FLAGSTAFF, Ariz.--Investigators from the U.S. National Transportation Safety Board
(NTSB) arrived in Tucson, Ariz., Tuesday (Jan. 22) as the focus of
electronics problems that have grounded Boeing's 787 Dreamliner
shifted to battery-charging electronics.
The NTSB team is visiting Securaplane
Technologies, an avionics unit of U.K.-based Meggitt, to
test the company's battery-charging devices as well as wire bundles
and battery-management circuit boards. Japanese and U.S. airlines
grounded their 787s on Jan. 16 while investigators in both nations
looked into battery and other problems that have plagued the
airliner in recent weeks.
The Dreamliner, a huge business and
technology bet for Boeing, leverages advanced electronics, millions
of lines of software code and lithium ion battery technology to
claim 20 percent better fuel efficiency than the Boeing 767 it's
intended to replace.
Investigators said they examined the 32V battery using X-ray and CT
scans, disassembled the APU battery into its eight cells for
detailed examination and documentation. Three of the cells were
selected for more detailed radiographic examination to view the
interior of the cells prior to their disassembly, the NTSB said in a
statement. The agency added that an examination of the
flight recorder data from the JAL B-787 airplane indicate that the
APU battery did not exceed its designed voltage of 32 volts.
Securaplane, acquired by Meggitt in 2011, has been in business since
1986and offers a variety of avionics including aircraft
security systems, LiOn batteries and chargers and inverters--the
last leveraging Vienna rectifiers, power MOSFETs/IGBTs, planar
transformers, ultracapacitors, and advanced pulse width modulation
techniques, according to the company's web site.
The charger in question, the Boeing
787 BCU (schematic pictured above), uses advanced DC to DC conversion technology,
patented charging algorithms, comprehensive diagnostics and fault
isolation to charge the APU battery, according to company
Measuring 14.7 x 5.0 x 7.7 inches and weighing 11 pounds, the
rack-mounted system features 23VDC minimum and 80 Amps maximum input
power and 1500W max output power.
I don't understand the comment about exceeding the design voltage. You wouldn't expect an overcharge to result in higher battery voltage, unless they're talking about the charging voltage being higher than spec.
Overcharging LiOn cells is manifested by continuing to run too much current through them after the charging current has dropped to a low level, like 3 percent of the cell's rated current capacity. But the voltage, at that point in the charging cycle, has been at a constant level for most of the charging period. It quickly ramps up in the initial charging period, then settles as the reverse current through the battery drops steadily.
Perhaps they meant exceeding the charging voltage, during the charging cycle?
Overcharging a battery is not the same as overfilling a gasoline tank, if you are equating voltage with gallons of gasoline.
If this is designed improperly or to close to the margins...not good
If the power cycle confuses the algorithm for the timer...not good
If Boeing took at a ground wire out to save weight and lightning strikes create 100's to 100's of differential voltage across the fuslage reverse biasing FETS in contol paths (that are grounded at differennt points than the battery, this can reverse feedcurrent into battery without any way to control it...Not good
Except for the words "lithium" and "fire" there is no similarity to the single Volt issue last year. In that case the battery pack was intentionally damaged by smashing it into a steel beam and leaving it for a week. They speculate that current through the leaked coolant caused the fire.
If you want to make a comparison without facts a better one is the lithium laptop batteries that were spontaneously catching fire and exploding a couple years ago.
I work with two way radios,all use now Li batteries,I do not think there is an overcharging problem, all this batteries, have a small protection PCB inside, that drains 4 uA. This PCB limits the charging and discharge current to safe limits, and it also controls the voltage allowed to reach the battery. But I did received once a radio case with a burned battery that was not in use. The only explanation is the failure of internal parts, that shorted BOTH sides of the battery. To limit this problem, all shipments by plane today limit the battery capacity to a safe 30% level. This is new technology, and those who lead, take a risk.
The latest public statement from the Japan Transport Safety Board says the battery did not experience overvoltage, but rather, a sharp and unexpected drop in voltage. So no evidence of overcharging, but perhaps over-discharging into a hard short for a period of time. Hopefully they can quickly get to the root cause.
I have a friend that worked at Securaplane and he witnessed one of these battery systems catch fire and burn down the building he was working in. Seems nothing was reported on the incident - perhaps a cover-up?
There is something fishy about the details in this story. According to the NTSB's own website, there are two pictures of the damaged battery and one picture of an undamaged battery (see here):
If you zoom into the undamaged battery picture, you can clearly see that it was manufactured by a French company called Thales (where they boast about being the first to deploy this technology on the B787):
Both undamaged and damaged battery appear to be identical in dimensions, color, and decal location.
According to the story above, the battery system is provided by a company called Meggitt/Securaplane and the battery charger schematic (shown above) looks completely different than the one shown on the NTSB website. Why the discrepancy about the real battery manufacturer?
