787 Batteries and Chargers - Part 1
Join Date: Jul 2007
Location: Virginia, USA
Age: 86
Posts: 77
Likes: 0
Received 0 Likes
on
0 Posts
@saptzae (#369)
Thanks for commenting on my post (#364) about charging arrangements and balancing. The TI seminar talk on this was most interesting:
I was struck by how much remains unknown about how much and when to apply the balancing, with the resulting mention that balancing can do more harm than no balancing if done incorrectly.
Obviously, in those parts of the charging cycle where the correct action is unknown, there is the possibility of unsuspected harm from the balancing. Altho presumably small each time, it could accumulate. There is also mention that balancing of charge is more effective and more needed approaching fully charged; but that this part of the cycle is short, which limits the effectiveness. And similarly, balancing of discharge is recommended; but set up to drive a starter motor, this is not possible in the 787.
This 8-page talk has nine figures (9th didn't download for me), and is well worth reading. It is general to several battery types, but follows with some detail specific to lithium batteries. It is undated (unless I missed it). The examples seem to be small batteries (300-500mAh) in my impression, but some of the data can be extrapolated I think, altho not the experience.
One thing that concerned me was the MOSFET regulator chip idea-- if that is general and widely used, the concept would appear to vary the effective bypass by intermittently applying a fixed bybass. (Obviously a 787 battery regulator could not fit on a single chip.) During the no-bypass intervals, the effects to be avoided (e.g. lithium plating out at over-voltage on one cell) would still occur even if for reduced time. Another problem is that this circuit cannot be applied in discharge without consuming a portion of the charge capacity, if it were somehow applied except when cranking the APU.
Therefore, @USMCProbe (#460);
i agree that no charging should be done in the air, let the tow vehicles power the movement lights; and add that sufficient time and sensors should be provided to assure that the battery is cooling down after being charged.
Beyond all that, I wonder if these DC chargers are putting out currents with too much AC ripple on it (the smoothest filtering requiring heavy choke coils inconvenient to carry aloft)?
Again, I'm inclined to more effective system of very pure DC being supplied from the ground for balancing both up or down on charging, supplied as a regulated current output (not primarily voltage regulated) for each cell, based on a digital demand generated by voltage testing by the aircraft CBs. More elaborate and expensive (surely $16k would cover), but eliminates a couple more uncertainties. Of course, that may not do it, and this is not the time to be running tests (ie, in revenue service) as has been noted.
Thanks for commenting on my post (#364) about charging arrangements and balancing. The TI seminar talk on this was most interesting:
Here it is how balancing is done in practice.
http://focus.ti.com/download/trng/do...0and%20How.pdf
http://focus.ti.com/download/trng/do...0and%20How.pdf
Obviously, in those parts of the charging cycle where the correct action is unknown, there is the possibility of unsuspected harm from the balancing. Altho presumably small each time, it could accumulate. There is also mention that balancing of charge is more effective and more needed approaching fully charged; but that this part of the cycle is short, which limits the effectiveness. And similarly, balancing of discharge is recommended; but set up to drive a starter motor, this is not possible in the 787.
This 8-page talk has nine figures (9th didn't download for me), and is well worth reading. It is general to several battery types, but follows with some detail specific to lithium batteries. It is undated (unless I missed it). The examples seem to be small batteries (300-500mAh) in my impression, but some of the data can be extrapolated I think, altho not the experience.
One thing that concerned me was the MOSFET regulator chip idea-- if that is general and widely used, the concept would appear to vary the effective bypass by intermittently applying a fixed bybass. (Obviously a 787 battery regulator could not fit on a single chip.) During the no-bypass intervals, the effects to be avoided (e.g. lithium plating out at over-voltage on one cell) would still occur even if for reduced time. Another problem is that this circuit cannot be applied in discharge without consuming a portion of the charge capacity, if it were somehow applied except when cranking the APU.
Therefore, @USMCProbe (#460);
i agree that no charging should be done in the air, let the tow vehicles power the movement lights; and add that sufficient time and sensors should be provided to assure that the battery is cooling down after being charged.
