Home grown BMS ideas !

How do you store and manage your electricity?
User avatar
Johny
Senior Member
Posts: 3732
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny »

BTW. Nice visualisation of Jack's premise weber.
antiscab
Senior Member
Posts: 2680
Joined: Mon, 26 Nov 2007, 05:39
Real Name: Matthew Lacey
Location: Perth, WA

Home grown BMS ideas !

Post by antiscab »

alternatively, you could wire those BMS modules to a contactor that switches out the charger of each sub pack (could even do this on the AC side).

Matt
Matt
2017 Renault zoe - 25'000km
2007 vectrix - 156'000km
1998 prius - needs Batt
1999 Prius - needs batt
2000 prius - has 200 x headway 38120 cells
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle »

yeah it'd be pretty easy to keep the functionality for charging but bypass it for discharge.
You could also make it switch to a low current charge (below shunt current) to avoid the cycling on and off while balancing.
The Engine Whisperer - fixer of things
User avatar
weber
Site Admin
Posts: 2784
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber »

Nevilleh wrote:weber's extension of the analogy is good, but more complex. I used to be a lecturer in a past life and I found simple, if not complete, was always a good starting point.
Hey! I was a lecturer in a past life too. In renewable energy technology. I totally agree that you started in the right place with your analogy.
BTW, did you derive the pseuodonym "weber" from the BBQ maker, the carburettor maker, or just lines of force?
Image
(Hot Stuff, get rid of petrol, or just force your way through!!)
(meant in the nicest possible way, the pun(s) is(are) quite elegant!)
I don't mind which. But it started as the latter, although I think "lines of force" is a misnomer and would prefer to say "magnetic flux".

Coulomb started it. We are both in awe of the Tesla Roadster and the Tesla Motors company. And, well, we are converting a roadster ourselves, a Mazda MX-5 (known in Japan as the Eunos Roadster).

As well as being the name of the inventor of the induction motor, the tesla is the SI unit of magnetic flux density (equivalent to a weber per square metre). Coulomb named himself after the SI unit nearest to being the electrostatic analog of the tesla. And I thought, "That's a good idea" and named myself after the actual electromagnetic analog of the coulomb. And I figure tesla motors probably has expertise equivalent to at least a weber per square metre too. [;-)]

So we have the nice fact that webers and coulombs work together inside electric motors to make them go, and "Coulomb & Weber Roadster" has a nice ring to it.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Home grown BMS ideas !

Post by Squiggles »

weber wrote:
So we have the nice fact that webers and coulombs work together inside electric motors to make them go, and "Coulomb & Weber Roadster" has a nice ring to it.


That's good, people will be able to hear it coming.
User avatar
Johny
Senior Member
Posts: 3732
Joined: Mon, 23 Jun 2008, 16:26
Real Name: John Wright
Location: Melbourne
Contact:

Home grown BMS ideas !

Post by Johny »

Thanks antiscab, Electrocycle - I was trying to avoid extra work but you are right. Hacking the chargers sounds like fun - I guess Image.
User avatar
weber
Site Admin
Posts: 2784
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber »

Squiggles wrote: Now just imagine I am dumb....go on you can do it...
...
Do you really need to balance? What does it really gain? The chain is only ever a strong as its weakest link!

Ah the pitfalls of reasoning from an analogy. Like most chains, this kind of chain can certainly never be stronger than its weakest link, but unlike other chains, it can in fact be a heck of a lot weaker than even its weakest link, purely because of misalignment of cell's states of charge.

I guess my other diagrams of balanced and unbalanced packs didn't do it for you. So here is an exagerated case of unbalance.

Image

Note in the left diagram the cell second from the left is fully discharged. That's why this is the minium SoC for the whole pack. In the right diagram the cell on the right is fully charged. That's why this is the maximum SoC for the whole pack.

The smallest capacity cell is the one in the middle, but it can never be fully charged or discharged. Less than half its capacity can be used.

The worst case would be if you put a fully charged cell into a pack in which there was a completely discharged cell, or vice versa. The usable capacity of the pack would then be zero, because you can't put any charge in without damaging the fully charged cell, and you can't take any out without damaging the completely discharged cell.

