Found in the news, Boeing's dremliner LiIon battery unit burned out picture

On a sunny day (Thu, 17 Jan 2013 20:58:25 -0500) it happened "P E Schoen" wrote in :

Ah, I see R2 is now 10 Ohm, much better.

I was unsure whether to use PWM to switch the MOSFET so only R1 will get

hot, or to use linear mode and let it dissipate some of the heat. I really don't know if using PWM to shunt the charge is OK for the battery.

The PIC has an internal voltage reference that can be set to 1.024 or

2.048V, so Vdd change won't be a major issue. Thanks for the observation.

Paul

Yes, still a work in progress. Easy enough to change or remove components on a PCB.

I am still unsure about the reverse polarity diode D1. I agree that the cell is probably toast (preferably the cold dry kind) by the time it has weakened to this point, and the load should have been removed well before it dropped to a critical low voltage level, but if all else fails, and especially if the cell opened up in a 300 or 600V battery pack, at least it would prevent that voltage from being imposed on the cell. It may very well fail shorted, which would be OK. But if not, the Schottky should be able to handle up to

40 amps and may prevent a major destructive event with 40*300 = 12kW in a cell the size of a paperback book!

Paul

On a sunny day (Fri, 18 Jan 2013 19:52:01 -0500) it happened "P E Schoen" wrote in :

What happens if somebody shorts the battery terminals? Schottkys have a low reverse voltage, Can a cell that then goes defective (open circuit) cause this reverse voltage to dissipate ???kW in that D1? ----- |short | | - | --- now - Vmax | |----------- a | open cell D1 | |----------- k | - now + | --- | | | - | --- | | | etc etc |_____ |

:-)

----- |short | | - | --- now - Vmax | |----------- a | open cell D1 | |----------- k | - now + | --- | | | - | --- | | | etc etc |_____ |

Well, shorting the battery terminals is always a rather severe insult to its integrity, and the only reasonable way to protect against that is with the proper fuse. The MBR2045CT is rated at 20A nominal and 150A surge, so with both devices in parallel it might handle 300A. For my purposes, this pack would be only 2-10Ah rating with expectation of up to 3C or maybe 5C, so a fuse of 5A-30A would be sufficient, and should blow instantaneously at

300A. But fuses for 300VDC or 600VDC which are common pack voltage are rather expensive, like $20-$30 each. But that's way cheaper than a vehicle fire, and battery packs are generally thousands of dollars. df

There are large fuses of several hundred amps designed for EVs, as ultimate catastrophic safety devices, but there is a known problem with EVs when it comes to crashworthiness and danger to occupants and first responders from high voltage and high current battery packs. Some people have installed contactors which isolate the packs into 48 or 60 volt sections which are

less hazardous. I'm thinking about a modular design where each pack is

24-60V and no more than 10-20 Ah and have a BMS built into each module. It may also have a contactor which isolates the main terminal of the pack until a control signal is applied, so that in case of a major failure, hopefully an interlock system would deactivate all of the packs, and if they are designed to limit damage from catastrophic failure, it would not spread to other packs and cause the chain reaction meltdown that has been known to

happen.

Paul

Thanks. I had no idea it had all-electric brakes as well. Those were never bleed air either, but I'm guessing it makes the ABS far smarter and easier.

Still think the most impressive braking job I have seen is the

777 rejected takeoff test

--
A host is a host from coast to coast.................wb8foz@nrk.com 
& no one will talk to a host that's close........[v].(301) 56-LINUX 
Unless the host (that isn't close).........................pob 1433 
is busy, hung or dead....................................20915-1433

Here's an article by an expert on lithium technology and BMS units, speculating about possible causes:

fire

He is a regular on the

forums.

Paul

I am mistified. Even the simple 3.7V packs in our equipment is fitted with "BMS" protection PCB's that uncouple the pack in case of under/over voltage and overcurrent charge / discharge. Granted, no 500+Amps rated, but even a ye olde heavy metal relay could do that.

If 1000A+ charge and 500A+ discharge currents can occur unchecked - ... well I hope they put more thought in the rest of the plane.