Nothing fishy. In aerospace, it's normal to have primes and subs. From what I saw on the Thales web site, they were the prime contractor on the power conversion & battery system, and Securaplane was a sub-contractor that provided the integrated battery system -- battery, charger and battery management unit. It appears that GS Yuasa provided the actual lithium ion cells (LiPO4), perhaps as a sub to Securaplane, but that last point isn't completely clear.
"...the APU battery did not exceed its designed voltage of 32 volts. " do not guaranteed each cells inside the APU battery have not been overcharged.
It is dangerous ,if the charger over charge protection were controlled by voltage of the APU battery , instead of voltage of each cells.
The NTSB site linked to in the article has some interesting info - it shows a report that has a photo of a battery cell with a "damaged electrode - internal s/c."
The battery chemistry used is Lithium Cobalt Oxide. I would have expected the safter Lithium Iron Phosphate to be used instead (or another, safer-chemistry type available) in this safety critical application. For my money, this is one of the key factors that I would expect to see them change in future designs.
I tried to see from the photos what (if any) power conversion electronics were in the battery box, but couldn't see any high-power, only two low power, circuit boards. Where is the battery charger located? Is it underneath the cells? The bus bars disappear to the underside of the housing and it is difficult to see what is down there. The battery charger power electronics is of course another suspect in the picture I am sure.
Igor1327 is right - normally LiIon cells have integrated OV, UV and Over-Current protection, per cell. External OV protection per cell and external over current protection is also added. I believe you can see the OV protection in the photo of the undamaged unit - it is the loom of small gauge wires connected to each cell's bus bar. Also, integrating a thermistor per cell is standard practice to prevent over-heating. Electrically this should protect it. What I wonder if the fault was in the electrochemistry of the battery + electrodes, then maybe the protection electronics is out of the picture because it cannot protect against a chemical reaction that has started, only against volts, amps. Looking at the picture of the fire damaged battery box, then the cell top-right corner, second one down, looks more damaged than the others.
Please stop thinking Lithium Iron Phosphate won't have same issues. Byd and Volts cases are all Lithium Iron phosphate batteries. They can cause fires even the "trigger points" might be different. It is not clear to me how to prevent it re-ocurrung for BYD or Volt cases yet. And GM has a burnt battery lab. All Li batteries need to be very careful about electronics-charge/discharge/BMS design and very good thermal dissipation design.
I will assume the batteries used for airplane power start, the cranking current is quite high. Only a few chemsitry left for such application and at low temperature environment LiFexOy is not a good choice. If it is charger or thermal design issues, same thing will happen in any chemistry. A great engineering challenge for us to fix this problem. I am more concerned that Li-battery might be dumped in Boeing design before they figure out the reasons.
It seems that no one suspects the effects of flight. Batteries and charging systems were probably tested extensively on the ground - at sea level, but possibly not at altitude. Also, the comment about lightning strikes seems particularly relevant.
The Japan airlines fire started when the plane was on the ground and being cleaned. What event happened during that time I would like to know. Is this when the charger kicked in? I don't know. Also wonder how cold the batteries get when the plane is in the air, or if there is a battery heater function in the design.
Is there any mechanism to monitor voltages in individual cells rather than just the series sum of all the cells? If one or two cells are faulty (or shorted out), the standard charging voltage across the entire battery could produce an over voltage condition in the remaining functional cells.
PHW_#1 I disagree. http://en.wikipedia.org/wiki/Lithium_iron_phosphate_battery gives the reasons why I said LiFePO4 is the safer chemistry. It would be my guess that there is a heater circuit in this battery, because very cold batteries perform poorly.
DrQuine - in a multi-series-cell battery, then it is usual to monitor the individual cell voltages. The Dreamliner battery is comprised of 8 series cells and I think you can see pairs of wires connected to each cell that I guess are for monitoring cell voltages. See slide 9 and zoom in http://www.ntsb.gov/investigations/2013/boeing_787/JAL_B-787_1-24-13.pdf
I agree with DrQuine. Yes, you are right. Every individual cell inside the battery compartment of Li-Ion pack must be individually monitored using the fuel gauge Li-Ion cell over charge/discharge algorithm embedded in the SOC (system on chip) to monitor the state of charge (SoC) of individual Li-Ion cell. If one of the cells over charged or discharged for some reason, one can't prevent the cell from deterioration as the over charged/discharged situation initiates electrolyte (aportic) decomposition which will result in the gas evolution and pressure being built up in every further charge/discharge cycles till it explodes. I think this must have been the issue in Dreamliner case. Battery chemistry must be associated with external electronic control system such as a charger with CC-CV protocol. Hence, Li-Ion cell, although possessing wondering high energy density with high working voltage, the danger comes from many different angle but with one particular issue (over charging)!! Moreover, Anode side (perhaps graphitic carbon or Si-Sn-C composite) should not be over lithiated due to over charging as it the anode in Li-Ion cell is limited (so-called anode limitation to avoid formation of metallic lithium. Hence, a variety of cell issues emanated pausibly from the cell itself. If the latter is ensured, SoC monitoring circuit must have the appropriate CC-CV (constant current-constant voltage) protol. Any comments?