Author Topic: Battery balancing - big cells  (Read 1452 times)

0 Members and 1 Guest are viewing this topic.

Offline Miyuki

  • Regular Contributor
  • *
  • Posts: 145
  • Country: cz
Battery balancing - big cells
« on: April 18, 2018, 09:22:14 pm »
Hi,

I want to add battery to solar system as prices already drop to level when it can be cheaper to have battery than buy electricity from grid at night

Main goal is to keep cost as low as possible so simple passive balancing/monitoring like provide LTC6811 looks goods and little decrease in charging efficiency is not a big problem

I wan to ask for experiences and advices with this systems as I will use big cells with 58Ah each
 

Online MosherIV

  • Frequent Contributor
  • **
  • Posts: 945
  • Country: gb
Re: Battery balancing - big cells
« Reply #1 on: April 19, 2018, 05:57:38 am »
Quote
I will use big cells with 58Ah each

 :o

Do you mean cells or do you mean a module?
A quick search did not find any 58Ah cells but did turn up modules.

Yes, you MUST keep the cells balanced. This simply means keeping the cell voltages equal.

Normally, you get a Battery Managment System to do this for you.
Designing your own is not that easy, there are many parameters and effects to account for.
Eg cell voltage droop and recovery, when you draw current from the battery the voltage will drop as it drains but when the current draw stops, the voltage will recover a little. This must be account for.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #2 on: April 19, 2018, 01:12:12 pm »
Even Tesla uses resistive balancers. If the battery pack is in any good shape, the balancing rarely even comes into play.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 1000
  • Country: fi
Re: Battery balancing - big cells
« Reply #3 on: April 21, 2018, 03:19:44 am »
Yes, you MUST keep the cells balanced. This simply means keeping the cell voltages equal.

You are confusing balancing, which is (theoretically) only for maximizing energy storage capability of the aging pack, with cell-level monitoring, which is (again theoretically) for safety and protection. I could consider accepting the answer that the latter is a "must", but the former is clearly not.

Talking about "keeping the cell voltages equal" is nonsense. What is equal, anyway? Within 0.0001V? Within 0.1V? Even in a perfectly balanced pack, voltages will be significantly different over the operating range. For example, if the pack is balanced at 100% SoC as usual, then there are going to be significant voltage differences (>0.1V) at, say, 5% SoC, under considerable (over 0.5C, for example) load.

If this balancing is a must, why it's not universally used? Why most power tools don't do it?

That's about balancing. Then, even the cell-level monitoring is not actually a hard requirement; definitely not a generic "must". Companies such as BOSCH do not do it on many of their 6s and smaller li-ion packs.

Though, for such big cells (or paralleled cell groups - doesn't matter as long as the parallel connection is securely fixed during manufacture of the pack), cell-level monitoring could be important, but even more than the cell size, it depends on the number of cells connected in series. The more you have cells in series, the less relative contribution to the total voltage from any single cell. For example, almost no one uses cell-level monitoring on 2s packs, for which there seems to be total 100% industry expert agreement that it can be considered as a single 7.2V cell and managed as such. Above 2s, the viewpoints start to spread.

Now, you had one thing right - designing a BMS is not easy. It feels like an easy task, but the devil is in the implementation details. This can be easily seen by analyzing existing BMS's. Every single one I have seen has been faulty by design, and possibly dangerous. The most typical issues are: 1) high quiescent currents, deep discharging and killing the cells in typical use cases that the designer still didn't anticipate, 2) lack of really necessary protections due to the limited lithium ion chemistry knowledge of the designers; instead, all kinds of fancy "we-think-we-need-this" features are implemented, 3) stuck on balancing resistors, either just killing the cell, or worse, when combined with not-designed-for-worst-case thermal misdesign, heating up the adjacent cells over the thermal runaway onset temperature (about 150 degC for LCO).