Beyond all that, I wonder if these DC chargers are putting out currents with too much AC ripple on it (the smoothest filtering requiring heavy choke coils inconvenient to carry aloft)?
Again, I'm inclined to more effective system of very pure DC being supplied from the ground for balancing both up or down on charging, supplied as a regulated current output (not primarily voltage regulated) for each cell, based on a digital demand generated by voltage testing by the aircraft CBs. More elaborate and expensive (surely $16k would cover), but eliminates a couple more uncertainties. Of course, that may not do it, and this is not the time to be running tests (ie, in revenue service) as has been noted.
Last edited by Old Engineer; 5th Feb 2013 at 06:25.
Join Date: Mar 2012
Location: In a remote jobless corner
Posts: 76
Likes: 0
Received 0 Likes
on
0 Posts
Well so I've been following news closely and I cannot seem to fathom how boeing missed out on such a minor glitch, how did they get the FAA's approval and by how much does this set Boeing back in terms of cost of replacement , spare parts , deliveries etc. IMO Boeing were counting on the 787 to do exceedingly well in the market ..?!
Join Date: May 2010
Location: EU
Age: 54
Posts: 22
Likes: 0
Received 0 Likes
on
0 Posts
exercise of imagination
I wonder why they can't just INOP the battery chargers during flight until
they find the problem.
- Change the batteries after each flight with precharged and monitored/tested ones for at least ... days;
- After any battery kicks in, monitor the electronic bay /and land asap;
- Disable the charging circuit and install one.. /two..(as required) safe type batteries as stand by.... Eventually keep one on board and connect in really need only (If these bateries are indeed safe enough to be 'transported' in an aircraft and not to sensitive to fly (position, pressure variations, atmosphere, ..... not only temperature)
I wonder why they can't just INOP the battery chargers during flight until they find the problem.
Join Date: Aug 2011
Location: hong kong
Age: 63
Posts: 93
Likes: 0
Received 0 Likes
on
0 Posts
The only component that "for now" seems less likely, is GS Yuasa cells.
Last weeks update and this weeks activities: Press Release February 1, 2013
Last weeks update and this weeks activities: Press Release February 1, 2013
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes
on
0 Posts
I'm still concerned with the 6 sub-cells of each of the eight cells.
They are never equal/matched perfectly. Current oscillations may occur, but even a totally unmonitored thermal runaway...
Check this:
Serial and parallel battery configurations
They are never equal/matched perfectly. Current oscillations may occur, but even a totally unmonitored thermal runaway...
Check this:
Serial and parallel battery configurations
A high resistance or open cell is less critical in a parallel circuit than the serial configuration but the parallel pack will have reduced load capability and a shorter runtime. It's like an engine running only on three cylinders. An electrical short would be more devastating because the faulty cell would drain the energy from the other cells, causing a fire hazard.
A weak cell will not affect the voltage but provide a low runtime. A shorted cell could cause excessive heat and create a fire hazard.
Is charging the problem?
Is the problem the charging of the cells?
Following the different threads the batteries are not used under normal ops, there schould be no charging cycle necessary. But nearly 150 batteries have been changed by the operators due to being in a critical low charging state. It can be assumed, that most of those low charging conditions were observed and most probably also took place on the ground, as otherwise we would have known about it earlier by some diversion reports. Both incidents happened on the ground or shortly after departure.
Something must have caused the unplanned discharging of those batteries, what we observe in the incidents is most probably the outcome of this prior failure.
Therefore forgive me when i dont think it is wise to look for the general problem in the battery or in the charging system. The battery should be protected against turning into some fireworks if for some reason a critical low charging state happens, but it shouldn´t drain either that low on a regular basis.
Therefore the problem has to be somewhere in the wiring and design of the whole instalation into the aircraft, an integration problem.
The questions being:
- What causes the draining of the batteries?
- When does the draining process start?
- Why are the built in provisions to prevent this low state ineffective?
- What systems have to be reengineered to prevent excessive discharging?
and at last
- How can the battery be protected against going fireworks.
