You're playing fast and loose with the facts again - the root cause of the BA038 rollback problem was discovered and proved by the AAIB using a test harness.

The presence of manufacturer representatives on investigating teams is nothing new, and they're usually welcomed because they know the systems and airframe better than anyone else. As yet, no manufacturer has directly interfered with the investigatory process - because the risks of doing so considerably outweigh any potential benefit.

You are too quick to judge.

I make no conclusion, only that Boeing plays the primary role in certifying, and a de facto lead in investigation. The two become inextricably interwined when one or the other merely exists....

You immediately jump to your patent obsessions with authority, control, and sensitivity to views other than your own....

Boeing played a huge role in 038, designing and building very expensive ground test formats. The conclusion was Ice in fuel, due improper design of FOHE, a fix was a trimback to eliminate the Fuel inlets extra length....

Although Boeings testing failed to reproduce what was believed to be the actual problem.

Otherwise, nice to see ya again...:ok:

Wouldnt they be able to test the APU system on the ground?

To me it would be prudent to test the hell out of that system on the ground, show its operational and safe, before even asking to go airbourne...

I can't imagine they haven't been doing that for the last three weeks. Also, it's not just the APU system, but the main aircraft battery. It can fly without an APU battery for 10 days, limited to 180 mins ETOPs.

Yesterday, I read a big section i'd missed on the Tech.forum thread/

THE CELLS ARE EACH COMPOSED OF SEVERAL (7?) SMALL CELLS IN PARALLELL

In essence, this works like your car-battery, several plates alternately stacked pos-neg-pos-neg to get the surface-area for the capacity and discharge-rate required.....all the plates of each polarity are linked together, so there's only 1 Pos and one Neg connection.

The problem with these Lithium Cells, is the charger ONLY MONITORS THE CELL_GROUP AS A WHOLE.
I don't pretend to know how sensitive the monitoring-system is, but now believe that the internal configuration and control-boards of the Thales "power storage box" needs a serious looking at......OOPS They're going to do that.

The charging-unit MUST be able to see yhe individual sub-cells in a stack which form one of the 8 cells in the battery.....once a small sub-cell runs -away, it would appear a domino-effect teces place and the thermal-runaway cascades through the entire battery.

Poor battery design, for the application and poor charger design which fails to adequately monitor and counter the inadequacies in the battery-pack it's supposed to control.......
all just my opinion ,of course.

OK...I'm very interested in that idea...would it be practical to design an 8-cell battery (with the same energy capability), that comprises 8 separate single cell batteries that are all charged/controlled by separate chargers.
That would seem to get around the problem of charge/discharge of each single cell, and limit the effects of a thermal runaway.

@cockney steve

Yes, the eight "cells" of the 787 battery contain each 6 sub-cells in parallel.
Paralleling increases capacity. So each sub-cell has about 11Ah makes roughly the nominated 65Ah. These eight cell packs in serial a 3,7 V give you 29.6 V/65Ah.

Concerning the comparison with a car-battery, the 6 cells in serial of 2V each (lead-acid) give you 12V, but serial arrangement doesn't increase the capacity, just voltage.

Otherwise I fully agree with your post:ok:

This "cells in parallel" fixation is a red herring. There is no substantiative difference between having a two-metre long electrode folded up in one cell and two one-metre long electrodes in two cells connected in parallel: the electrical conditions on each square centimeter (or whatever) of electrode are identical.

There may be some slight differences due to the separate electrolyte, but the electrolyte in these batteries is largely immobilised in the porous separator anyway.

Sorry, I can't seem to find it anywhere.

Is this "6 subcells per cell" claim substantiated anywhere?

I'd be grateful for a pointer to an authoritative source.

