The Dreamliner Project

The meltdown of the Li-ion battery on the Boeing 787 that grounded now 50 planes worldwide is turning Boeing's dreamline into a nightmare. The choice for Li-ion batteries is obvious: high energy, lightweight and quick charge/discharge make them very suitable for use in planes. The backdrop is also clear - Li-ion cells are trouble makers and we as electronic engineers know why… don't we?

[ibimage==2386==Unchanged==none==self==null]

If you really know why and if you have ideas or suggestions of how to get the Dreamliner's Li-ion problems solved, please tell us about it. What is a typical safety circuit you might expect? How to dimension Li-ion batteries, and more important, are Li-ion batteries really safe enough?

Discussion (12 comments)

ClemensValens 11 years ago

Provided by Trevor Langevin, Canada.

I would like to submit more information to your Solve the Dreamliner project. A few members quoted that more information is needed. I have provided what I have found.

The battery manufacturer is GS Yuasa Corp.

http://www.bloomberg.com/news/2013-02-05/boeing-787-dreamliner-battery-maker-gs-yuasa-s-profit-declines.html

The control system of the battery is made by Thales in France (Much closer to home)

http://www.theusdaily.com/articles/viewarticle.jsp?id=2415178-0

Case has no venting to eliminate condensation

http://www.theaustralian.com.au/travel/news/airlines-switched-many-dreamliner-batteries/story-e6frg8ro-1226565731266

NTSB investigation

http://articles.latimes.com/2013/jan/24/business/la-fi-dreamliner-ntsb-20130125

Manufacturer contact

sales@gsyuasa-lp.com

High res images, and official investigation.

http://www.ntsb.gov/investigations/2013/boeing_787/boeing_787.html

My own theory;

I have 24 years in the commercial aviation business. The rear cargo hold where the batteries are located is not pressurized nor heated, only the front hold is. This battery box looks like from all the pictures is an enclosed nonvented compartment. During the takeoff and landing cycles, condensation will form inside those compartments, and it can compound without proper ventilation over repeated take offs and landings. Aircraft specifically have drain holes in the bottom of the fuselage to compensate for this.

Compression of atmosphere. When an aircraft descends it goes from a reduced pressure to a higher pressure atmosphere in an enclosed container (the fuselage). There is also a temperature difference that enhances the difference in materials effect that some can enhance condensation while it takes longer for other materials to react. It happens frequently with older aircraft, that when an aircraft is in descent, condensation will drip onto passengers, or down the sidewalls.

Its usually a coffee stain color from mixing with repeated condensation cycles and dust.

Batteries and water (condensation don't mix)!

If you look at the pictures of the battery terminals on the pictures, those are not lead or steel contacts their threaded. Commonly aircraft materials are a magnesium aluminum mixture to save weight. If you have ever washed an old car engine, you can pour water directly onto both battery terminals and you will not cause a short, or get electrocuted.

This is because the terminals are made from lead, which has enough resistance that water will not allow a direct short. Batteries can even be submerged in water because of this and still work. This is not the case with aerospace related equipment. light weight is the overriding specification.

The terminals on the GS Yuasa batteries look like they went for light weight over lead Lead terminals can not be threaded cause lead cannot take the stress of a threaded connection. There is a plastic shield over the batteries that can trap and hold moisture because of the difference in materials and condensation buildup during descent and landing. Its not as frequent as metal, but this is probably why the shorting has not happened more frequently.

Before an aircraft ever flies, the systems get years of testing on the ground in rigs designed to simulate heavy real world usage. To my knowledge none of these rigs have ever simulated pressurization or condensation cycles with hard simulation. That has to be tested in the air. Several years ago British Airways lost a B777 on landing in Heathrow, this was also a temperature related problem, that the temperature from normal flight operations dropped to a point where the water in the fuel turned to slush. A heat exchanger, that was not properly designed held the slush forward of the heating section and too much slush flowing clogged the fuel system, starving the engines.