How ironic!

GS Yuasa Searched After Boeing 787s Are Grounded By Chris Cooper and Anna Mukai on January 22, 2013 Tweet

Japan?s transport ministry is conducting a second probe of GS Yuasa Corp. (6674)?s headquarters in Kyoto and sent a team to a U.K. supplier to investigate faults that have grounded Boeing Co. (BA)?s

787 Dreamliner fleet.

Officials returned to offices in Kyoto today after a first inspection was undertaken yesterday with U.S. regulators, Akihiro Ohta, Japan?s transport minister, told reporters in Tokyo. Investigators were also sent to the U.K. to probe a valve actuator maker for the 787, he said, without identifying the company.

GS Yuasa?s batteries are the focal point of the investigation into the causes of a fire on a Japan Airlines Co. (9201) plane and an emergency landing by an All Nippon Airways Co. (9202) jet. Chicago-based Boeing can?t deliver more 787 jets until the U.S. Federal Aviation Administration confirms the safety of the Dreamliner.

?I can?t think of any example where the equipment provider gets sued in a case like this,? said Edward Stacey, a London- based analyst at Espirito Santo Investment Bank. Regulators, airlines, and aircraft makers have responsibility for safety, giving suppliers protection unless their product doesn?t meet requirements, he said in a phone interview.

Battery tests start today in a center run by the Japan Aerospace Exploration Agency, Ohta told reporters. An NTSB group of four officials and two Japan transport officials will do the inspection at the agency, Masahiro Kudo, an investigator in the ministry?s Japan Transport Safety Board said.

Fully Cooperating The battery will be taken to Kyoto for further analysis after checks including a computed tomography scan at the agency, Kudo said.

Yesterday?s search at GS Yuasa started at 11 a.m., and the ministry sent one official, while the FAA sent two people to Kyoto, Shigeru Takano, a director for air transportation in the ministry?s Civil Aviation Bureau, told reporters in Tokyo yesterday.

The company is fully cooperating with the authorities, Tsutomu Nishijima, a spokesman for the lithium-ion battery supplier, said by phone yesterday.

GS Yuasa, the world?s biggest motorcycle-battery maker, fell 1.9 percent to 312 yen in Tokyo today, after two consecutive days of gains.

Shares of Christchurch, England-based Meggitt Plc (MGGT), parent of Securaplane Technologies Inc., rose less than 1 percent to 430 pence at 2:59 p.m. in London. Securaplane?s Tucson, Arizona- based unit provides elements of the 787?s lithium-ion battery system.

Battery Charger In another part of the investigation, the U.S. National Transportation Safety Board will be visiting an Arizona firm Tuesday to probe the battery charger on the jet that caught fire in Boston.

Securaplane made the battery charger on the jet that was involved in the Jan. 7 fire in Boston. The NTSB, which investigates accidents and has no regulatory authority, will be working with the firm for help in testing the battery charger, Kelly Nantel, an NTSB spokeswoman, said in an e-mail.

The NTSB also hopes to capture data that may help the investigation from a memory module connected to the battery system, according to an e-mailed release from the agency.

Other components involved in the Boston jet fire have been sent to Boeing and to unidentified manufacturers in Japan for analysis, according to an NTSB release.

The Japan Air plane that caught fire in Boston on Jan. 7 didn?t exceed its intended voltage, according to the U.S. National Transportation Safety Board. Information released Jan. 20 in an e-mailed statement by the agency suggests the fire that precipitated a grounding of Boeing?s newest jet wasn?t caused by overcharging. The NTSB statement didn?t identify a cause of the battery fire.

Car Batteries GS Yuasa?s multiyear, multimillion dollar contract to supply batteries to Thales (HO) SA, announced in June 2005, was an opportunity to offset losses from sales to carmakers. While Kyoto-based GS Yuasa said in 2009 that lithium-ion batteries for vehicles will become a core business for the company, the battery maker since then hasn?t made a profit on the technology.

