Low cost BMS

How do you store and manage your electricity?
User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

The cheap inductive balancers I saw, always worked over the whole voltage range. There was no way to tell them to only work above a certain voltage. If there was a sufficient voltage difference, they shuffled charge. But yes, if you were designing them into your own CMUs you could make them work for LFP. However I don't see the need. The energy wasted in resistive balancing is minuscule.

I haven't seen a need for anything like 5 amps of balancing. We successfully use 0.6 A for up to 400 Ah cells. Our maximum imbalance in standby current between CMUs is 1 mA. Even if they don't get balanced for 5 days, that's still only 5*24*0.001 = 0.12 Ah of imbalance, which will only require 0.12/0.6*60 = 12 minutes to balance out.

In the car, we control the charger to back off the current, as you say. But in the off-grid solar, we use a trick that extends the balance time without having to control the charge current. We don't use a constant cut-in voltage for the bypass resistors. We vary it with current, in such a way as to compensate for internal resistance, so they bypass at approximately 98% charge irrespective of current.

And recently, in discussions with Pascal (Warrick Beattie) I came up with a way to extend the time even further, which we haven't implemented yet: It does not require internal-resistance compensation.

We know that if all the bypass resistors are turned on, no balancing is happening. So when it happens, the master unit (BMU) can recognise the situation and tell all the CMUs to raise their bypass thresholds, until at least one of them is not bypassing again.

Alternatively, the master can tell all the CMUs what the average cell voltage is, and they can all bypass at that voltage, provided it is above the rested 98% SoC value.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

User avatar
4Springs
Site Admin
Posts: 610
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

Re: Low cost BMS

Post by 4Springs »

T1 Terry wrote:
Mon, 27 Jan 2020, 14:52
The resistor type balancers work when the cell reaches X voltage and burns what ever value the resistors used can handle. That method relies on the charge current being reduced to less than the resistor burn off rate plus the load at the time ...... that would require accurate cell voltage monitoring to control the charge rate so the highest cell doesn't exceed the maximum safe voltage.
With that sort of charge current control, either type balancer would work fine and the inductive charge shuffle could actually move more current between cells, with no heat transfer into the cell, than the resistor type balance boards could handle.
I do have a number of the EV Works 5 amp resistor boards that I trialled to see if I could make an easy fit balancer for the bigger house battery packs, but the heat generated was incredible. The only way I could see these boards could be used without damaging the cells was using a heat exchanger and that further complicated the build, so I gave it a miss.
I have a fix for the problem in this system and maybe it will turn out to be the answer to the problem, only time will tell I guess.

T1 Terry
So far so good with my house battery. It is a grid connected battery, which may be different to what you are describing. In a normal day it is charged up by lunch time, so spends many hours each day fully charged. The inverter goes into a low current charge mode when the first cell reaches shunt voltage, and stays that way until all cells are shunting, at which point it shuts off. This low current mode seems to last only a few minutes. I guess the cells are very similar, and they are all in the same environment.
Bypass current is about 500 mA. The boards get hot to the touch (perhaps 50°C), but they are well proud of the cells, and certainly don't heat the cells at all. After your previous comments I made a point of testing the temperatures with a non-contact thermometer. Even at the terminal (the part of the cell closest to the resistors) there was no discernable temperature increase. The balancing doesn't last long, so not much power wasted.
Will be interesting to see how it goes in the winter, when our input is less and output is more. I've set the inverter to do a weekly 'balancing' charge, which it will do from the mains if there is not enough solar.

User avatar
4Springs
Site Admin
Posts: 610
Joined: Thu, 23 Dec 2010, 01:14
Real Name: Christopher Walkden
Location: Selbourne, TAS

Re: Low cost BMS

Post by 4Springs »

weber wrote:
Mon, 27 Jan 2020, 17:28
Alternatively, the master can tell the CMUs what the average cell voltage is, and they can all bypass at that voltage, provided that voltage is above the rested 98% SoC value.
Ah, now that is interesting. And something that I could implement...
In my case the master tells each CMU when and how much to bypass. And it has access to all cell voltages. Currently it just tells each cell above a threshold to bypass, by a percentage that increases to 100% at another threshold.