Trevor Langevin

Vancouver, Canada

17pv5bla-91.jpg (360kb)
2013-02-05t112709z-2-cbre9140u7x00-rtroptp-2-boeing-dreamliner-japan.JPG (31kb)
613735-dreamliner-battery.jpg (55kb)
la-blm-ntsb-boeingjpg-20130124.jpg (68kb)
5096254.jpg (94kb)
inside-no-822.jpg (46kb)
products-pic-lithium-04.jpg (6kb)
xlarge-bat-pack.jpg (94kb)

8 Attachment(s) 0 Comment(s)

KappesBuur 11 years ago

Couldn't the slipstream be used to power a dynamo which then provides the plane with enough electricity.

I remember as a kid I had a dynamo on my bicycle. Worked great. :-)

Raymond_Vermeulen 12 years ago

That method is already often used as a backup. http://en.wikipedia.org/wiki/Ram_air_turbine

1 Comment(s)

wh 11 years ago

Hi All, many thanks for your contributions so far. I am reading interesting comments on this subject. We don't have any access to circuit diagrams or more specific information on temperature conditions or vibrations. I also hear rumors the Li-On technology is just not stable enough for use in planes or to pass through all the security checks. But, let's focus on the electronics part of this - what minimal sensors would be required to make a real save system with Li-On batteries... temperature, humidity, vibrations, usage in terms of current, voltage and cycles.. we will probably end up with a complete computer system! Perhaps every Li-ON battery must be equipped with its own linux distribution to make this all run.

keep your ideas coming - appreciated

Wisse Hettinga

0 Comment(s)

drovenier 11 years ago

Hello,

as pointed out by Occam49, it is mandatory to know some more informations. I don't know if they are available, or aren't to the general public but available and can be shared, but they are very important and, among the others, there are:
1) electrical dimension of the whole pack, e.g. nominal voltage, nominal capacity;
2) cycle of usage: approximate time evolution of the standard utilization of the pack, with the surge, burst, etc... that is requested;
3) approximate time evolution of all the external ambient parameter at least of temperature, vibrations and air pressure;
4) if the pack is, or has to be, redundant at cell level;

(I did not mentioned other data I just found by myself reading the newspapers/Internet; nor other data that are directly related to others available, like from the chemistry we are able to know max/min voltages and so on).

Abovementioned parameters are needed in order to model the battery pack ad a first level.

Then, other informations are required in order to better dimension the pack and the electronics: these are mainly mechanical dimension and weight of the requested pack;

As an example of a request: battery pack, chemistry Li-Ion-Cobalt, 30 volt nominal, 50 Ah nominal;
pack typical cycle 30' @ 77°F idle (500 mA for the electronics), then 10 x 0.5 sec (separate 3 sec) 30A burst @ 77°F, then 6h @ 5°F external temperature and 2A mean drain, then 1h @ 113°F 3 A. Charge available (up to 1C, to be managed) in the third and fourth stage of utilization. Temperature cycle as stated, pressure during 1st stage, 2nd and 4th 1000mmHg, 3rd stage 300mmHg, vibrations 0-100 Hz in every stage is ___ (1st and 2st stage is low, 3rd and 4th stage most), pack has to be redundant at cell level, mechanical max dimension (mostly to undestand where to put the electronics) 1 ft x 1 ft x 10 in, weight max 30 lbs.

Numbers are just as an example.

Are these data available for the packs that had the troubles?

Thanks and best regards.

0 Comment(s)

Capper 11 years ago

All electronics in a critical safety system would have required a Failure Mode and Effects Analysis or FMEA. This document can be many hundreds of pages long and would discuss in detail all of the possible failure modes of each component of the battery and the effects of these on the system. All failure modes that could be serious would have then required failure mode testing to determine probability of detection. If the probability of detection is low, then the engineers would have designed an acceptable method of detection. The bottom line - either the particular failure mode for this system was not thought of by the design engineers, or several failures occured at the same time such that the detection mechanism was also disabled. An example would be a high-powered radio or other EMI device coupling electronic noise on critical wires monitoring the battery or controlling the charging system.

