Boeing Lithium Battery Problem

[quantumtangles]

Materials scientists have said adding cooling fans to the battery layers (each being about the size of a laptop - stacked one on top of the other) would help, as the internal 'laptops' cannot 'heat-sink' the same way battery layers near the outside of the unit.

There is also talk of thermal runaway...hitting a critical temperature in electrochemical terms, causing a change of state (to gas) putting pressure on the unit case, breaking seals and leading to thermal runaway.

But these materials science ideas do not make that much sense to me, not least because they are too obvious. Boeing engineers are really first rate. Amongst the very best engineers in the world, so I suspect the problem is more exotic.

Based on insufficient data, my first wild guess was a problem in the manufacture of the circuit board or a mismatch by a supplier in terms of the voltage regulators. But again, these engineers are too good to make such mistakes ...parts manufacturers may not have followed Boeing spec sheets of course.

But a reported fact that really caught my attention (if indeed it is a fact) is that one of the dysfunctional batteries overheated when neither charging nor discharging. This is a clue if ever there was one, but what does it mean?

Before learning of this, I thought maybe the circuit board clearances were not wide enough or there was some sort of malfunction or mismatch of the voltage regulator ICs. A supplier might have made a mistake.

But if the malfunction happened when the battery was neither charging nor discharging...wtf?

When you exclude the impossible, what you are left with, no matter how improbable, is the truth. On that basis, might other electromechanical components on the aircraft have caused inductive spikes (due to collapsing magnetic fields) in proximity to the battery circuit boards?

Circuit board clearances on the battery circuits containing the voltage regulators and short circuit protection components may have been fine, ceteris paribus...but in the presence of inductive spiking from proximate devices, could they have overloaded the battery regulation circuits, caused a polarity short and led to thermal runaway?

Interested in your views. So much want to see this miracle of engineering fixed and ready to go. They take safety so seriously I am absolutely confident they will fix it perfectly.

Edit: Inductive spike exceeds datasheet spec of voltage regulator integrated circuit?

 

[BobK]

I'm still betting on a software glitch, but then that is because I am a software engineer.

Edit: Software is really good at producing irreproducible errors that rarely occur.

Bob

 

[quantumtangles]

I'm still betting on a software glitch, but then that is because I am a software engineer.

Edit: Software is really good at producing irreproducible errors that rarely occur.

Bob

I like it

 

[Raven Luni]

Theyre probably running windoze or some ****

 

[DieselTwitch]

Are the batteries Li or LiFePO4's?

 

[(*steve*)]

Are the batteries Li or LiFePO4's?

My information is that that are one of the higher density types. So yes, one of the less stable variants.

 

[DieselTwitch]

So maybe LiNiMnCo or LiCoO2? I'm a little shocked they would run such an unstable battery in an airframe. I worked on Ah-64 and we have been using LiFePO4 for years at least 1995

 

[(*steve*)]

Cobalt rings a bell for the 787. They're in the 380 as well.

However, there's a difference in scale.

 

[sirch]

Apparently not the battery itself -

http://www.bbc.co.uk/news/business-21230940

 

[(*steve*)]

No fault with the battery (other than it is now a charred pile of scrap).

Interestingly, the finger was pointed at the charger. Then someone claimed it wasn't that. Now they are pointing to that again.

Anyone else think they don't really know yet?

 

[shrtrnd]

Last I heard, they were investigating the MONITORING SYSTEM manufacturer.
The guys who made the system that supposed to monitor the battery status.
Makes me wonder if it's the batteries, or the support system for them.

 

[BobK]

Last I heard, they were investigating the MONITORING SYSTEM manufacturer.
The guys who made the system that supposed to monitor the battery status.
Makes me wonder if it's the batteries, or the support system for them.

My guess of software problem is looking better.

Bob

 

[BobK]

New article with a lot more info:

http://www.nytimes.com/2013/01/30/business/boeing-aware-of-battery-ills-before-the-fires.html?hp

The failures were not that uncommon:

Officials at All Nippon Airways, the jets’ biggest operator, said in an interview on Tuesday that it replaced 10 of the batteries in the months before fire in one plane and smoke in another led regulators around the world to ground the jets.

And appear to be caused by sudden over-discharge:

In five of the 10 replacements, All Nippon said that the main battery had showed an unexpectedly low charge. An unexpected drop in a 787’s main battery also occurred on the All Nippon flight that had to make an emergency landing in Japan on Jan. 16.