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

4Springs wrote:
Mon, 27 Jan 2020, 18:33
weber wrote:
Mon, 27 Jan 2020, 17:28
Alternatively, the master can tell the CMUs what the average cell voltage is, and they can all bypass at that voltage, provided that voltage is above the rested 98% SoC value.
Ah, now that is interesting. And something that I could implement...
In my case the master tells each CMU when and how much to bypass. And it has access to all cell voltages. Currently it just tells each cell above a threshold to bypass, by a percentage that increases to 100% at another threshold.
I'm glad you like it. :)

But I've never seen the point of variable bypass. I figure if it's over the lower threshold and it's not bypassing the maximum current, then it's just wasting time. They automatically end up PWMing with a period of a second or two, with whatever duty cycle they need.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

The induction balancers we use switch off once the cells are within 10mV and don't switch back on until the differential is greater than 100mV. That way they are not a constant load on the battery. We tried small current balancers that did the same thing but the balancing was so slow the cells required being held at a higher than optimal voltage for an extended period. Holding the cells at a high voltage causes the electrolyte to heat, then adding more heat from the "bypass diode" into the cell can only have a negative affect on cell cycle life.
To determine just how much heat is generated by the balance boards, disconnect the negative end from the cell terminal and reconnect using a wire that is capable of handling the current expected. Now you will be able to see just how much heat is generated and then you will know how much of the heat is being transferred into the cell via the negative terminal.
Some what deceiving to refer to it as "bypass current" it is a controlled short across the cell between a higher voltage and a lower voltage and that current is causing a conversion from electrical energy to heat energy. That heat has to go some where .....

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

T1 Terry wrote:
Tue, 28 Jan 2020, 10:38
The induction balancers we use switch off once the cells are within 10mV and don't switch back on until the differential is greater than 100mV.
Ah yes. That was the other reason I told my customer/friend that these balancers are useless for LFP. One cell could be at 80% SoC and 3.33 V and its neighbour could be at 100% SoC and 3.43 V and it still would not balance!
We tried small current balancers that did the same thing but the balancing was so slow the cells required being held at a higher than optimal voltage for an extended period. Holding the cells at a high voltage causes the electrolyte to heat, ...
We bypass at 3.41 V (plus IR compensation). What voltage do you bypass at? You need to figure out why your cells are getting so far out of balance so quickly, that they need long periods of balancing. I only know of two reasons for imbalance: differences between cells in self-discharge current and differences between cells in current taken by the BMS (or other individual cell loads).

I see self-discharge for LFP quoted at < 1.5% per month. For a 400 Ah cell, that's 1.5/100 * 400 Ah/(24 h*30) ≈ 8.3 mA. But it's not that figure that causes imbalance, it's the difference in that figure between the worst and best cell. But let's imagine a worst case where the best cell has 0 mA of self-discharge and the worst has 8.3 mA. After 24 hours without balancing, the difference in charge between those cells will be 8.3 mA * 24 h = 200 mAh = 0.2 Ah.

If the bypass current is 600 mA it will take 200/600 = 1/3 h = 20 min to balance that out.

If the bypass current is 5 A it will take 200/5000 = 1/25 h = 2.4 min to balance that out.

The total amount of heat generated in the bypass resistance will be exactly the same in both cases. If we bypass at 3.5 V this will be 0.2 Ah * 3.5 V = 0.7 Wh of heat. Utterly negligible.
... then adding more heat from the "bypass diode" into the cell can only have a negative affect on cell cycle life.
To determine just how much heat is generated by the balance boards, disconnect the negative end from the cell terminal and reconnect using a wire that is capable of handling the current expected. Now you will be able to see just how much heat is generated and then you will know how much of the heat is being transferred into the cell via the negative terminal.
That's not true. You still won't know how much goes into the cell and how much goes into the air and convects up and away from the cell. But even if it did all go into the cell it should be negligible, unless your cell-to-cell variation in self-discharge plus BMS discharge is way larger than it should be. We had this discussion 5 years ago.
Some what deceiving to refer to it as "bypass current" it is a controlled short across the cell between a higher voltage and a lower voltage and that current is causing a conversion from electrical energy to heat energy.
Of course it produces heat. But during charge, it still has the effect of letting some current bypass full cells so it can charge cells that aren't full.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

macks
Noobie
Posts: 11
Joined: Sat, 26 Oct 2019, 19:55

Re: Low cost BMS

Post by macks »