Capper 11 years ago

http://news.yahoo.com/video/boeing-says-dreamliner-fix-eliminate-125332870.html

1 Comment(s)

Occam49 11 years ago

Need specs, schematics, BOM and telemetry data from the battery units ... without this we are wasting out time here playing silly guessing games

0 Comment(s)

iwhobbit 11 years ago

Some years ago I came across an innovative battery technology from Altairnano (www.altairnano.com). They showed a demo video on their website that demonstrated their battery's iimproved immunity to catching fire. Not sure of all the details, but it might be worth contacting them.

Mike Sells

0 Comment(s)

Ivo Krieg 11 years ago

Intrinsic safety (IS) is the acseptens of
murphy's law.

for example:

spred the cells, use two makes of setups, Use also mechanik surveillance, use the same setup in cars....

And first of all be honest.

Because the setup exploded already in the factury. As reported.

Emploi the experts of Endless-sphere, Teslar, and TWIKE

http://www.treehugger.com/cars/hyper-efficient-twike-human-electric-hybrid-vehicle.html

Dont give up

Ivo Krieg 12 years ago

http://www.golem.de/news/dreamliner-tesla-chef-haelt-akkus-der-boeing-787-fuer-konstruktionsfehler-1301-97266.html You see

1 Comment(s)

Roland3 11 years ago

You can design the best battery controler of the world, it can always gets troubles. It's only electronics !. You can't change the battery caracteristics.

I propose to change the batteries for LiFe batteries which are more tolerant in case of charging errors and less temperature sensitive.

0 Comment(s)

Roland3 11 years ago

You can design the best battery controler of the world, it can always gets troubles. It's only electronics !. You can't change the battery caracteristics.

I propose to change the batteries for LiFe batteries which are more tolerant in case of charging errors and less temperature sensitive.

Ivo Krieg 12 years ago

Thats right, eeven better with ceramik seperator.

1 Comment(s)

NECV20 11 years ago

Normally safety circuits shut a battery down to protect it.

But this is very nasty in mid-air flight, to be without electrical power for steering and instruments.

So we may need to override the safety - a broken battery is less a problem then a crashed plane.

Ivo Krieg 12 years ago

Funny idea: In mid air, get the Mobiles of the passengers and link them together.

1 Comment(s)

Raymond_Vermeulen 11 years ago

I haven't read any details on the cause of those issues.

But I would say that the battery capacity of li-ion cells is very temperature dependant. Airplanes operate at temperature extremes.

So perhaps it is an issue with battery temperature control? I can imagine some sort of heater to keep the cells from cooling to much.

So here's the wiki:

http://en.wikipedia.org/wiki/Boeing_787_Dreamliner#Groundings

Floris 12 years ago

Haha, it wouldn't be very smart if they forgot put in a balancer. I know they want to save extra weight but then they will play with the most important rule on a airplane ..... Safety !!!

Ivo Krieg 12 years ago

Forgetting a balancer is a crime. Forgetting a waste heat heating is a crime of the cost controler.

NECV20 12 years ago

Boeing has not forgotten to implement a balancer, or did they?

Michael Mader 12 years ago

As far i saw from some Internet article, boeing use the Lithium Cobalt cell. You know this kind of battery from modeling airplanes. The Lithium Cobalt cell has the highest energy and lowest dimension and weight. BUT if you hurt the cell or overcharge the cobalt cell than you got a huge flame, when the cell is fully charged. So also with an bms that works fine , this cell can got into a fireball. With other Lithium cell like mangan oder fepo4 you can avoid this problem( is not 100% true but this kind of lithium cell are more saftier) But mangan or fepo4 cell have less energy, more weight and more volume. thats my point of view

Floris 12 years ago

Li-ion cells are a good choice but at these extreme conditions in a wide range of temperature and heavy vibrations they should be monitored at temperature and voltage per cell continuously. Even if they are high industrial quality batteries, they can be or become defective . If they are charging, there should be at least a report of the amount of charge that has been charged in every singel cel, so you can see loose of capacity very quick and keep them safe from overcharging. The risk of overcharging without knowing is als higher if cells are connected in parallel.