I have designed my own, as well, and sold it in fairly small numbers for suitable clients for specific purposes, and I'm quite darn sure I still have some bugs there, as well :). In any case, the correct order to design a BMS is:
1) Make sure it doesn't heat up or catch fire in any possible use or misuse scenario - do the thermal analysis assuming it's coupled with the cell, then thermally insulated for little cooling effect
2) Make sure you have correctly learned and identified all critical safety parameters of the cell chemistry. Unfortunately, forum posts and easy-to-digest fake information sites such as Battery University are not enough, you need to attend some courses or read enough research material on actual cell chemistry.
3) Implement monitoring on these critical safety parameters and react by shutting down any operations if these are exceeded. Make sure you have this completely working before implementing anything else
4) Implement non-critical "comfort" features such as balancing. Make sure these do not prevent point 3) from working.

Last but not least, modern li-ion cells from the brand manufacturers are very well protected on chemical and physical level. This is necessary, since the industry experience has shown that the BMS's do not protect reliably due to so much misdesign and misuse; and even if correctly designed, there are limits on what they can do.


« Last Edit: April 21, 2018, 03:22:42 am by Siwastaja »
 

Offline woodchips

  • Frequent Contributor
  • **
  • Posts: 426
  • Country: gb
Re: Battery balancing - big cells
« Reply #4 on: April 21, 2018, 03:39:22 am »
Since it is stationary why not be sensible and use lead acid batteries? Get one of those cells up to thousands of Ah!
 

Online MosherIV

  • Frequent Contributor
  • **
  • Posts: 945
  • Country: gb
Re: Battery balancing - big cells
« Reply #5 on: April 21, 2018, 08:09:31 am »
Quote
You are confusing balancing, which is (theoretically) only for maximizing energy storage capability of the aging pack, with cell-level monitoring, which is (again theoretically) for safety and protection. I could consider accepting the answer that the latter is a "must", but the former is clearly not.

Talking about "keeping the cell voltages equal" is nonsense. What is equal, anyway? Within 0.0001V? Within 0.1V? Even in a perfectly balanced pack, voltages will be significantly different over the operating range. For example, if the pack is balanced at 100% SoC as usual, then there are going to be significant voltage differences (>0.1V) at, say, 5% SoC, under considerable (over 0.5C, for example) load.

Ok, I admit I did over simplify on my statement. Let clarify.
Balancing is a MUST  when many cells (say >5) are in series to prolong life.
Why? To avoid the potential of reverse biasing single cells in the stack.
In extremely large stacks it is essential to make sure that all cells are closely matched in charge and discharge rates. Further more they must be kept balanced (equal in voltage) to stop one cell discharging before others and therefore being reverse biased. Reverse biasing leads to over discharging the cell and the voltage drops to 0V.
Once 1 cell goes to 0V, it brings all other cells around it down for the same reason.

How balanced do they need to be? I am not sure I can say for commercial reasons.
I currently work for a battery company developing 400V to 700V systems.


Quote
If this balancing is a must, why it's not universally used? Why most power tools don't do it
Depends on the application and cost of the packs.
Power tools are commodity, so are made to the lowest price possible. So long as the packs last the 1 year warranty period, after that it is expected that the user bears the cost of replacement.

The high power, high voltage systems I work on have warranty periods of around 5 years, so it is worth making sure they have mechanisms to prolong life. The extra expense of the measuring and balancing circuits is cost effective in this application.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #6 on: April 21, 2018, 10:34:28 am »
Since it is stationary why not be sensible and use lead acid batteries? Get one of those cells up to thousands of Ah!