Thankyou guys for your feedback on your respective systems. My perspective has shifted. I have time to mull over design and will be for waiting actual batteries and how they will be housed before final design. (and yet another open source project - github'ed :) )

Cheers
Andrew

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

There will always be different camps when it comes to cell balancing and BMS systems in general. I guess time will tell who has got it right and who hasn't by the cycle life of the battery packs. We are in the 9th yr now with a number of off grid battery packs and at the 8yr point they still tested at over 100% of the manufacturers stated capacity using their 0.5CA capacity testing regime. Basically, the battery discharged till a cell reaches the 2.5vdc under load cut off point, if 2 hrs comes up before the battery isolator kicks in, the battery capacity is still more than the manufacturer stated and that is all we care about. If you bought 600Ah and you can draw 600Ah out in 2 hrs without the isolator kicking in, you are still ahead of the game.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

No Terry. Even if my batteries die tomorrow, it won't prove anything about our respective BMSs because there are way too many other variables.

I never said there was anything harmful about having 5 amp balancing currents, or the idea of inductive charge shuffling. I just wanted to make sure that others, who may be designing their own BMS, understand that these things should not be necessary. Particularly since this topic is about low cost BMS.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

weber wrote:
Wed, 29 Jan 2020, 12:06
No Terry. Even if my batteries die tomorrow, it won't prove anything about our respective BMSs because there are way too many other variables.

I never said there was anything harmful about having 5 amp balancing currents, or the idea of inductive charge shuffling. I just wanted to make sure that others, who may be designing their own BMS, understand that these things should not be necessary. Particularly since this topic is about low cost BMS.
Ummm..... the whole purpose of a BMS is to extend the cycle life and retain usable capacity isn't it? The issue is cell internal temperature, the BMS should be designed to avoid the cells operating outside the safe voltage range that causes the heat generation in the first place. If the ambient temperature is going to exceed the point where the cells can dissipate the generated heat, then the BMS needs to incorporate some method of heat transfer to keep the cell internal temperature within the electrolytes safe limits.
The bit I'm trying to point out is, last thing a cell suffering from heat stress needs is more heat stress. A cell already at an elevated voltage is generating heat internally, trying to drop that voltage by converting the electrical energy into heat energy and using the cell to absorb the heat is not going to end well. If the chosen method of balancing is to burn off the excess capacity from the high voltage cell, then do it away from the cell where the heat generated can dissipate without heating the cell in question.

Simply bring all the cells voltage to the same point isn't balancing the cells to the same SOC unless that voltage represents the voltage of that chemistry cell when it is at the saturation charged point. If you charge an LFP or LYP cell to 3.5v and stop the charging and the voltage drops below 3.4v, the cell is not saturation charged yet. Simply raising that voltage to 3.8v is only going to increase the electrolyte temperature, it will not get the cell saturation charged unless that last bit between 3.4v and 3.8v was at a trickle charge rate compared to the cell group capacity. If that is the case, then there is no need to push the cell to 3.8v and cause the electrolyte to heat up unnecessarily, the 3.5v point will do the same job but not generate heat in the electrolyte.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