I remember seeing an exploded drawing (can't find that either) actually showing a "folded" cell, and even GS Yuasa makes no such claim in their "specifications".

http://www.mlit.go.jp/jtsb/flash/JA8...116-130205.pdf

Page 5?

have a read of:

Lithium Battery Manufacturing

and:

Materials and Processing for lithium-ion Batteries

hetfield,

thanks for the link. But it does not say anywhere that there are subcells. In fact, when I look at the lab table photos from the NTSB press releases, it appears there are 6 very long strips, indicating folded cells (now unfurled), and two shorter ones, which I take to be cells that were so severly damaged that there was no continuous strip any more.

So, this sub-cells hypothesis is not substantiated anywhere, but just a wild guess that somehow stuck in these forums.

fizz57

Bingo. The "Case" contains eight batteries. Each battery has three "cells". Each "cell" is one electrode, ten metres long, 150 millimeters wide, wound (folded) to fit in the prismatic container. Each electrode has 35k square centimeters of surface area. Each time a pin sized aperture in the polypropylene separator melts or is occluded, efficiency drops, heat is created, and the battery loses efficiency.

The "cell" in its multiple "folds" presents as a brain, or kidney would, layer upon layer of organic material.

Hence the CT scan. Not that it would help an in service Cell.

Once again, I repeat. A circular, cylindrical wind is preferred, "folding" is problematic for the very thin separator.

http://www.nfpa.org/assets/files/pdf...rieshazard.pdf

It's in there. I posted the pertinent part from the website in TechLog. #525

fizz57 et al:

When using Li-Ion (as well LiPo) cells it is important to have each cell voltage charged (or at least balanced) separately.
This means that when nominal cell voltage is 3.7V and you have e.g. 3 cells in SERIES to get a 11.1V potential, it won't do the job to charge them with a single 12.9V supply UNLESS each potential is balanced (i.e. 3 x 4.3V charge voltage applied to each single cell level).
If cells are in parallel, then it is of utmost importance that both cells have the same charge/discharge curve, otherwise current balancing is not possible between the cells, which may lead to under- or overvoltage discharge or overcurrent draw of single cells.
Thats at least how I learned it when designing large-ish power applications for Li-Ion batteries. And they still work fine, no issues, no interruptions, no anything.

Parallel Cells
 

Not as big an issue as series batteries. Theoretically, you can view one cell as being 10, one meter long 'batteries' connected in parallel. Or 100, 10cm batteries, etc.

Discharge of paralleled batteries tends to be self balancing. As one battery provides more than its share of charge to the load, it's charge is reduced below that of is neighbors. And as its charge is reduced, its voltage drops and internal resistance rises, reducing its further contribution to the load until the others 'catch up'.

The problem arises when cell (or a section thereof) voltage rises with temperature faster than its discharge lowers it. That provides a positive feedback, causing that cell to discharge even faster, generating more heat. Now you have that thermal runaway.

Putting three paralleled cells in one can reduces the thermal gradient between them and reduces this effect. But the effect would still be present with only one cell per battery, given that the outer layer would lose heat to the container faster than those toward tthe center.

The photographs and notes posted by others suggest the configuration is probably 6P8S.

In other words there are 48 cells. Six of these are wired together in parallel to make sub-batteries (increasing the capacity), then 8 sub-batteries are wired in series (increasing the voltage) to make the main battery.

The 8 sub batteries are balanced but the 6 cells in parallel in each don't appear to be. However that's not usually a problem because cells in parallel will tend to auto balance...

Cell Balancing and Battery Equalisation

Update from the NTSB
 

http://www.ntsb.gov/investigations/2...787_2-7-13.pdf

Quick solution does not appear to be just around the corner ...

llagonne:

Thanks for that. Some nice new pics, but no real substance otherwise ?...

Correction to NTSB images.

The cells represented as "LEFT SIDE" and "RIGHT SIDE" are reversed.

Unless you are looking at the BATTERY from BEHIND.

"INSIDE THE BATTERY" page 8.....

syseng68k
 

The real substance is that they are still trying to understand what when wrong at every possible level (design, manufacturing, certification, etc.).

If you read between the lines, NTSB appears to take a tough stance and FAA looks on the defensive as they have certified the A/C while apparently overlooking something ...