?They had been hoping to make up for the lack of sales to carmakers by selling to Boeing,? said Jun Yamaguchi, an analyst at Credit Suisse AG in Tokyo. ?Any inability to sell in the aviation market is going to make the lithium-ion battery business even more unprofitable for GS Yuasa.?

od-news-for-boeing

fire

He understates the damage of the battery from the ANA flight that made the emergency landing in Japan. Approximately ten pounds of electrolyte leaked from the battery, through the mounting containment, and through the floor t o the outside of airframe. The electrolyte is conductive as well as highly corrosive, capable of destroying physical integrity and functionality of po ssibly critical flight control components such as wiring bundles it comes i n contact with. Also, I think his estimate of peak charge currents are a bi t off as Boeing literature claims a 73 minute charge time, which, allowing for 80% efficiency, would make the charging current peak at no more than 65 AH/(73*0.8)x60= 67 Amps. I'll add a fifth hypothesis in that the connect orization was some auto industry kluge that broke the BMS-aircraft link, Bo eing did afterall import Detroit management into the Dreamliner job. The in vestigators are barking up the wrong tree looking for anything even remotel y close to a design or manufacturing flaw at Yuasa. But Securaplane is anot her story, they burned down their entire manufacturing facility a few years back doing lithium battery overcharge testing, of all things.

the

leaked

floor

minute

charging

fifth

that

management

tree

flaw

entire

overcharge

There is an interesting test of high current discharge of a 170 Ah LiFePO4 prismatic cell at 1500 amps for about 15 minutes without damage (except to the wrench that was used as a load in a bucket of water):

(starting at time mark

1:02)

(at time mark 1:30 or so)

I figured that the internal resistance of the cell used in the short circuit test was about 470 uOhms and a true bolted short would be about 6800 amps. There's a good picture of a lithium cell fire in the thread:

The thread itself is interesting:

896.html

And there is a chart in this thread showing measurement of short discharge time as a measure of peak current capacity. The battery in the discharge

test had an 89 second figure. The range is from about 12 seconds to 220 seconds:

Paul

probably takes a

the event of power loss in flight. It is the key piece that gets them ETOPS certification without which the 787 would not be allowed the more fuel saving flight routes, among other things. How much of a drain on the batteries is starting the APU is what I want to know if the manufacturer's news release brags about a 75 minute fast charge time??? I hope this thing glides well.

That is more a function of the size of the company/organization, large aerospace companies have lots of doofuses (BTDT).

Believable, but not necessarily so. There is a lot of available power in the ram air. Just the same, motorizing the generator to bring it up to speed quicker can take a lot more VA than real power.

According to the BMS expert you linked, the battery was internally fused. Those kinds of current would open any respectable fuse in milliseconds.

On a sunny day (Wed, 23 Jan 2013 22:40:46 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

kinds of current would open any respectable fuse

I have wondered if temperature changes, combined with humidity changes, could cause moisture to be sucked into the battery, and then causing problems, or when expanding (melting ice) could perhaps break the casing, causing the thing to start leaking. Bit of expanding ice in cracks can break a rock.

As to charging, I have read that the charge voltage never exceeded 33 V. (give or take journalist interpretation), from teh flight recorders.

. Those kinds of current would open any respectable fuse

ould

Those measurements are really gross measures of battery state of health. Th e problem with this technology is the highly reactive nature of the electro de materials and electrolyte, all kinds of aging characteristics, all kinds of transformations and growth processes that can take place as a result of impurity concentrations in the raw materials or introduced during manufact uring, all kinds of potential for the development of weak spots/hot spots t hat don't show up in monitoring sensors like thermistors and cell voltage m onitors, the list is endless. All the safety backups internal to the cells like shut-down separator, tear-away tab, pressure vent and thermal interrup t are easily defeated.

On a sunny day (Thu, 24 Jan 2013 10:16:14 -0800 (PST)) it happened snipped-for-privacy@gmail.com wrote in :

That is a lot of chemistry...

What would the consequences or possibilities be for Boeing to replace those Li-ion batteries with for example NiMH or maybe even a fuelcell [1]?