T1 Terry wrote:
Wed, 29 Jan 2020, 15:04
Ummm..... the whole purpose of a BMS is to extend the cycle life and retain usable capacity isn't it?
No. That's only part of its purpose. Another part is to stop your house burning down. Another part is to extend the calendar life and retain low internal resistance. But a BMS can't control everything that might shorten the life of a battery. It usually can't control the climate. And it can't control the nature of the loads. It can't control what make and model of cells you bought. Or whether one cell got badly assembled at the factory. Those are some of the other variables I referred to.
The issue is cell internal temperature,
Yes. We agree it is important to keep internal temperature low.
the BMS should be designed to avoid the cells operating outside the safe voltage range that causes the heat generation in the first place.
Yes, it should be designed to avoid the cells operating outside the safe voltage range. But no, this is not to avoid heat generation. Voltage alone cannot cause heat generation. For that, there must be current flow. But even then, it doesn't necessarily cause heat generation. It can instead cause a chemical reaction to take place, that stores the energy in chemical bonds. When charging at a low rate, an LFP cell will store the energy chemically in a reversible manner, and it will actually consume some heat, cooling the cell internally (endothermic). This effect is small and can usually be ignored, which is just as well, because there's no free lunch and the heat must be given back on discharge.

Most of any internal heating that does occur, occurs in normal charge or discharge, and the rate corresponds to the current squared, times the internal resistance. P = I²R. Note that the cell voltage doesn't appear in that equation. The internal resistance of these cells is very low, and decreases further as their temperature rises. So self-heating is somewhat self-limiting.

When the cell voltage exceeds about 3.6 V, damage occurs directly. It doesn't have to heat the cell. The energy gets stored by a number of chemical reactions that cannot be reversed, and which cause permanent loss of capacity and increase in internal resistance. Two of these reactions are lithium metal plating and electrolyte breakdown (which forms non-condensing non-dissolving gasses and thickens the SEI layer). Similarly for the damage caused when the voltage subceeds about 2.5 V, but with different irreversible reactions.
If the ambient temperature is going to exceed the point where the cells can dissipate the generated heat, then the BMS needs to incorporate some method of heat transfer to keep the cell internal temperature within the electrolytes safe limits.
That's done in Tesla batteries, but a low cost BMS, when the cells are too hot, can't do much more than throttle back the charge current or disconnect the loads.
The bit I'm trying to point out is, last thing a cell suffering from heat stress needs is more heat stress.
Agreed. But there's no reason why a cell in a home solar power system should be suffering from heat stress at the end of charge, when the charge current will be tapering off.
A cell already at an elevated voltage is generating heat internally,
Not much, if the current is low, as it should be when balancing is taking place.
trying to drop that voltage by converting the electrical energy into heat energy and using the cell to absorb the heat is not going to end well.
Why do you keep saying we're using the cell to absorb the heat? We're not doing that at all. As I keep saying, most of the heat from a bypassing CMU goes directly into the air from the resistors. And almost all the rest is conducted to the high current copper links or cables and then into the air. This warm air then rises upward away from the cells, and cooler air comes in to take its place.

I looked up the heat capacity of LFP cells. Even if all the heat from my worst-case daily balance scenario went into the cell, and the cell was thermally insulated so it couldn't get out again, it would only raise the temperature of the cell by 0.25 °C — a quarter of a degree.
If the chosen method of balancing is to burn off the excess capacity from the high voltage cell, then do it away from the cell where the heat generated can dissipate without heating the cell in question.
Sure. Feel free to do so. But recognise that that has problems too. Mostly fire safety problems due to thin unfused wires. And recognise that cell-top CMUs add negligible heat to the cells.