Lead acid is not that good for regular cycling. LiFePO4 is the way to go for regular cycling.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Offline Gregg

  • Frequent Contributor
  • **
  • Posts: 321
  • Country: us
Re: Battery balancing - big cells
« Reply #7 on: April 21, 2018, 10:56:39 am »
When dealing with larger capacity batteries, lead acid still gives the best bang for the buck.  All of the big data center UPS units and all of the telco terminals use lead acid batteries. The secret to long life is maintenance, proper sizing and proper charging.  If lead acid batteries cannot get a full charge on a daily basis, or if they are discharged too deeply, their life and capacity will diminish.  Lead acid battery banks do not require separate cell monitoring or balancing; there are lots of UPS systems with 500 volt battery banks without individual cell monitoring or balancing.
Rule of thumb with lead acid is not to discharge them below 1.75 volts per cell and to recharge them as soon as possible.  A minor downside is the initial voltage drop upon loading lead acid batteries; google coup de fouet.
In any case some research before spending lots of money is always prudent.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #8 on: April 21, 2018, 11:32:13 am »
All of the big data center UPS units and all of the telco terminals use lead acid batteries.
Those don't cycle the batteries by any significant amount every day or even every week. Keeping the batteries at full charge until needed for backup power is about the ideal use case for lead acid. Contrast that to lithium which prefers to stay around half charge.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Online ikrase

  • Regular Contributor
  • *
  • Posts: 106
  • Country: us
Re: Battery balancing - big cells
« Reply #9 on: April 21, 2018, 04:52:55 pm »
I am actually somewhat suprised and frustrated by the lack of really good BMS options for medium and large cells. Closest would be the Electrodacus system.

 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #10 on: April 21, 2018, 05:00:11 pm »
I am actually somewhat suprised and frustrated by the lack of really good BMS options for medium and large cells. Closest would be the Electrodacus system.
They're pretty easy to find for common 4.2V lithium and 3.6V LiFePO4. What's tricky is LiMn which is fully charged at around 4.1V, most commonly in the form of Nissan Leaf battery modules. I ended up developing my own BMS using a PIC for my 4S pack.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 1000
  • Country: fi
Re: Battery balancing - big cells
« Reply #11 on: April 21, 2018, 10:09:59 pm »
Since it is stationary why not be sensible and use lead acid batteries? Get one of those cells up to thousands of Ah!

Because lead acid battery capacity per actual kWh has been far more expensive than lithium ion for many years now, especially the complete lifetime cost, at least by factor 2-3.

The fact that some large backup packs still use (or were using a few years ago; this is changing as we speak) lead acid is not because the lead acid itself makes much sense, but because of tradition and supply chain. Some big things change slowly: something needs to be cost-effective and reliable for many years if not a full decade before being accepted as a replacement for a traditional low-tech usage.

Li-ion has been considered a luxury thing, so that the pack constructor / maintainer company can add 500-1000% price over the bare cell price, while the equivalent lead-acid constructor / maintainer only adds, say, 200-500%. IMHO, this has already been changed in the recent years.
« Last Edit: April 21, 2018, 10:16:32 pm by Siwastaja »
 

Offline Siwastaja

  • Super Contributor
  • ***
  • Posts: 1000
  • Country: fi
Re: Battery balancing - big cells
« Reply #12 on: April 21, 2018, 10:11:38 pm »
Balancing is a MUST  when many cells (say >5) are in series to prolong life.
Why? To avoid the potential of reverse biasing single cells in the stack.

I suggest you reread my reply, or study elsewhere. Cell voltage monitoring (specifically, per-cell Low Voltage Cutoff) is for this purpose. Balancing is not, and it cannot even address this problem at all!
« Last Edit: April 21, 2018, 10:17:07 pm by Siwastaja »
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Battery balancing - big cells
« Reply #13 on: April 21, 2018, 10:25:42 pm »
I am actually somewhat suprised and frustrated by the lack of really good BMS options for medium and large cells. Closest would be the Electrodacus system.
They're pretty easy to find for common 4.2V lithium and 3.6V LiFePO4. What's tricky is LiMn which is fully charged at around 4.1V, most commonly in the form of Nissan Leaf battery modules. I ended up developing my own BMS using a PIC for my 4S pack.

Whats so difficult on a 4.1V LiMn battery? :-O There is a metric shitton of fully programmable BMS ICs on the market.

