handling. 280AH cell pack, 275A/300A fuse is acceptable.

IN Parallel the "Battery Packs" will level each other during the charge cycle and they will flutter up & down during charge (normal) and Post Full charge they will settle and level up.

Cell Balancing is internal to the Battery Pack itself and does not affect external packs.
- Passive Balancing is only effective on lower AH cells which are perfectly Matched, Batched & Binned. It also only burns off Hi Voltage from cells.;
- Active Balancing with a 5A Capacitance Balancer will transfer Hi Volt to Lo Volt cells and generally can level up cells to <10mv differential.
- Very FEW BMS' offer Active Balancing at this time and they come at quite a Premium. Additionally, the higher the Amperage of an Active Balancer increases cost exponentially.
REF: http://liionbms.com/php/wp_passive_active_balancing.php

Known Good Active Balancers:

This product is no longer available.

This product is no longer available.

cncdheltec.en.alibaba.com

Alibaba Manufacturer Directory - Suppliers, Manufacturers, Exporters & Importers

Alibaba Manufacturer Directory - Suppliers, Manufacturers, Exporters & Importers

deligreen.en.alibaba.com

My Production Packs have QNBBM's and my Utility Packs have the Heltecs.

One of Andy's Videos with a Capacitance Balancer, is pretty straight forward.

Adding Packs in Parallel.
- You can add several without issue BUT be forewarned that after 5 Packs things get a tad more complicated.
- Do NOT have a wide AH differential between packs. Do keep within 50AH difference to reduce issues.
- The Lower AH packs will reach Low State quicker which will cause the larger packs to try & back charge, thereby sacrificing a percent of capacity in doing so.
- During Charge, the Lower AH Pack will attain FULL SOC 1st. Pending on Charge & BMS profile this could result in that Pack being cutoff and then remaining Amps will be pushed to the larger Packs. OR if poorly configured, it could result in HiVoltDisconnects due to Cell Over Voltage trips. (very bad).

A 48V System @ 250A Draw can output 12,000W. The general unwritten rule is to NOT exceed 250A draw to a single inverter system. If more amperage for Watts is required, Stackable Inverters & SCC come into play.

PLEASE Review the Links in My Signature and do download the resources, they will help you going forward.
Hope it Helps, Good Luck
Steve

 

[xTratrestrial]

Many use the JK Heltec 200A Active BMS. A bit more expensive but it is a BMS and an active balancer too.
If you do the initial top balance correctly the parallel strings will work fine.
Not a good idea to add different chemistry or different Ah LiFePO4 string to the battery.
If you put a 150Ah string parallel to 280Ah string(s) then 150Ah will go empty faster and then the 280Ah strings will charge it until they go empty too. Messy.
And when you charge the battery the 150 Ah string will be full fast and then you float this cells until 280Ah strings are full too. Even more messy.

If there is a case that is ready with compression ... I am interested too

This is what I've always thought. Logically it's what makes sense. It seems like mismatched parallel strings used together would cause discharging to the lower capacity strings as they discharge quicker. The larger strings would "balance" the smaller strings by charging them as voltage drops and you'd be able to use the extra capacity from the smaller strings, but the constant discharging from larger strings to keep the entire bank balanced would probably wear hard on them over time and drastically shorten cycle life expectancy. Then, there are probably issues with charge/discharge and internal resistance as well, but that's a whole other story. Just wanted to confirm what I've always believed to be the truth in everything I've read thus far.

As far as compresssion goes, I'm wondering what other folks are building or using, not so much a ready to go case. From what I understand, it's pretty important to keep these cells compressed, no? They can have major issues if allowed to buldge that will lead to shortened cycle life expectancy from what I've read.

 

[mrzed001]

This is what I've always thought. Logically it's what makes sense. It seems like mismatched parallel strings used together would cause discharging to the lower capacity strings as they discharge quicker. The larger strings would "balance" the smaller strings by charging them as voltage drops and you'd be able to use the extra capacity from the smaller strings, but the constant discharging from larger strings to keep the entire bank balanced would probably wear hard on them over time and drastically shorten cycle life expectancy.

Also could happen that the smaller becomes over discharged. And overcharged. Both are bad.

Then, there are probably issues with charge/discharge and internal resistance as well, but that's a whole other story. Just wanted to confirm what I've always believed to be the truth in everything I've read thus far.

As far as compresssion goes, I'm wondering what other folks are building or using, not so much a ready to go case. From what I understand, it's pretty important to keep these cells compressed, no? They can have major issues if allowed to buldge that will lead to shortened cycle life expectancy from what I've read.

Everyone is building wooden frames and rods to compress them together.
I would love to see a solution where you put 4-8-16 EVE/LISHEN/CATL 200-280-310Ahcells in a box, fix them with screws, busbars, BMS, cable management ... and ready.
Like Seplos does in this (case?) product:

 

[xTratrestrial]

Couple of things to clarify.
You want to build 3, 48V/16S Battery Packs to put into a Paralleled Battery Bank.

Paralleling Cells within a Battery Pack can be done but ONLY with Properly Matched & Batched Cells, NOT with Bulk/Commodity cells. Most vendors say they are Matched, THEY ARE NOT, no Handheld tester like a YR1035 is capable of such. Most vendors sell Voltage & IR Matched at storage voltage only.

Yes, you are correct. I will be building 3, 48v/16s packs that I will place in parallel with each other to create one bank. Also, they will not be going into an already existing bank. I will not be trying to enter parallel-series wire cells in any configuration where it would look like 3x paralleled 3.2v cells all going into one series connected battery. I can see where that could be an absolute mess.

Paralleled packs will split/share Load & Charge capacity between packs.
Each Battery is "Independent" unless a Distributed BMS system is used (very $$$$ not needed for ESS).
Each Battery Pack should be setup to be able to operate as "The Last Soldier Standing". Meaning as a Standalone capable of accepting 100% Charge & Discharge capacity of the system. This is for Failsafe / Fail-Over purposes to ensure overall reliability.
Each Battery Pack to Common DC Bus should use the exact same length & gauge if Battery Cable (both Pos & Neg).
Each Battery Pack should be Fused to the Max Capacity Handling. 280AH cell pack, 275A/300A fuse is acceptable.

Thank you. A couple good reminders and clears up a few things. What would the benefit of a distributed BMS be? I'm sure I don't need to go down that path for this build, but I'd like to know what the benefit would be. My assumption by the name is that it would balance all cells across all 3 packs to ensure they all match identically rather than trying to match all cells internally to each independent pack. I can't see where the benefit would lay in doing this though, as the 3 paralleled packs would balance themselves without assistance anyway by nature.

Also, could you define ESS?

IN Parallel the "Battery Packs" will level each other during the charge cycle and they will flutter up & down during charge (normal) and Post Full charge they will settle and level up.

Cell Balancing is internal to the Battery Pack itself and does not affect external packs.
- Passive Balancing is only effective on lower AH cells which are perfectly Matched, Batched & Binned. It also only burns off Hi Voltage from cells.;
- Active Balancing with a 5A Capacitance Balancer will transfer Hi Volt to Lo Volt cells and generally can level up cells to <10mv differential.
- Very FEW BMS' offer Active Balancing at this time and they come at quite a Premium. Additionally, the higher the Amperage of an Active Balancer increases cost exponentially.
REF: http://liionbms.com/php/wp_passive_active_balancing.php

How much energy is lost via passive balancing using something like an Overkill BMS vs. an active balancer?

Known Good Active Balancers:

This product is no longer available.

This product is no longer available.

cncdheltec.en.alibaba.com

Alibaba Manufacturer Directory - Suppliers, Manufacturers, Exporters & Importers

Alibaba Manufacturer Directory - Suppliers, Manufacturers, Exporters & Importers

deligreen.en.alibaba.com

My Production Packs have QNBBM's and my Utility Packs have the Heltecs.

One of Andy's Videos with a Capacitance Balancer, is pretty straight forward.

I see that the Heltec balancer is 5a. Is that the same 5a rating as thinking about a passive balancer at 100a or is 5a simply the current the capacitors are able to hold for discharge from one cell to transfer to another? I'm probably not understanding the differences between a balancer and a BMS properly. Would an active balancer typically offer all the other monitoring/conditional options that most of the passive BMS out there do? I'm going to watch the video you posted after I finish this reply and I'm sure that will probably clear this up a little.

Adding Packs in Parallel.
- You can add several without issue BUT be forewarned that after 5 Packs things get a tad more complicated.
- Do NOT have a wide AH differential between packs. Do keep within 50AH difference to reduce issues.
- The Lower AH packs will reach Low State quicker which will cause the larger packs to try & back charge, thereby sacrificing a percent of capacity in doing so.
- During Charge, the Lower AH Pack will attain FULL SOC 1st. Pending on Charge & BMS profile this could result in that Pack being cutoff and then remaining Amps will be pushed to the larger Packs. OR if poorly configured, it could result in HiVoltDisconnects due to Cell Over Voltage trips. (very bad).

A 48V System @ 250A Draw can output 12,000W. The general unwritten rule is to NOT exceed 250A draw to a single inverter system. If more amperage for Watts is required, Stackable Inverters & SCC come into play.

This clears up a ton regarding my thoughts on future additions to the pack. That means that in several years, I need to capacity test if I'm planning to add to the pack and ensure that my pack hasn't lost more than 50ah capacity if I plan to enter a 4th identical 280ah pack as the first 3 into the bank. It also clears up questions that I had in wondering where the "rules" lay within whether it's okay to do any mixing at all, which sounds like it would be a pretty significant mistake, which is what I assumed.

PLEASE Review the Links in My Signature and do download the resources, they will help you going forward.
Hope it Helps, Good Luck
Steve

Thank you for your time. This has already been very helpful!

 

[Bob B]

If you want to get up to speed on cell compression .... this thread will give you all the info you want.

EVE-280 cells should these be clamped tight or spaced for expansion?

I working on my 4s 280ah battery plans. I thinking of my battery box. should these cells have a air space between them or be bonded tight together with a hose clamp or something similar.

diysolarforum.com

If you are really serious about using compression to extend your pack life .... then the following thread might interest you ..... see the 2nd post in that thread for some shortcuts to relevant information about compression.

Pack / Cell compression Optimized By Using Springs.

I am opening this thread to split it off from the general cell compression thread. That way those looking for info don't have to sort thru a thread that has many paths for pack compression. If you are new to this idea, it stemmed from an EVE cell specification that indicated a significant...

diysolarforum.com

 

[Steve_S]

What would the benefit of a distributed BMS be?

When getting into such large banks, you will be requiring interaction with Inverter/Charger systems for proper management, These typically use CanBus protocols to change charge rates and voltages plus controlling each BMS (which have more capabilities). Battrium BMS is an example of Distributed.

Balancer Type info:
Passive balancing slowly burns off voltage from Hi Volt cells within a pack, typically at 1.2A or less.
Active transfers it from Hi to Low and at a %a Capacity as I suggested (is enough) but you can get 10A transfer capacity (much quicker & reactive) but you go from $45 to $400
WhitePaper.

Li-Ion BMS - White Paper - Dissipative vs. nondissipative balancing (a.k.a.: Passive vs. Active balancing)

ESS = Energy Storage Systems