Internal self-heating isn't much of a problem either. If an LFP cell at 25 °C was thermally insulated so that all the self heat from a 0.5C charge over 2 hours was kept inside the cell. It's temperature would only rise by about 5 °C, to 30 °C. And of course we don't thermally insulate our cells, so they don't rise anywhere near that much. The real killer, at least in Brisbane, is heat from the environment.
Simply bring all the cells voltage to the same point isn't balancing the cells to the same SOC unless that voltage represents the voltage of that chemistry cell when it is at the saturation charged point.
There isn't really a specific saturation charged point. There's a degree of arbitrariness in what rested voltage to call 100% (the two papers below disagree on that voltage), but yes, to balance cells to the same SoC their rested voltage has to be in a part of the voltage vs SoC curve where voltage rises noticeably with SoC. The higher you go, the steeper the curve gets and so the less accurate you have to be about measuring the voltage. You have to be sure that you're no longer on the upper plateau. You can be pretty sure of this by about 3.36 V rested. We use 3.41 V to be really sure.
If you charge an LFP or LYP cell to 3.5v and stop the charging and the voltage drops below 3.4v, the cell is not saturation charged yet.
I agree that it's not charged enough for balancing to be valid. But I remind you that we have internal-resistance compensation of our bypass voltage. The 3.41 V value is only when the current is near zero. When the charge current is higher, the voltage at which the bypass resistors are switched in is higher. Maybe going up to 3.45 V at 0.5C.
Simply raising that voltage to 3.8v is only going to increase the electrolyte temperature, it will not get the cell saturation charged unless that last bit between 3.4v and 3.8v was at a trickle charge rate compared to the cell group capacity. If that is the case, then there is no need to push the cell to 3.8v and cause the electrolyte to heat up unnecessarily, the 3.5v point will do the same job but not generate heat in the electrolyte.
I totally agree with that (except for the stuff about voltage alone generating heat, which I debunked above). But who ever suggested taking an LFP cell to 3.8 V?! They shouldn't be allowed to go over 3.65 V. Our absorb voltage is set to 3.45 V per cell.

The above figures are based on experimental data from here: download/file.php?id=1410
and here: http://www-personal.umich.edu/~hpeng/DSCC2013_Weng.pdf
And you can see how we did our own temperature-rise experiments here: https://www.aeva.asn.au/forums/viewtopic.php?p=12118
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

Not sure how many cells you have killed testing your theory, but the collection I have here back up my claim that the internal temperature generation is voltage related, the other affects are secondary.
As far as loads, as long as the load does not exceed the ion exchange rate, the voltage will remain within the safe range, exceed that current load and the voltage will drop, the electrolyte heats up and starts to deposit material on the plate surface and this causes the initial increase in internal resistance. Reduce the current to maintain the minimum cell voltage and the heat increase stops, simple as that. If the current is not reduced, excess current and heat will then start to effect the thin copper and aluminium plates that act as the conductors and the tops of these are not submerged in electrolyte but simply clamped to the terminal block with a couple of screws, they start to oxidise and that is the beginning of the end. Every heat affected cell I have cut the top off has oxidised plate tabs and have started the process of breaking away, the outer layers first.
The electrolyte is the only medium in the cell to conduct the heat away to the cases and it is very temperature sensitive and will start to degrade if the temperature rises, this process can not be reversed. Maybe, if replacement electrolyte was available, the electrolyte could be changed and this would improve the length of cycle life in a partially damage cell. That is the basis of an experiment I plan to do next winter, that is the only time it is cold enough down this way to decant the electrolyte from dead cells into not completely dead cells to see if the capacity improves.

As far as the heat generation created by the short circuit across the cell and into the negative terminal, you might be correct regarding your particular cell top boards, but not all cell top boards. The majority appear to use a large area of copper where the negative terminal bolts through and this is the area they dissipate the heat. I mentioned in an earlier post, to determine just how much heat is generated, disconnect the negative end and solder a thin wire (so it can't conduct the heat away) to the negative terminal. Now hold the cell top board between finger and thumb at the negative end while the board is in the "bypass" mode, this will give you an idea just how much heat is being pumped into that metal block that makes up the terminal where the negative cable attaches. The bulk of that metal block is inside the cell and has no means of dissipating that heat other than through the thin metal plates that are the conductors the whole cell relies on the move the current in and out.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

Terry, I'd be happy to test your theory, that voltage, without current, can cause internal heating of an LFP cell. Can you describe a simple experiment I can do, involving say a single (possibly sacrificial) cell, a thermometer, some thermal insulation, a power supply and some multimeters? What voltage should I hold it at, to demonstrate this near-zero-current heating?

What are the bad voltage conditions that you think our BMS is allowing to happen?