To the OP: I have no personal experience with LTC6411, but it is a successor of the LTC6804 which I have a lot of (not the greatest) experience.  We did use these for a couple of larger 300 to 400V LiIon battery packs. The LTC6804 did not behave as it should have a lot of times, we got several BMS burned for no apparent reason (did work at the time of installation absolutely fine, only to find later it fucked itself for no clear reason).  The LTC6804 are discontinued and I think my experience with these just tells the reason. As the LTC6811 should be compatible if I remember correctly, I guess it is a corrected revisit of the 6804 and the 6811 should be more or less fine.

However we switched recently to using the new 16cell BQ series chip from texas instruments. So we'll see how these will do in the near future.
 
The following users thanked this post: Siwastaja

Online MosherIV

  • Frequent Contributor
  • **
  • Posts: 945
  • Country: gb
Re: Battery balancing - big cells
« Reply #14 on: April 22, 2018, 05:13:13 am »
Quote
Quote
     Balancing is a MUST  when many cells (say >5) are in series to prolong life.
    Why? To avoid the potential of reverse biasing single cells in the stack.


I suggest you reread my reply, or study elsewhere. Cell voltage monitoring (specifically, per-cell Low Voltage Cutoff) is for this purpose. Balancing is not, and it cannot even address this problem at all!

Yes, you are right that there is a min cell voltage cut off BUT you do not want that to shut down the battery pack during operation because one cell is lower voltage than the rest. That is why balancing is a MUST for systems where you have many cells in series.

If you do not beleive that cell balancing is necessary, ok.
I am try to explain why it is necessary in a civil curtious manner.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #15 on: April 22, 2018, 05:18:13 am »
Whats so difficult on a 4.1V LiMn battery? :-O There is a metric shitton of fully programmable BMS ICs on the market.
Hard to find a reasonably priced module for a 4S pack because there's little demand. The PIC is doing a lot of other things with power management so I needed it in my design anyways.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Battery balancing - big cells
« Reply #16 on: April 22, 2018, 09:48:51 am »
Well finding an off the shelf module for them may be hard for sure.
 

Offline woodchips

  • Frequent Contributor
  • **
  • Posts: 426
  • Country: gb
Re: Battery balancing - big cells
« Reply #17 on: April 22, 2018, 07:09:37 pm »
Ok, so where are Li-ion batteries being cheaper than lead acid obtained?

Just done a quick check at a distributors prices and lead acid is far cheaper. Cheapest Li-ion batteries were DeWalt 54V 6Ah strangely.

A forklift battery must be the cheapest lead acid, and they have a life measured in years, mine must be 30 years old, down to 1/5 of rated capacity though.
 

Online ikrase

  • Regular Contributor
  • *
  • Posts: 106
  • Country: us
Re: Battery balancing - big cells
« Reply #18 on: April 22, 2018, 10:55:13 pm »
The initial cost is still much higher.
 

Online MosherIV

  • Frequent Contributor
  • **
  • Posts: 945
  • Country: gb
Re: Battery balancing - big cells
« Reply #19 on: April 22, 2018, 11:10:31 pm »
Quote
Ok, so where are Li-ion batteries being cheaper than lead acid obtained?

Hi, someone else has already pointed out that we are in the transition period, where LiPo are still at a premium price.

It will also depend on application. Unlike SLA, I doubt that you will be able to buy raw modules like a SLA batteries, any Li based battery module will require some kind of BMS. Think of laptop batteries, they all come as a pre sealed module with the BMS embedded.

The company I work for is already developing replacement modules for SLA batteries in electric pallet trucks. They are complete modules and the end user has no access to the batteries inside. The manufacturer of the electric pallet truck is driving this change over. The volume price of LiPo is now slightly cheaper than SLA. Ironically, for pallet/forklift the weight of SLA is an advantage, the LiPo modules need ballast weight in the housing.

The need for the BMS is because when Li battery is abused, they vent gas which is highly dangerous so the cells must be protected from abuse.