[Edit: "unbalanced cells" -> "unbalanced packs", "second from the right" -> "second from the left"]
Last edited by weber on Tue, 24 Nov 2009, 02:24, edited 1 time in total.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
jrickard
Noobie
Posts: 27
Joined: Fri, 20 Nov 2009, 10:56
Real Name: Marion Rickard
Contact:

Home grown BMS ideas !

Post by jrickard »

weber wrote:
Ah the pitfalls of reasoning from an analogy. Like most chains, this kind of chain can certainly never be stronger than its weakest link, but unlike other chains, it can in fact be a heck of a lot weaker than even its weakest link, purely because of misalignment of cell's states of charge.

I guess my other diagrams of balanced and unbalanced cells didn't do it for you. So here is an exagerated case of unbalance.

Image


There are indeed pitfalls in reasoning from analogy - the very heart and spirit of "typing yourself smart."

The question is, how did they get out of balance?

I just went through this discussion on DIYElectric forum. I have never balanced the Speedster, top or bottom, and not only don't have any problems with it, but after an actual 100% discharge drive, found all 72 cells (two banks of 36) to fall between 2.8 and 2.9 v. They've NEVER been balanced by any action I know of.

The concept discussed in the other forum was of cell balance "drift". As it hasn't appeared in a year and 5000 miles, I would suggest it isn't much of a problem.

I understand the theory. If we charge and discharge the least capacity cell to the maximum, we must be charging the highest capacity cell to some lesser value on both ends. As it appears cell life is increased by both lower charge finish voltages, and less DOD, it would appear the strong get stronger and the weak get weaker.

I just can't detect such with test instruments. And I would suggest that there are a variety of other factors at play that baasically get us into cancelling errors and noise.

For example, we also know that hot cells diminish in capacity more rapidly than cool ones, and hot cell areas in the car can be a real problem. So where the cell is positioned may be more of an issue than what it's original capacity was.

And we don't know if the diminishment by percentage on a higher capacity cell is either greater or lower than the absolute quantity measured from cell to cell.

So it kind of reduces from theory and analogy to practice. I do not detect "cell creep" or "balance creep" at all. And I would suggest that if it is a problem, it's more likely to be an annual problem than a daily problem. And I would advocate bottom balancing as the cure if it is necessary at all.

Jack Rickard
http://evtv.me
jrickard
Noobie
Posts: 27
Joined: Fri, 20 Nov 2009, 10:56
Real Name: Marion Rickard
Contact:

Home grown BMS ideas !

Post by jrickard »


In fact, I wonder if this is an argument for buddy pairs in general. That way, you get the option of "manually balancing" the pack, pairing the worst with the best, next worst with next best, and probably the others don't matter much. If you notice the pack getting out of balance, find the best and worst pairs and swap buddies. Repeat as necessary.
[/QUOTE]

Peter Senkowsky called yesterday to rave a bit about our latest video and the bottom balancing thing. Peter is the designer of the Raptor controller and has a number of years working with the Corbin Sparrow and some Ford Rangers.

This is PRECISELY what he advocates. Hand select your cells by capacity, and do a string of parallel pairs matching up the high capacities with the low capacities.

It's an interesting notion. Adds a bit of work to building an electric car, and most don't have the equipment to test capacity of a cell this size. But it has merit.

Jack Rickard
http://evtv.me
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor »

What a thread !! A few new faces.. welcome..

Can I attempt a summary from my point of view ?

A battery managment system (BMS) that warns (and takes action if required) of overvoltage, undervoltage and does some form of voltage balancing is required for pretty well any battery chemistry. This may be chemical in some part with NiMH chemistry for instance.

Without some shunt charge balancing, the above weber scenario is the outcome in extreme cases an to a lesser extent at all times.

Now remember here that Jack has clearly stated that he has killed packs\cells by fitting unbalanced replacement cells. Proof that balancing is actually required (at least at some stage).

Maybe the above example of 100% DOD means that a bottom balance has been performed (rather brutally !)