I totally agree that one should not design a cell-top management unit to use the cell terminals as heatsinks for its bypass resistors.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

"I totally agree that one should not design a cell-top management unit to use the cell terminals as heatsinks for its bypass resistors."
At least we can agree on something :lol:

The first part of the experiment involves temp sensor placement. Useless on either terminal because that is just a block of metal attached to the top of the case and out of the electrolyte. Granted, it will heat up as the conductor plates heat up and transfer that heat into the metal block, but the delay and the fact the heat will be transferred through both conductor plates and through the electrolyte to the case, the readings are not indicative of the electrolyte temperature. Taking the measurement through the case delays the actual reading compared to the time the heating occurred and is more an average than an actual.
Cutting the top off the cell to place the probe centre of the positive plate tabs and negative plate tabs just above the separator sheet and then resealing the case requires the cell and ambient temp to be around 0*C or lower, otherwise the electrolyte evaporates. You will require access to a cold room to do that in Brisbane, we just have to wait for a winter morning in the shed down here :lol:

Next is the ambient temp where you do the testing, if the ambient temp is low enough then the heat can be dissipated out of the cell.
We have some fairly varied ambient temps down this way, sub zero at times in winter and 50*C + in summer :lol: The temperatures in a tin shed where the off grid batteries are often situated can climb well above the ambient temp as do the batteries in a caravan mounted in the front boot. If the electrolyte is heated in these cells it just stays hot, there is no cooling effect no matter how much air movement there is because the air is just as hot. We have had inverters shut down at 70*C because they just can't cool using air movement alone.
This means the heat generated while charging to the higher voltage will remain trapped in the electrolyte, so the minimal current required to maintain that voltage continues to heat the electrolyte .... anyway, you will be able to see this for yourself when you do the testing.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

Terry, You sound like you're making excuses as to why you haven't done this experiment yourself. ;)

I note that your proposed cell opening, to insert a temperature sensor, would not only require a low temperature, but also zero humidity and no oxygen, if you don't want to damage the cell.

But measuring the electrolyte temperature won't be a problem for me. I have some 50 Ah LFP cells with aluminium cases. I can enclose one in a block of polystyrene foam and take a temperature reading off the aluminium, in the middle of a broad face, under the polystyrene foam.

You haven't told me what voltage I should hold it at, to demonstrate heating at minimal current?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

weber wrote:
Thu, 30 Jan 2020, 14:54
Terry, You sound like you're making excuses as to why you haven't done this experiment yourself. ;)

I note that your proposed cell opening, to insert a temperature sensor, would not only require a low temperature, but also zero humidity and no oxygen, if you don't want to damage the cell.

But measuring the electrolyte temperature won't be a problem for me. I have some 50 Ah LFP cells with aluminium cases. I can enclose one in a block of polystyrene foam and take a temperature reading off the aluminium, in the middle of a broad face, under the polystyrene foam.

You haven't told me what voltage I should hold it at, to demonstrate heating at minimal current?
OK, sounds good, just don't do what Professor Jay Whitacre did :lol: Hold the cell at 3.6v initially for 12 hrs and watch for a temperature increase and current decrease, then move up to 3.8v, the most common voltage used by many cell top resistor board balancer builders. The other interesting thing is to watch how much current is required to maintain 3.6v compared to the 3.8v with the insulation on and off.

T1 Terry
Green but want to learn

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

Thinking more about this test, is it possible to charge the cell to the 3.41v you suggested to start the "bypass" balancing? I believe you were suggesting holding this voltage until all cells reach the same 3.41v and bypassing and then stop the charging ..... have I understodd that correctly?
I'm interested in how many Ah flow into the cell after the 3.41v was reached and the 3.6v with minimal current flow, then the additional Ah to push the voltage from 3.6v to 3.8v until a steady minimum current flow was reached.

My power supply isn't steady enough to hold a constant controlled voltage no matter what the current, it creeps up as the current flow reduces and messes up the figures. I'm assuming you have access to a lit better equipment than I have available here. I'm more set up for building battery packs than testing now I've moved to Mannum SA.

I hadn't considered the oxygen and humidity problem because the factories don't appear to have any facility for excluding either during assembly, they just evacuate the cell and fill with the specific volume of electrolyte, then fit the safety valve, spring and cap ..... well in the case of Winston cell construction anyway.