According to investigators in Japan, the battery on the jet that made the emergency landing showed a stable reading of 31 volts, near its full charge capacity, until 15 minutes into the flight when pilots detected a strange smell. About that time, sensors detected a sudden unstable discharge of the battery to near zero for reasons that Japanese investigators still cannot explain.

But that makes little sense, where did all that power go without blowing something else up?

Edited to add:

But you have to admire the depth of analysis from Boeing:

Boeing officials said that improperly connecting a battery can also render it unusable.

Bob

 

[PTaylor]

Probably the wrong root cause, but think outside the box for a second. I've been curious if altitude/pressure is contributing to the problem chemically.

Anything else, say software or hardware, that should be reproducible on the ground in a lab.

I've been thinking about it too, and I keep going back to the batteries that were in Laptops and phones a few years ago that was catching fire and exploding. I don't know if they were the same type or constructed the same way, or with similar materials.

I think in all cases it was ruled as a defect, how did they come to that conclusion?

 

[quantumtangles]

Probably the wrong root cause, but think outside the box for a second. I've been curious if altitude/pressure is contributing to the problem chemically.

Anything else, say software or hardware, that should be reproducible on the ground in a lab.

I've been thinking about it too, and I keep going back to the batteries that were in Laptops and phones a few years ago that was catching fire and exploding. I don't know if they were the same type or constructed the same way, or with similar materials.

I think in all cases it was ruled as a defect, how did they come to that conclusion?

The effects of pressure variations must have been tested. The FAA would have insisted upon it even if the engineers themselves did not (beyond a peradventure the engineers also insisted on testing this).

More importantly, pressure decreases as altitude increases. The pressure at the base of an olympic sized swimming pool full of seawater will be low if the pool is only a couple of centimeters deep.

In contrast, the pressure at the base of a tiny pool that is 25m deep will be very considerably greater. Pressure (Pascals) = height (m) x rho (say 1020kg/m3) x gravity (9.81 m/s/s)

Specifically, height (25m) x density of seawater (1020kg/m3) x gravity (9.81 m/s/s)
= 250,155 Pa gauge pressure at the base of the pool

Adding atmospheric pressure of 101,350 Pa would give absolute pressure of 351505 Pa at the base.

But as an airplane rises (air is also a fluid), pressure decreases.and decreasing pressure is associated with cooling, not heating.

Only by increasing pressure (illustrated by clamping the output nozzle of a bicycle pump with your finger before operating it) does the temperature of fluids increase (assuming other sources of heat are not applied).

You suggest pressure differentials in flight affected the electro-chemistry of the batteries, but is there any authority for the proposition that decreasing pressure...or minor but continuous fluctuations in pressure during flight...affect electrochemistry? I do not know the answer.

Can you provide an example of a situation where changing pressure...and in particular decreasing pressure ...has been found to affect electro-chemistry?

If I am wrong in thinking inductive spikes from collapsing magnetic fields in proximate electro-mechanical devices exceeded the Vmax of the voltage regulator IC, or exceeded the Imax of circuit board clearances, you may be correct in suggesting (by necessary implication) that shaped and possibly sealed lithium packets in the battery leaked due to changes in pressure. That contracting and expanding pockets of air in the battery compartments burst the battery skin due to pressure changes, causing leakage.

Consider also the possibility that shaped lithium batteries (preferred because they can be molded to shape ...fitted into otherwise useless crevices) ended up having sharply pointed features on otherwise smooth Gaussian surfaces, leading to a build up of charge on the sharply pointed features. Is this a Gaussian surface issue?

Finally, could this be a Kelvin Battery effect. Interconnected lithium batteries may be pressed against the metal fuselage. Fast flowing ionized air may somehow have created a superabundance of negative or positive ions, per Kelvin Battery though without there being a need for ionized water. In such circumstances, could an unintended spark gap have caused meltdown of the Voltage Regulator IC or Imax of the circuit board clearances? Did the aircraft that exhibited problems fly through rain or dense cloud such that ionization of water may have created a Kelvin battery effect as between two batteries on either side of the fuselage?

 

[(*steve*)]

I am pretty sure you'll find that the batteries are held within the pressurised compartment that is the fuselage of the aircraft.

The pressure changes would exist, but outside of very unusual situations would not have a pressure altitude above 8000 feet.

If it were outside the pressure vessel, smoke would not have been smelt from inside the cabin.