Following the different threads the batteries are not used under normal ops, there schould be no charging cycle necessary. But nearly 150 batteries have been changed by the operators due to being in a critical low charging state. It can be assumed, that most of those low charging conditions were observed and most probably also took place on the ground, as otherwise we would have known about it earlier by some diversion reports. Both incidents happened on the ground or shortly after departure.
Something must have caused the unplanned discharging of those batteries, what we observe in the incidents is most probably the outcome of this prior failure.
Therefore forgive me when i dont think it is wise to look for the general problem in the battery or in the charging system. The battery should be protected against turning into some fireworks if for some reason a critical low charging state happens, but it shouldn´t drain either that low on a regular basis.
Therefore the problem has to be somewhere in the wiring and design of the whole instalation into the aircraft, an integration problem.
The questions being:
- What causes the draining of the batteries?
- When does the draining process start?
- Why are the built in provisions to prevent this low state ineffective?
- What systems have to be reengineered to prevent excessive discharging?
and at last
- How can the battery be protected against going fireworks.
Last edited by RetiredF4; 5th Feb 2013 at 12:23.
Join Date: Aug 2011
Location: hong kong
Age: 63
Posts: 93
Likes: 0
Received 0 Likes
on
0 Posts
@hetfield
Appreciated, a sub cell consists of an anode-separator-cathode triplet. All but the smallest cells consist of many paralleled sub cells. It's like that for more than 100 years.
Sub cells shortens --> cell short + heat. Sub cell opens --> reduced cell capacity and increased inner resistance with potential for cell reversal.
Important here is that BMS+charger look after the cells, keeping them in their operating envelope and mitigate cell faults.
Mitigate: DISCONNECT charger on any cell fault. When in doubt, do not charge.
Appreciated, a sub cell consists of an anode-separator-cathode triplet. All but the smallest cells consist of many paralleled sub cells. It's like that for more than 100 years.
Sub cells shortens --> cell short + heat. Sub cell opens --> reduced cell capacity and increased inner resistance with potential for cell reversal.
Important here is that BMS+charger look after the cells, keeping them in their operating envelope and mitigate cell faults.
Mitigate: DISCONNECT charger on any cell fault. When in doubt, do not charge.
Last edited by saptzae; 5th Feb 2013 at 12:22. Reason: cathode, mitigate
Join Date: Jan 2008
Location: lancs.UK
Age: 77
Posts: 1,191
Likes: 0
Received 0 Likes
on
0 Posts
^^^^^ 466 the answer is, surely, a fusible link within the cell itself.
On more than one occasion I've seen a lead-acid automotive battery where the internal lead straps have melted inter-cell or individual plates have self-disconnected, simply by melting the internal connection
I made a couple of posts on the R&N thread....synopsis...charging unit and battery unit don't interface properly. boeing appear to have failed to ensure they did.....otherwise you wouldn't change out so many rechargeable batteries in such a short space of time..
On more than one occasion I've seen a lead-acid automotive battery where the internal lead straps have melted inter-cell or individual plates have self-disconnected, simply by melting the internal connection
I made a couple of posts on the R&N thread....synopsis...charging unit and battery unit don't interface properly. boeing appear to have failed to ensure they did.....otherwise you wouldn't change out so many rechargeable batteries in such a short space of time..
Last edited by cockney steve; 5th Feb 2013 at 12:26.
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes
on
0 Posts
Appreciated, a sub cell consists of an anode-separator-cathode triplet. All but the smallest cells consist of many paralleled sub cells. It's like that for more than 100 years.
Important here is that BMS+charger look after the cells, keeping them in their operating envelope and mitigate cell faults.
They have their own social life. The rich feed the poor until dead.
The questions being:
- What causes the draining of the batteries?
- When does the draining process start?
- Why are the built in provisions to prevent this low state ineffective?
- What systems have to be reengineered to prevent excessive discharging?
and at last
- How can the battery be protected against going fireworks.
- What causes the draining of the batteries?