Jack, I do feel that your concern about top vs bottom balancing is interesting but QUITE irrelevant if there is voltage monitoring BMS in place.
Electrocycle is pointing this out several times. Why create fear about one cell killing another ?... just use a BMS ??? problem gone. No more scare tactics ?

My concern is that to make the economic theory of value for money work with my TS cells (vs lead acid for instance), I must achieve 2000 maybe 3000 cycles. That will push out to 10 to 15 years (!!!) and not just one year as in the examples you suggest.
It is easy to have a battery pack work OK in it's infancy, but I am looking for the long haul.

To suggest that I will kill the pack by using top balancing is quite unfair information to the rest of the EV world and technically not true if there is voltage monitoring BMS in place.

Minimising DOD should be the aim for all chemical battery users IMHO. Is there data to convince people otherwise ? If not then trust in a smart BMS to manage the pack.

All you want to know is that one cell is weak and that will most likely be the first one to fail (or reach too low a capacity for expected range). You do not want to use the other 100 or 200 cells at less than best SOC because of one cell that will fail anyway.
Rise to the highest level and don't be dragged to the lowest common denominator !!!

However, if I can see data that TS cells for instance will last longer if they are never taken to full charge I would be interested. It already appears that 4.2V or 4.35V or whatever is no go for ling life, but is 100% SOC also an issue ?
My experience with many chemistries is that there is a 'float voltage' that if the cell is held there for some hours the SOC will be mighty close to 100% SOC without pushing out to the OV point. This seems as low as 3.3V on TS and 13.6V on LA.
This is where shunt/top balance/reduced charge rate come into the charge plan, without need to go near OV or max V.

Hey, not flaming anyone, just questioning the logic.

Image
iMiEV MY12     110,230km in pure Electric and loving it !
jrickard
Noobie
Posts: 27
Joined: Fri, 20 Nov 2009, 10:56
Real Name: Marion Rickard
Contact:

Home grown BMS ideas !

Post by jrickard »

acmotor wrote: What a thread !! A few new faces.. welcome..

Can I attempt a summary from my point of view ?

A battery managment system (BMS) that warns (and takes action if required) of overvoltage, undervoltage and does some form of voltage balancing is required for pretty well any battery chemistry. This may be chemical in some part with NiMH chemistry for instance.

Without some shunt charge balancing, the above weber scenario is the outcome in extreme cases an to a lesser extent at all times.

Now remember here that Jack has clearly stated that he has killed packs\cells by fitting unbalanced replacement cells. Proof that balancing is actually required (at least at some stage).

Maybe the above example of 100% DOD means that a bottom balance has been performed (rather brutally !)

Jack, I do feel that your concern about top vs bottom balancing is interesting but QUITE irrelevant if there is voltage monitoring BMS in place.
Electrocycle is pointing this out several times. Why create fear about one cell killing another ?... just use a BMS ??? problem gone. No more scare tactics ?

My concern is that to make the economic theory of value for money work with my TS cells (vs lead acid for instance), I must achieve 2000 maybe 3000 cycles. That will push out to 10 to 15 years (!!!) and not just one year as in the examples you suggest.
It is easy to have a battery pack work OK in it's infancy, but I am looking for the long haul.

To suggest that I will kill the pack by using top balancing is quite unfair information to the rest of the EV world and technically not true if there is voltage monitoring BMS in place.

Minimising DOD should be the aim for all chemical battery users IMHO. Is there data to convince people otherwise ? If not then trust in a smart BMS to manage the pack.

All you want to know is that one cell is weak and that will most likely be the first one to fail (or reach too low a capacity for expected range). You do not want to use the other 100 or 200 cells at less than best SOC because of one cell that will fail anyway.
Rise to the highest level and don't be dragged to the lowest common denominator !!!

However, if I can see data that TS cells for instance will last longer if they are never taken to full charge I would be interested. It already appears that 4.2V or 4.35V or whatever is no go for ling life, but is 100% SOC also an issue ?
My experience with many chemistries is that there is a 'float voltage' that if the cell is held there for some hours the SOC will be mighty close to 100% SOC without pushing out to the OV point. This seems as low as 3.3V on TS and 13.6V on LA.
This is where shunt/top balance/reduced charge rate come into the charge plan, without need to go near OV or max V.