I remember Apollo spacecraft had a fuel cell for power (exploded in Apollo 13), and 4 sure weight must have been important there. I also remember some aircraft already have fuel cells.

Also Airbus has planned Li-ion in their new design, what will the consequences be for theirs?

Sure Being will want to fly again ASAP, so changing battery type could be a possibility?

[1] From: Greener aircraft and reduced emissions Parker Aerospace is a longtime Airbus supplier with special competencies in multifunctional system integration. Parker is partnering with Airbus to develop fuel cell technology as an alternative energy source for on-ground and in-flight electrical power supply. Within this partnership, Airbus will be responsible for the overall aircraft system architecture

Th=

tro=

nds=

of=

act=

s t=

e m=

ls =

rup=

se Li-ion

The particular brand of LiCoO2 they're using now has energy density of 200 versus NiMH running at 60. So they would need about 200 lbs of NiMH to repl ace the existing 60-pound battery. Of course the aircraft interface has to change, and, although straightforward, will be extensive rework.

o 13),

If they're smart, those plans will be canceled.

a possibility?

Looks like it's going to be a requirement. The FAA is in the hot seat right now. Congress Senate Commerce, Science and Transportation Committee is in the process of setting up special hearings right now, and the same expertis e and industry groups that came out against the FAA special rule in 2007 al lowing Boeing to use the battery will be there in force to continue their f ight to get the lithium battery removed.

in multifunctional system integration. Parker is partnering with Airbus to develop fuel cell technology as an alternative energy source for

, Airbus will be responsible for the overall aircraft system architecture

-fire

The cell specifications are downright frightening, capable of currents in excess of 5000 A. And the whole box handling currents in excess of 1000 A as well. There are not a lot of switches that can interrupt that kind of current at 32 V DC.

?-)

It's even more crazy in an EV where it is common to have multiple cells in series for a pack voltage of 150, 300, or even 600V, with 100 Ah cells, so total energy can be upwards of 50 kWh (or a gazillion Joules). There was a discussion of an electric semi truck conversion with a range of 750 miles at

80 MPH and we figured it would be something like 750 kWh. And the peak currents of these cells are typically 5C to 10C under normal operation for acceleration, and short circuits may well be 50C or more. 10,000 amperes at 600V? More than enough for a flux capacitor and a ticket to the future!

Fortunately, there are fuses that can break 1000 VDC and 10,000 amps interrupting current. They are also peak limiting types which react to the rate-of-rise so they start opening before the current reaches its potential maximum based on the circuit impedance. This is one place where AC is safer, since it crosses zero 100-120 times a second and it is limited by inductive reactance as well as resistance. For a DC fault, the inductance still limits the rate-of-rise, but then when the protective device opens the stored energy will cause an arc.

I'm looking into a comprehensive system to minimize the dangers of Li-Ion batteries. If you can detect dangerous conditions before they become catastrophic, you can break the current with an IGBT, which switches much more quickly than an electromechanical relay. But it does not provide true isolation. So I figure that you can have an IGBT and a relay in series, and stop the current with the solid state device, and then open the relay for true isolation. If the relay does not have to switch any significant current, it does not even need to be DC rated. You can get a 250 VAC or even a 600 VAC relay with large contacts capable of a hundred amps, and they will provide enough spacing to hold off 600 VDC or more, especially with two or more poles in series.

High voltage DC rated relays generally have magnetic arc blowouts and/or are sealed with vacuum or an insulating gas like SF6, and are really expensive. See Kilovac and Gigavac relays. Maybe there are also Megavac and Teravac but I haven't seen them... :)

Most DIY EV folks use LiFePO4 cells which are generally much safer than the Li-Ion, but they are twice as expensive. Probably when the Yuasa/Boeing battery packs were initially designed, the LiFePO4 were fairly new, and because of the long time needed for detailed design, testing, and approval, the newer and safer technology was unable to be implemented quickly enough for the management types, so they pushed it through, sort of like what they did when they allowed the Challenger to take off.

Paul