Now I'm even more confused, if there is no oxygen in the cells and no oxygen produced in the chemical process used in the ion exchange in an LFP/LYP cell, then why do cells that have died due to over voltage all show oxidised copper and aluminium conductor plates that have actually failed from the outside layers in once the cell top is removed?

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

@T1 Terry You still haven't told me what to do voltage-wise, to demonstrate internal heating of a cell at minimal current. I note that I am not willing to hold any LFP cell at a voltage of more than 3.65 V, because I know this will accelerate the rate of loss of capacity. Our LyteFyba BMS will not allow this to happen for more than a few seconds. [Edit: I did say "possibly sacrificial" earlier, but I had in mind the 10.5 year old plastic-case cells, not the new aluminium case ones]

Would you expect to see a significant temperature rise over an hour, if I take a rested fully-charged thermally-insulated cell and hold it at 3.65 V for that hour?
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

The test is not to prove if your BMS works or not but rather to verify the heating of the cell and resultant damage at a higher voltage. I'm saying that 3.6v is about the limit before the electrolyte starts to heat up, the electrical energy is converted to heat energy and dissipated through the cell walls. This means to maintain the higher voltage than 3.6v until all cells reach that voltage and are therefore "balanced" a lot of heat will be generated within the cell so any balance board that uses a resistor to burn off the over voltage will be adding heat into an already hot cell causing the gradual destruction of that cell.

It seems every time I mention the problems with charging a cell over 3.6v, the post vanishes, so someone doesn't agree. A test that verified there really is damaging heat generated yet no additional capacity gained by charging a cell above 3.6v would verify my argument or put it to bed as incorrect.

T1 Terry
Green but want to learn

User avatar
weber
Site Admin
Posts: 2702
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Low cost BMS

Post by weber »

@T1 Terry, Why are you wasting your time telling us not to do things that no one in this thread is proposing to do? No one in this thread is proposing to bypass LFP cells at a voltage above 3.6 V. And no one in this thread is proposing to use the cell as a heatsink for the bypass resistors.

I'll be very surprised if you can find any present-day BMS for LFP cells that bypasses at a voltage higher than 3.6 V. Please post a link to its specification. Otherwise, you're just battling straw men.

You can find any number of voltage versus charge curves for LFP cells online. I gave links to two such documents above. I don't know of any that go to 3.8 V because LFP cell manufactures tell us not to go that high, but it's bleeding obvious that the curve is so steep at 3.6 V that the difference in charge between 3.6 V and 3.8 V would be utterly negligible. You can also tell from those graphs, that there's only about 2% of capacity between 3.41 V and 3.6 V. See figure 2 of: http://www-personal.umich.edu/~hpeng/DSCC2013_Weng.pdf
T1 Terry wrote:
Sat, 01 Feb 2020, 11:37
I believe you were suggesting holding this voltage until all cells reach the same 3.41v and bypassing and then stop the charging ..... have I understodd that correctly?
That's certainly one way to do it, except that you also need to wait for the charge current to fall to about 0.01C before you know they are all about 98% full. But the better way that I described was to let the bypass threshold rise in proportion to the charge current times the internal resistance of the cell, up to 3.45 V, or maybe 3.5 V if the weather's cold. That way it's all over sooner.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

"I'll be very surprised if you can find any present-day BMS for LFP cells that bypasses at a voltage higher than 3.6 V. Please post a link to its specification. Otherwise, you're just battling straw men."
https://www.ev-power.com.au/product/bms-cbm400/ see the heatsink on the negative terminal? These units are used in a lot of EV batteries (including the ones in my Prius PHEV battery until I removed them because they were causing the cells to vent) These boards or copies of these boards are also used in a lot of the "drop in" 12v lithium batteries that are flooding the market at the moment. Look through here https://www.ev-power.com.au/ev-powerpak/ and you can see some ready built with a see through lid, most manufacturers use a black lid so it looks like an AGM battery or is even built inside an AGM battery case.