Hey, not flaming anyone, just questioning the logic.

Image


I've seen this "we're really all saying the same thing" approach several times. We're really NOT saying the same thing.

Ok, you're going to be fine top balancing because your BMS is going to "alert" you to a cell going over the knee of the curve, so we're really all winding up at the same place.

No.

The problem is the time zone. We're accustomed to observing our cells charging at the top at 20 amps. Unfortunately, when we're driving, they too often are discharging at 200 amps, or 500 amps. Now I've explained this, and graphed this. If you don't get it, they're your cells - take my advice and do whatever you like with the cells. But it is NOT the same thing.

As long as you drive to 70% you'll probably never have a problem. But then you don't need a BMS of any kind. If you wait until you get an alert that you have a low cell, while you're driving, you've most likely lost one. And maybe three.

As to the premise that you can extend life by not fully charging, that's already not only a known, but a manufacturers recommendation. They rather clearly state (in Chinglish) that you can further extend cycle life by charging to 4.10 vdc instead of 4.25 vdc. That would rather imply that shallower charges lead to longer cycle functions.

Finally, if 10 or 15 years is your time frame, and a year's experience isn't enough for you to deal with, then there's really no necessity for a daily fix is there?

You will extend life by either not balancing or bottom balancing. And you will decrease it by top current shunt balancing. Period.

Summarizing what I've said, to represent that it really means something else, isn't going to change anything. If you want to shunt balance every third cell, and sacrifice goats to every second cell by the light of the moon, I say go girl friend. Just report your results.

I cannot even count how many e-mails I've received from people now admitting privately that they have been losing cells all along with this goofy top balance current shunt scheme.

It can be embarrassing to report failures, but it is important. We're individually invested in a device from China that costs $10,000 and comes with documentation that says "your glad acceptance is our warmest happiness." It's a little unclear to me that Thundersky has actually used their own cells in an electric car. And so we're as a group climbing a wall of learning curve.

But its all good. These cells make electric cars viable in a way that they just were not with AGMs. That we have to work out how to deal with them is a good kind of problem.

Jack Rickard
http://evtv.me
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor »

jrickard wrote: ...

You will extend life by either not balancing or bottom balancing. And you will decrease it by top current shunt balancing. Period.
...


Sorry Jack, that claim does not have any technical basis in my thinking.
I am open to be convinced otherwise, but not by those statements !
Keep putting your case though.

Surely we do not question that cells of any chemistry should not be taken over or under voltage ! Your whole concern is the discharge 'knee' and the available capacity to cause damage ? But with BMS this is irrelevant, as it should be.
The valid point you raise is 'should the cells be operated at a lower SOC'. This I would need to see evidence for in life expectancy.
This will be hard since the lithium I use now will be rather old hat in 5 years time when early data has started to surface !

Cells will always fail in a pack. Eventually they all fail. There is no magic to solve that.
Understanding the failure mechanism is more important for us at this stage.
What I seek is the 3000 cycles. Nearly any cells will do a few hundred without special care from the user !!!

I don't follow your 'time zone' concern ? A cell will drop off the knee just as fast under any charging/BMS regiem. It is the BMS's job to protect you from damage.

True, people are disappointed at a cell failure. Maybe they should look at the factory not the charging technique ?
People may seek a scape goat for a cell failure. I just hope they think about where they cast blame !

Yes, manufacturers do say reduce the max V with lithium for max life. I see that often. Even Tesla recently dropped thier max V in line with that thinking.
I did point out that though that this is different to reducing the max DOD in many (most) chemistries. There may also be only 1% useable SOC between 3.6 and 4V in TS anyway.

Now the 'select matching Ah cells' trick is rather old hat and is just an attempt to make up for a lack of BMS. This works for the early part of a pack's life but as cells age differetially or are treated to different temperatures around the vehicle for instance, the system no longer works in the years to come.
It is also not practical for most people to 'select' capacities for this inital balance anyway as they are not inclined to order twice as many cells as required for their EV.