- When does the draining process start?
- Why are the built in provisions to prevent this low state ineffective?
- What systems have to be reengineered to prevent excessive discharging?
and at last
- How can the battery be protected against going fireworks.
Towing on Battery only for too long.
Fuelling on battery only for too long.
APU switch left in 'On' position (to vent APU core)
In short a lack of education by Boeing to it's students during training.
I for one have only learned about the major problem with this battery (IE don't let it drop below a certain charge state or it's u/s) on pprune. Not Boeing.
lack of training
@turin
That would be an explanation for some batteries, but not for 150. After the first unplanned changes normal people learn from mistakes. Is the true problem masked by some of these causes as you describe them?
And are you saying, that the battery changes have nothing to do with the known two incidents from ANA and JAL?
Boing and the 787 operators loose big money, and a lot of highly qualified people are looking into the problem. There are some possibilities imho:
1.) they know the cause already, but there is no quick solution at hand /approved by FAA and therefore there is no need to tell anybody, or
2.) they are looking in the wrong corner, are still looking at the symptoms and have´nt found the causes yet.
If an easy work around (like diconnecting from chargers) would be practicable, we would see it already. They checked the batteries, they checked the chargers, they are checking the PCB´s. They have hundreds of used and new batteries available for testing, and the grounded aircrafts as well. They came out with nothing until now.
Whatever we will see, it will be very interesting.
That would be an explanation for some batteries, but not for 150. After the first unplanned changes normal people learn from mistakes. Is the true problem masked by some of these causes as you describe them?
And are you saying, that the battery changes have nothing to do with the known two incidents from ANA and JAL?
Boing and the 787 operators loose big money, and a lot of highly qualified people are looking into the problem. There are some possibilities imho:
1.) they know the cause already, but there is no quick solution at hand /approved by FAA and therefore there is no need to tell anybody, or
2.) they are looking in the wrong corner, are still looking at the symptoms and have´nt found the causes yet.
If an easy work around (like diconnecting from chargers) would be practicable, we would see it already. They checked the batteries, they checked the chargers, they are checking the PCB´s. They have hundreds of used and new batteries available for testing, and the grounded aircrafts as well. They came out with nothing until now.
Whatever we will see, it will be very interesting.
Last edited by RetiredF4; 5th Feb 2013 at 13:05.
Join Date: Aug 2011
Location: hong kong
Age: 63
Posts: 93
Likes: 0
Received 0 Likes
on
0 Posts
@Old Engineer
Cell balancing is nothing new, used with large telco installations for the better part of 100 years.
In small batteries (up to starter battery size) Ni*, Pb* took care of itself well enough for practical use. However, cell imbalance is present there too. In a typical old car battery, one or two cells will typically cause the need for replacement. These cells were often impaired by imbalance.
What's newer is the need for precise management of smaller Li* batteries. Every small battery has circuitry (these days often a single chip) built in.
For small cells up to a few Ah, balancing by resistive means is sufficient. The differences amongs cells are just a few percent, a balancing current in the 10s of mA is sufficient. Important is to prevent cell over/under voltage.
Discharge balancing is never done in high load applications. The available capacity is limited by the weakest cell.
Individual cells, which dropped capacity by a few percent are routinely replaced in large installations. Some, like stationary 48V Pb telco battery cells can be refurbished. I guess that it would be possible to replace individual cells of reduced capacity in the 787 battery.
Cell balancing is nothing new, used with large telco installations for the better part of 100 years.
In small batteries (up to starter battery size) Ni*, Pb* took care of itself well enough for practical use. However, cell imbalance is present there too. In a typical old car battery, one or two cells will typically cause the need for replacement. These cells were often impaired by imbalance.
What's newer is the need for precise management of smaller Li* batteries. Every small battery has circuitry (these days often a single chip) built in.
For small cells up to a few Ah, balancing by resistive means is sufficient. The differences amongs cells are just a few percent, a balancing current in the 10s of mA is sufficient. Important is to prevent cell over/under voltage.