T1 Terry
Green but want to learn

rhills
Site Admin
Posts: 421
Joined: Fri, 25 Jul 2008, 01:57
Real Name: Rob Hills
Location: Waikiki, WA

Re: Low cost BMS

Post by rhills »

Well, you got my attention @T1 Terry, because the LiPO cells I've installed in both yacht and motorhome for house batteries are from EV-Power and use his BMS, though I have to confess I don't know which cell-top units I have in each system. The system in my yacht has been there for just over 8 years now with no issues and no obvious loss of capacity despite being kept in fully charged state for much of that time (it spends far more time in the pen than it deserves :-( ) and I don't think we've ever discharged them below about 70% full.

I note in the description section for the cell-top unit you linked above that it commences balancing bypass at 3.5V, with a bypass current of up to 2A so maybe it rarely gets over 3.6V? I also note that it signals overvoltage at 4.0V, which seems a bit high in the context of the very educational discussion you and @weber have been having about this. I don't actually have a temperature probe in my system and I'm thinking that is probably unwise and in need of rectification despite my lack of problems to date.

Assuming the cell-top units I have are the same as these, they seem to have stood the test of time in my marine application so far anyway.

Cheers,
Rob Hills
AEVA Webmaster
2014 Mitsubishi Outlander Aspire PHEV
Petrol Usage to last refill: Jul 2014 - Dec 2019
Total Petrol: 692.1L
ODO: 57409
Av Consumption: 1.21 L/100km

User avatar
coulomb
Site Admin
Posts: 3919
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Low cost BMS

Post by coulomb »

rhills wrote:
Sun, 02 Feb 2020, 14:21
the LiPO cells I've installed...
Rob, do you really mean LiPo (as in Lithium Polymer), or LFP (Lithium Iron Phosphate, LiFePO₄)? It makes a big difference.

LFP is nominally 3.2, and should not go much beyond 3.6 V.
Nearly all other lithium chemistries are nominally 3.7 or 3.75 V, and should not go beyond 4.2 V.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
160 W solar, 2.5 kWh 24 V battery for lights.
Patching PIP-4048/5048 inverter-chargers.

T1 Terry
Senior Member
Posts: 1072
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Low cost BMS

Post by T1 Terry »

The units I linked to are the same as units I purchased some yrs back from one of the WA suppliers, not sure which at the moment. They most definitely don't start bypassing till the cell reaches 3.8v and stop bypassing at 3.6v. These units shift 5 amps with a cell voltage of 3.8v or higher and the heat generated will burn that grid imprint into the finger and thumb after a few minutes ... yes, but only did it the once :oops: I mounted them on separate bits of 3mm thick aluminium and Jaycar heatsink to find out just how much area was required to dissipate the heat. Glued to a sheet of Lexan Thermoclear so one side could radiate the heat directly to the air and the other side provided good airflow and an insulating surface to mount the units away from the cells, it requires a piece of 3mm thick aluminium 80mm x 80mm to keep the heat down to the point you could hold a finger on it ... still hot to the touch but didn't burn an impression into the skin. the fact they didn't send out a warning signal until over 4v ended my plans to use these units as a backstop for the larger house battery installations.
The small units I removed from my 40Ah cells in the Prius showed signs of the clear encapsulating resin on a number of the boards being seriously heat affected and the cell terminal were quite hot once the small red lights indicated they were bypassing. A check with the multimeter confirmed that some cells were reaching the 4v mark yet the charging continued at 5 amps until the upper end voltage was reached. The cells are damaged and vent every time they are charged now, a real pain in the butt and they are going to be expensive to replace, so I want to make sure the next balancing method is not one that will destroy the cells they are meant to protect.

T1 Terry
Green but want to learn

User avatar
brendon_m
Senior Member
Posts: 465
Joined: Sat, 28 Oct 2017, 11:00
Real Name: Brendon McCarrol

Re: Low cost BMS

Post by brendon_m »

I have these https://www.ev-power.com.au/product/cbm100-180/
in my car and they work seamlessly and don't get noticeably hot when balancing. Although they are only 700mA.

Your ones almost sound like they are set up for halfway between lifepo4 and NCM chemistry. Or they were faulty/built wrong

Post Reply