Lets be clear here.... a good BMS solves this balance problem.
A BMS is real world catering for differeing Ah and ageing.

Some chemistries use chemical design at the high and low end of operating voltage to perform balancing and so packs can seem to remain in balance. NiNH/NiCad use heat for OV shunt (1.5V). SLA uses elecroltye recycling to use up energy to perform OV shunt (14.5V). Lithium warms slightly but seems to be the most BMS dependent chemistry. All these methods would appear to result in shortened cell life if they are relied upon for balancing in my experience, thus my feeling that all chemistries benifit from a good BMS.

I would also point out that I have gone for a few amps shunt capacity as I need to cater for recharging in regen operation.

I might have it all wrong mind you ! But yes, I'm on the side of the world's leaders in battery systems.

Image
iMiEV MY12     110,230km in pure Electric and loving it !
jrickard
Noobie
Posts: 27
Joined: Fri, 20 Nov 2009, 10:56
Real Name: Marion Rickard
Contact:

Home grown BMS ideas !

Post by jrickard »

acmotor wrote: [

Sorry Jack, that claim does not have any technical basis in my thinking.
I am open to be convinced otherwise, but not by those statements !
Keep putting your case though.

Surely we do not question that cells of any chemistry should not be taken over or under voltage ! Your whole concern is the discharge 'knee' and the available capacity to cause damage ? But with BMS this is irrelevant, as it should be.
The valid point you raise is 'should the cells be operated at a lower SOC'. This I would need to see evidence for in life expectancy.
This will be hard since the lithium I use now will be rather old hat in 5 years time when early data has started to surface !

Cells will always fail in a pack. Eventually they all fail. There is no magic to solve that.
Understanding the failure mechanism is more important for us at this stage.
What I seek is the 3000 cycles. Nearly any cells will do a few hundred without special care from the user !!!

I don't follow your 'time zone' concern ? A cell will drop off the knee just as fast under any charging/BMS regiem. It is the BMS's job to protect you from damage.
Surely you're kidding. What amount of TIME do you think it takes to go from 3.00 vdc to 2.9vdc at 20 amps. What time do you think it takes to to verom 3.00 to 2.9 vdc at a 200 amp discharge. And at a 500 amp discharge.

Now do the same thing from 2.9 to 2.8, and from 2.8 to 2.7

This is an experiment. Something you can duplicate at home. Not unctious words of its all really the same and if you had a good BMS like mine you would not have to worry about it. Go measure it. Time it. And then rejoin the conversation.

Jack Rickard


v
Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh »

TS do use their own cells in EVs. They supply buses and motorcycles that I know of. I have an electric motorcycle supplied by TS - I have re-named it EVT 6000, so it fits in with my EVT brand here.
It has 20 x TS60AH cells and a "BMS" which I have pulled to bits and discovered that it monitors cell voltages and temperatures only. It doesn't do any balancing at all. It does make a tiny, squeaking sound if a cell gets down to 2.5V. I suupose it might do the same if they ge too hot.

The machine does 100 kph flat out and draws about 60A, according to my clamp ammeter. Varies a wee bit, depending I suppose, on how flat the road is and what the wind might be doing, but 60A seems a good average. That's about 3.8 kW input to the motor.

The charger is a big box that supplies 80V at 20A and that's it. It shuts off when the current drops to about 5A. Note the 80V for 20 cells. TS themselves supply a charger that just does 4.0V per cell.

The BMS also shows watt-hours consumed/left but it is pretty inaccurate and needs to be zero'd fairly regularly, else it creeps up.

So, no balancing whatsoever. Good enough for TS, good enough for Jack!

My measurements so far indicate that the cells fresh out of the box match each other for voltage (and hence SOC) pretty well; a variation of less than .1V showed across my 135 SE cells. It seems that TS rely on this to get started and then don't worry any more. Their warranty is only 12 months, so why care?

Even Jack admits the need for balancing when you have to replace one cell in a battery. How did you do that Jack?

Looking at the discharge curve for the LFP60AHA at 1C, or 60A, the energy available to go from 3.0V down to 2.5V is approximately 1% of the total, or .6 AH. So the time to get down there at a 60A discharge, is 36 seconds.