Discharge balancing is never done in high load applications. The available capacity is limited by the weakest cell.
Individual cells, which dropped capacity by a few percent are routinely replaced in large installations. Some, like stationary 48V Pb telco battery cells can be refurbished. I guess that it would be possible to replace individual cells of reduced capacity in the 787 battery.
Join Date: Aug 2011
Location: hong kong
Age: 63
Posts: 93
Likes: 0
Received 0 Likes
on
0 Posts
@hetfileld
Battery is an assembly of cells, connected in series, each cell consists of one or more sub cells in parallel. Whether the sub cells are in the same housing does not matter.
It can't and it does not have to as the sub cells are all parallel.
Right, since life began, or energy that is.
Battery is an assembly of cells, connected in series, each cell consists of one or more sub cells in parallel. Whether the sub cells are in the same housing does not matter.
Yes, but how can a charger/BMS look after the sub-cells?
They have their own social life. The rich feed the poor until dead.
Last edited by saptzae; 5th Feb 2013 at 13:41.
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes
on
0 Posts
RetiredF4....
Almost certainly. And known cause is what caused the feverish replacement scheme. Buy time, hope for a fix.....with the blessing of the FAA.
Certainly "Flight Test" with people inside...
While Rolls R/Rs Intermediate shafts, the danger lurked..(QF32)
For all intents, a DENDRITE is an inadvertant "Fusible LINK"
1.) they know the cause already, but there is no quick solution at hand/approved by FAA and therefore there is no need to tell anybody...
Certainly "Flight Test" with people inside...
While Rolls R/Rs Intermediate shafts, the danger lurked..(QF32)
For all intents, a DENDRITE is an inadvertant "Fusible LINK"
Last edited by Lyman; 5th Feb 2013 at 14:47.
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes
on
0 Posts
On Feb 5th 2013 the JTSB released a second progress report in Japanese reporting that all 8 cells of the damaged battery, nominal voltage 29.6V, 75 Ah capacity at 28.5kg/63 lbs, showed thermal damage before the thermal runaway, particularly cells 3 and 6 are damaged. The positive electrode of cell 3 shows substantial damage and a hole, the internal wiring has melted down.
8 cells?
48 I would say....
Sub cells not monitored/balanced/controlled at all!
check # 466
Last edited by hetfield; 5th Feb 2013 at 17:46.
Join Date: Mar 2006
Location: Choroni, sometimes
Posts: 1,974
Likes: 0
Received 0 Likes
on
0 Posts
"Large cells without enough space between them to isolate against the cell-to-cell thermal domino effect means it is simply a matter of time before there are more incidents of this nature," Elon Musk, CEO of electric car company Tesla Motors, wrote in an e-mail to Flight International last week.
Very funny, eight Yuasa cases, called cells, contain 6 sub-cells each.
Join Date: Aug 2011
Location: hong kong
Age: 63
Posts: 93
Likes: 0
Received 0 Likes
on
0 Posts
Damage seems worse than at BOS. Even a terminal was burned off.
Failure scenario seems similar to BOS though. A cell short, over voltage on the other cells and then thermal runaway.
@RetiredF4
Could be pre-existing, caused by intervention or a short to the case from a current source such as a cell shorting to its case. In case of short, I would expect more melting and burned insulation though.
If pre-existing, could cause floating battery case and (very remote) possibility of EMI disturbing BMU (the PCB's).
Failure scenario seems similar to BOS though. A cell short, over voltage on the other cells and then thermal runaway.
@RetiredF4
Frayed and broken earthing wire
If pre-existing, could cause floating battery case and (very remote) possibility of EMI disturbing BMU (the PCB's).
Last edited by saptzae; 6th Feb 2013 at 04:59. Reason: Typo, EMI
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes
on
0 Posts
The ANA battery failure has a different failure signal, imho. Although we see the BOS battery after complete combustion, in ANA, there is evidence of a high velocity, high temperature "jet" of gas from the top slicing through the connector. The energy is impressive. Does anyone think as I do, that may be evidence of a metallic dendrite cooking off?