It will be a while before Li battery modules are available as COTS items that you can just order.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 3934
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: Battery balancing - big cells
« Reply #20 on: April 23, 2018, 12:08:27 am »
Ok, so where are Li-ion batteries being cheaper than lead acid obtained?

Just done a quick check at a distributors prices and lead acid is far cheaper. Cheapest Li-ion batteries were DeWalt 54V 6Ah strangely.

A forklift battery must be the cheapest lead acid, and they have a life measured in years, mine must be 30 years old, down to 1/5 of rated capacity though.

Cheaper *after accounting for cycle life*.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

"Get Naomi Wu back on television screens around the world or shoot earnhoney mining difficulty to the stars trying! Put those shiny new Ryzens and Coffee Lakes to good use!"
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3031
  • Country: us
Re: Battery balancing - big cells
« Reply #21 on: April 23, 2018, 01:57:42 am »
A few relevant points:

While it is true that when one accounts for total cycle life, lithium chemistry batteries are cost competitive or better in daily cycle applications, deep cycle lead acid batteriess still have one key advantage: they are very robust and will tolerate extremes of temperature and charge/discharge that will kill lithium batteries.  This gives them an advantage in some home solar power applications where temperature extremes may occur and power/weight is not a big issue.

If you choose to go with lithium, large, prismatic LiFePO4 cells with capacities of 100-200 Ah are the battery of choice for serious DIY home systems IMHO. Yes some people are making large DIY “Tesla power wall “ style packs out of the ubiquitous small 18650 LiPo cells, but that is still a niche, labor intensive, unproven approach with risk of catastrophic failure.

The large LiFePO4 prismatic cell approach was pioneered by DIY EV builders some time ago and in the last few years is becoming very popular for home PV and other renewable energy systems.

The issue of cell balancing large capacity lithium battery packs has been a topic of debate for some time. As the OP notes it is difficult or impossible to find BMS systems up to the task.

The good news is that it turns out that contrary to conventional wisdom,  for these system a BMS is not needed and may actually make things more prone to failure. The answer is to bottom balance the cells.

Bottom balancing has become the method of choice. You can read more about the rationale and concept HERE and HERE.
 

Offline Yansi

  • Super Contributor
  • ***
  • Posts: 1673
  • Country: 00
  • STM32, STM8, AVR, 8051
Re: Battery balancing - big cells
« Reply #22 on: April 23, 2018, 02:33:34 am »
Quote
deep cycle lead acid batteriess still have one key advantage: they are very robust and will tolerate extremes of temperature and charge/discharge that will kill lithium batteries.

Proof? Why Lithium based chemistry is the choice number one for the coldest of temperatures?

Quote
If you choose to go with lithium, large, prismatic LiFePO4 cells with capacities of 100-200 Ah are the battery of choice for serious DIY home systems IMHO. Yes some people are making large DIY “Tesla power wall “ style packs out of the ubiquitous small 18650 LiPo cells, but that is still a niche, labor intensive, unproven approach with risk of catastrophic failure.

Again, any proof of such statement? Stacking cylindrical Li-Ion based cells with NCA chemistry can yield the highest energy storage density, both volumetric and by weight when properly done.  Stacking cylindrical cells is much safer and easier.

It seems like you are talking out of your ass with these ones. Otherwise you have my approval on the rest.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3031
  • Country: us
Re: Battery balancing - big cells
« Reply #23 on: April 23, 2018, 03:54:37 am »
Quote
deep cycle lead acid batteriess still have one key advantage: they are very robust and will tolerate extremes of temperature and charge/discharge that will kill lithium batteries.

Proof?

It is the experience of many eBike and EV builders that charging Lithium batteries below freezing temperature causes irreversible damage. I personally damaged my eBike LiFePO4 pack that way.  This does not happen to lead acid batteries.

There is a reason why the Auto companies now building EVs are almost all including some active temperature management to prevent both low and high temperature extremes.  My Volt will automatically turn on the ICE to warm the battery pack when charging in very low temperatures. Why do that if it is not an issue?