Just extrapolating, I would expect a 160 AH pack being discharged at 500A to take about 4.3 seconds to drop the same voltage. As Jack says, its not a lot of time! Especially if all that happens is that an alarm sounds and the driver has to respond by backing off. But it is enough time, especially if the BMS backed it up by severely limiting the motor controller current if a cell gets to 2.5V, say.

But after all that, the cell capacity represented by the voltage range of 3 down to 2.5 is only 1%, so play it safe and stop everything at 3.0V. Same as running out of gas in an ICE'd vehicle.

You could be a bit more clever and have the BMS allow a "small" motor current which might be a sort of "limp home" mode, enough to get you across the intersection and to the side of the road safely, but that's about it.

There has certainly been a lot of animated and interesting discussion here, but let's not lose sight of the point of it all: we are trying to prolong the life of our expensive cells as much as possible and if a BMS is needed to do this, let's be very sure of what that BMS actually has to do. Defining the problem is 90% of the way to solving it!

Nevilleh
Senior Member
Posts: 773
Joined: Thu, 15 Jan 2009, 18:09
Real Name: Neville Harlick
Location: Tauranga NZ

Home grown BMS ideas !

Post by Nevilleh »

BTW, some people might "type themselves smart", but I like to think about what I want to get across before laying hands on the keyboard. I hope others do too.
"always make sure brain is engaged before putting mouth into gear" as they say.
User avatar
weber
Site Admin
Posts: 2784
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Home grown BMS ideas !

Post by weber »

jrickard wrote: The question is, how did they get out of balance?

I just went through this discussion on DIYElectric forum. I have never balanced the Speedster, top or bottom, and not only don't have any problems with it, but after an actual 100% discharge drive, found all 72 cells (two banks of 36) to fall between 2.8 and 2.9 v. They've NEVER been balanced by any action I know of.

Aw c'mon Jack. Don't you think you're overstating your case a little? They have to have been balanced at least once, at the start, if only by the action of the manufacturer before sending them to you.

And I've noticed you use a term "to marry a new cell into the pack". Isn't that just a euphemism for manual balancing? i.e. ensuring that the new cell has a similar state of charge to the existing smallest-capacity cell, so it will not reach full charge or complete discharge before that cell does, and thereby reduce the usable capacity of the pack.

Squiggles seemed to be asking why any balancing was necessary, not merely why automatic balancing on every charge might or might not be necessary.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle »

if they have to be balanced once, what's the harm in maintaining that balance on each cycle with the BMS?

Do you think new cells are discharged to the "knee" and then charged with a fixed number of amp hours, or are they all charged to 100% SOC?

The end of charge voltage differences are partly to do with charge time.
You can finish a full charge quicker with a higher cutoff voltage, vs charging for longer at a lower voltage - but the state of charge will end up the same.
The Engine Whisperer - fixer of things
User avatar
acmotor
Senior Member
Posts: 3605
Joined: Thu, 26 Apr 2007, 03:30
Real Name: Tuarn
Location: Perth,Australia

Home grown BMS ideas !

Post by acmotor »

Still on the 'time zone' point from Jack.... please lets talk in xC discharge not absolute Amps i.e. 3C rather than 500A
This helps with the understanding the concern.

Now, what is the difference with a cell passing 3C going over the knee and the state of charge of the rest of the pack ?
The pack voltage is dropping faster, so ? That doesn't change a BMS's monitoring of the single cell's voltage and the need to back off the discharge.
Sure, some other cells still have more capacity to push through your weak cell, but if your weak cell can pass 500A then it is not at 0% SOC or anywhere near it !

I would also note (as I think all battery users are aware) that you can pull a Lithium cell below the 2.5V point even at 100% SOC with high discharge.

What I'm saying is that the 'knee' voltage due to SOC is discharge rate dependent (nothing new there) and that it does not happen as fast as Jack is scaring us with. Particularly since, at the high discharge rates talked about, the cell is not at risk of reversing or something in the seconds suggested ! Perhaps if it is, the C rate is too high for the cell design ?
... and the charge method of the pack has nothing to do with it IMHO.

A good BMS is keeping track of the pack kWh/Ah and you should not be in the habit of going to 100% DOD. Nothing should come as a surprise at the discharge end. If the BMS really has it wrong, then a cell has failed and the charging/balancing method is not the scape goat. 'Saving' the cell does what ? it has failed already, or as the weakest link should be replaced so the other 100 or 200 cells can perform correctly.

It would be desirable to remain within the manufacturer's designed discharge rates as well. So with LFP TS this would be 3C cont.
I note that Nevilleh's 60Ah TS are discharged at 60A (1C) max. This has a lot to do with life expectancy and battery management requirements IMHO.
What is the charge current in cell C ? is this CV ?

EC notes "You can finish a full charge quicker with a higher cutoff voltage, vs charging for longer at a lower voltage - but the state of charge will end up the same." That's my impression as well. I think that is TS's thinking with the CV charge at 3.65V ?

BTW, I am of the opinion that most SLA battery premature failures are due to their lack of individual cell BMS. The batteries are cheap and work for the first 12 months without any BMS.
I really can't afford the LiFePO4 to last only 12 months, so I've gone for BMS. The suggestion is that the BMS can cause harm ????
So far, I can technically convince myself that full BMS is the correct path. Still waiting for any other thinking to be justified.    Image
iMiEV MY12     110,230km in pure Electric and loving it !
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle »

yep, not throwing out my BMS just yet :)

With the 600mA shunt current it's not going to be doing a whole lot of balancing if I'm charging at 20A+ anyway! (but I do plan to drop the charge to a balance friendly current whenever I have the time)
I will be monitoring SOC with a Cycle Analyst, and watching the 2.5v warning.
The Engine Whisperer - fixer of things
Mr. Mik
Groupie
Posts: 66
Joined: Fri, 08 May 2009, 04:41
Location: Gold Coast, Aus

Home grown BMS ideas !

Post by Mr. Mik »

Is there any problem at all with the odd cell having an increased self discharge rate (with LiFePO4 cells)?

With NiMH cells the weaker cells often self discharge faster than the others; therefore the top-balancing is needed to stop a differential SOC developing.
There is always a way if there is no other way!
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle »

I think self discharge would certainly come into play to some degree - and will definitely cause imbalance over time.
The Engine Whisperer - fixer of things
Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Home grown BMS ideas !

Post by Squiggles »

weber wrote:
Squiggles seemed to be asking why any balancing was necessary, not merely why automatic balancing on every charge might or might not be necessary.


Hmm...I might of asked that but it is not what I meant.
Obviously when new or when a cell is being replaced some form of balancing is required....although with a new cell that is near 100% the balancing might happen automatically when connected to the battery.
What I meant to ask was if regular balancing is really needed in normal operation. How do the cells get out of balance (if they do)? How long does this action take to show any significant effect?
Hypothetically if it takes 12 months for the effect to be of consequence why bother with the expense of balancing every charge. Just do it with your annual maintenance. Save all those wasted milliwatts lost as heat in your bypass resistance, make your BMS 50% cheaper, make your BMS significantly more reliable.

Somebody needs to set up a business providing annual battery maintenance. Test your cells and do a balance charge, refill the windscreen washer bottle, and apply some tyre black.
User avatar
Electrocycle
Senior Member
Posts: 985
Joined: Sun, 19 Oct 2008, 20:23
Real Name: Andrew
Location: Sydney
MSN: dumhed@dumhed.com
Contact:

Home grown BMS ideas !

Post by Electrocycle »

why do annual maintenance if the BMS can keep the batteries in balance every cycle?
The Engine Whisperer - fixer of things
Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Home grown BMS ideas !

Post by Squiggles »

why spend $1000 on a BMS if a $300 one will do the job?
Squiggles
Senior Member
Posts: 742
Joined: Wed, 22 Apr 2009, 03:19
Real Name: Neil
Location: Newcastle NSW

Home grown BMS ideas !

Post by Squiggles »

Why make a BMS less reliable by adding unnecessary complexity?
Post Reply