Nissan's well publicized failure to provide active cooling to the Leaf led to many premature battery pack failures.

If you want data here is some. I'm sure much more can be found with a search.

Quote
Why Lithium based chemistry is the choice number one for the coldest of temperatures?
Proof?   ::)

I suspect your confusion is with the difference between performance at low, non freezing temperatures versus their robustness - which means their ability to withstand very low temperature charging or very high temperature discharging without sustaining irreversible damage.  THAT is what I was referring too and why I specifically used the term "robust".

The fact -based on many decades of experience - is that deep cycle flooded lead acid batteries can withstand abuse in both extremes of temperature and charging/discharging without suffering irreversible damage. Period. Full stop. The same simply cannot be said of lithium batteries. Why else all the concern about cell monitoring, balancing, and use of BMS for lithium and not for LA?

That said - Lithium batteries do have many advantages which is why they are replacing LA in most applications.

Quote
Quote
If you choose to go with lithium, large, prismatic LiFePO4 cells with capacities of 100-200 Ah are the battery of choice for serious DIY home systems IMHO. Yes some people are making large DIY “Tesla power wall “ style packs out of the ubiquitous small 18650 LiPo cells, but that is still a niche, labor intensive, unproven approach with risk of catastrophic failure.

Again, any proof of such statement? Stacking cylindrical Li-Ion based cells with NCA chemistry can yield the highest energy storage density, both volumetric and by weight when properly done.  Stacking cylindrical cells is much safer and easier.

Highest energy storage density is not always the most important design consideration. In particular that is not true for home renewable energy applications. This is why using large prismatic LiFePO4 cells is the best solution there. Their energy storage density is not quite as good as other lithium chemistries but their other advantages (including safety) outweigh that.

It is a fact that stacking large numbers of 18650 or similar cylindrical cells and utilizing electronic BMS introduces more failure points. BMS is not easy to do correctly. Experience has shown that this can result in catastrophic failures.  Do I really need to point out the examples of this?

Early in this thread Siwastaja, who knows his shit when it comes to lithium battery chemistry, said as much:

Now, you had one thing right - designing a BMS is not easy. It feels like an easy task, but the devil is in the implementation details. This can be easily seen by analyzing existing BMS's. Every single one I have seen has been faulty by design, and possibly dangerous.

Of course it can be done safely - but at high cost and complexity.  That price is worth it in commercial EVs where energy density is so critical.  But not so much home RE system or some DIY EV applications.

« Last Edit: April 23, 2018, 04:01:38 am by mtdoc »
 

Online MosherIV

  • Frequent Contributor
  • **
  • Posts: 945
  • Country: gb
Re: Battery balancing - big cells
« Reply #24 on: April 23, 2018, 04:14:04 am »
Quote
Bottom balancing has become the method of choice. You can read more about the rationale and concept HERE and HERE.

I was intrigued to see what 'bottom balancing' is.

Turns out that the guy is destroying cells because his system is NOT balancing, ie trying to keep all the cells in a series stack all at the same voltage (ie all at the same SOC) and therefore some cells would get reverse biased and be destroyed.

Firstly, all cells in a large series stack should be monitored and the pack shut down if any cell reaches a minimum cell voltage. Yes, you do not get the max capacity but it stops cells from being destroyed.

His system is not using closely matched cells. It really helps in series stacks if all the cell charge and discharge at the same rate, even if it is only at the start of life. Again, this is where cell balancing helps.

His system is using the full capacity range of the cells. In EV systems, this is unusual because EV application normally require a minimum sevice life. The main strategy to acheive long life is to limit the max and min SOC to 20% for min and 80% for max. By doing this, again it avoids bringing cells down to the point where they get reverse biased.

Fyi, for those that have not read the article - top balancing is bring all cells in a series stack up to 100% SOC or a known Voltage which is known to be 100%.
Bottom balancing is bring all the cells in a series stack to the voltage that is 0% SOC
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf