Go Back  PPRuNe Forums > Flight Deck Forums > Tech Log
Reload this Page >

787 Batteries and Chargers - Part 1

Tech Log The very best in practical technical discussion on the web

787 Batteries and Chargers - Part 1

Old 16th Mar 2013, 05:47
  #1001 (permalink)  
Join Date: Aug 2011
Location: hong kong
Age: 60
Posts: 93

I second your post in its entirety.

Talking is just so much easier than understanding.

I read sites like PPRuNe and avherald pretty much since their beginnings. Read, to learn. This is the first thread i contributed to. It is almost regrettable. The amount of noise is astounding and it seems to increase.

The industry has seen its share of design induced failures and learned from it. Some examples with fatal outcomes are: Comet square window fatigue failures, Electra whirl mode wing failures, 737 rudder pcu reversal.

In this case, no one got hurt. Plenty of red faces though, this is good. I once had an engineer fly in to find my design error in 5 minutes. I shall never forget.

Technology gets ever more complex. it takes time to learn and apply.

My background is electro mechanical engineering, I went into micros age 17, assembling code by hand. I do mostly systems software (C/C++/JS) these days.

I am amazed how far young developers are removed from "deep" understanding of systems.

Things get abstracted. A does A, B does B, C does C.... Interfaces A<>B B<>C are defined but nobody will understand all permutations of interactions of A, B, C, especially when something breaks.

As we read often here, young pilots face similar problems.

It takes a positive mindset and do the best and learn and try again.

Back to topic.

Boeing needs to get the planes back into the air.

Their proposal is better management of the cells, more insulation and an additional layer, the firebox. IMHO, it will be sufficient to address the issues at hand, and mitigates further failures.

In short, the approved solution will be safe by the letter of the law.

I may wish to do more, for example it seems that the charger charged the failing battery for 3-4 seconds. I would like to avoid that. Anyway, improvement is incremental, it never ends.

The entire industry will benefit from this incident, as it did from incidents in the past.

Last edited by saptzae; 16th Mar 2013 at 05:52. Reason: Remove ref to 747 CWT
saptzae is offline  
Old 16th Mar 2013, 05:55
  #1002 (permalink)  
Join Date: Feb 2013
Location: Placerville, CA
Posts: 71
Looks like that link is dead at the moment. (Died of embarrassment?) I did see what I think was that image from an earlier published article though. The main change was in the descriptive callouts. And the fireproof top layer. Not sure if anything else changed in the image.
Maybe it was just intended to show where the changes would be made and not what they were.

Last edited by inetdog; 16th Mar 2013 at 05:57.
inetdog is offline  
Old 16th Mar 2013, 15:19
  #1003 (permalink)  
Join Date: Aug 2003
Location: FR
Posts: 222
The new battery enclosure will still be pressurized?
pax2908 is offline  
Old 16th Mar 2013, 15:20
  #1004 (permalink)  
Join Date: Feb 2013
Location: Seattle, WA
Posts: 78
I wonder how long the airlines will be able to stand what was already an apparently abnormally high battery replacement rate prior to the fire/overheats.
The cause of those replacements was predominately due to the battery being too deeply discharged by ground staff using it during operations like refueling and tripping safety systems.

Boeing is raising that limit so I guess that means when the battery cuts-off, it will be at a charge level that will allow the onboard charging system to bring it back up and it will no longer need to be removed.

Here is a YouTube Clip of part of the Boeing presentation on the changes they are making to the battery system -
Kiskaloo is offline  
Old 16th Mar 2013, 21:26
  #1005 (permalink)  
Join Date: Jan 2008
Location: lancs.UK
Age: 74
Posts: 1,191
full marks to the commentator who got through the whole of his spiel without once sniggering. Me- i laughed aloud!

Of course, it's Youtube, therefore, it's designed to appeal to the masses of mushrooms out there.

Are the rest of you convinced?

Forgot, the overflow pipe is Titanium, so that's OK then (looks like domestic plumbing to me!)
Per my remark on the R&N thread....poke the pipe out of the tailcone and the burning, toxic crap will eject into the jetstream......Hell, that commentator could keep a straight face and pass it off as automatic thrust-jet augmentation to get the aircraft back to land quicker.

See! Mr. Boeing! I can bullshit with the rest of them....got a cushy overpaid job for me?
cockney steve is offline  
Old 16th Mar 2013, 21:55
  #1006 (permalink)  
Join Date: Feb 2013
Location: Seattle, WA
Posts: 78
Hot, liquid electrolyte exited the forward EE bay of NH692 via the outflow valve and it did not damage the fuselage beyond leaving a brown streak on top of the white paint.
Kiskaloo is offline  
Old 16th Mar 2013, 22:06
  #1007 (permalink)  
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,123
The only closeup photo I saw of the forward scupper showed the aperture completely covered over with film, and tape.
Lyman is offline  
Old 16th Mar 2013, 23:28
  #1008 (permalink)  
Join Date: Aug 2001
Location: Ft. Collins, Colorado USA
Age: 87
Posts: 188
Thanks very much for you update and the Boeing video. I am persuaded that they have indeed produced a container sufficiently strong and heat resistant to hold the worst case failure. A steel plate box 0.250" thick does raise my eyebrows as to weight.

Speaking as a maintenance person of long experience, changing the battery has now gone from releasing a couple of hold-downs and a screw-on battery connector to removing and reinstalling a considerable number of fasteners around the periphery of the cover and maintaining a large seal as well. Will it require some pressure check after reinstalling?

I surely hope that the change in the on-board recharging spec by Boeing cuts down the number of battery changes required and that most of the previous changes were for that reason alone. The fact that the battery seemed to be easily depleted below that value used before by the simple use of it for ground servicing seems strange.

I salute Boeing's tenacity in trampling down the perceived problem and I hope the fixes put the whole thing to bed.
tonytales is offline  
Old 17th Mar 2013, 04:28
  #1009 (permalink)  
Join Date: Jul 2002
Location: 40N, 80W
Posts: 233
What Boeing wanted vs. what Boeing got.

I have been reading parts of the FAA docket at:
Accident Investigations - NTSB - National Transportation Safety Board.

Old stuff, but interesting.

The battery Boeing wanted.
Page 15 of the above docket refers to a Boeing Specification Document” (SCD) which is used to describe an individual airplane component.

The SCD referred to states that “The [APU] battery shall be capable of supplying 18kW constant power for 45 seconds to the Start Power Unit (SPU) with a minimum battery terminal voltage of 20V for three consecutive APU start attempts with 60 seconds rest between each start attempt”.

18kW at 32Vdc needs 562A, and at 20 Vdc needs 900A.

So Boeing was expecting the APU battery to supply between 562A and 900A for 45 seconds at a time.

In contrast,
The battery Yuasa supplied.
The 787 batteries are made up of several Yuasa LPV65 batteries in series. The same current passes through all the batteries.

These batteries are 75Ah types with a Yuasa specified maximum discharge rate of 5C, or 5x75 = 375A.

So Yuasa was not expecting to provide more than 375A.

How much current does an APU on a 787 actually take to start?
The FAA noted this disparity (562A or more called for by Boeing vs. 375A maximum available from the Yuasa batteries).

They then noted that the APU on the incident aircraft actually used less power than the Boeing SCD called for, and that there were sudden changes in power consumption during the start sequence, but that testing was done with “resistive banks”.

The FAA docket says that a start on the APU on the incident aircraft took 35 seconds and actually used power as follows:

Initially about 9.6kW (32V at 300A)
Then 8.7kW
Then 10kW for 10 seconds
Then 4kW
10.3kW for about 20 seconds
Then zero for 47 seconds
Then charging at 46A.

If the voltage available is 31V or less, then the maximum APU start power of 10.3kW requires at least 332A and probably more.

It would be interesting to know what factor limits the Yuasa battery current to 375A?
And is this limit independent of temperature, and of the duration of the current?

BTW, is there any information about how many of the 150 replaced batteries were APU batteries and how many were main batteries?

Last edited by PickyPerkins; 26th Mar 2013 at 15:56. Reason: Add bold and "took 35 seconds".
PickyPerkins is offline  
Old 17th Mar 2013, 05:52
  #1010 (permalink)  
Join Date: Feb 2013
Location: Placerville, CA
Posts: 71
If the voltage available is 31V or less, then the maximum APU start power of 10.3kW requires at least 332A and probably more.
This, at least for the APU start battery, indicates that some proposals here about using a battery design that would allow dropping a single failed cell out of the series string, thus allowing continued use of the battery if necessary, might not be feasible.

Last edited by inetdog; 17th Mar 2013 at 05:53.
inetdog is offline  
Old 17th Mar 2013, 11:55
  #1011 (permalink)  
Join Date: Dec 2001
Location: England
Posts: 1,388
So they propose adding insulators and tape and wrapping the whole thing in an outer steel box... Is there any additional cooling to compensate in hot climates or isn't that needed?
cwatters is offline  
Old 17th Mar 2013, 15:46
  #1012 (permalink)  
Join Date: Jun 2009
Location: Oxford, England
Posts: 297
PickyPerkins, #1009:

That's an interesting post, thanks, though are there typos in the battery
load timeline ?:
Initially about 9.6kW (32V at 300A)
Then 8.7kW
Then 10kW for 10 seconds
Then 4kW
10.3kW for about 20 seconds
Then zero for 47 seconds
Then charging at 46A.
If you have an accurate timeline, then you can calculate the total
watts hours needed to start the apu. It's quite easy to have profiled
current limiting in the apu starter electronics, since it's inverter
driven and start time trades off against max starter current. The
20v minimum seems a bit iffy, in that each cell's voltage will be 2.5v,
at that point. So, what will the cell voltages be when a start has
completed from 20v ?. Less than a min safe voltage, or what ?. The final
point being that the cell data sheet only lists discharge characteristics
up to 250 amps, with nothing after that, yet they are driving these cells
at 300 amps ?. My take on it is that they skimped on the cell spec and
should have used cells of much higher ampere hour capacity for this
application. The weight saving obsession must have it's limits at some

Aviation kit is usually designed to have a higher margin of safety than it's
commercial equivalent. I would have thought that they would at least have
doubled the cell capacity in relation to max expected load. To do this would
mean larger cells, but perhaps longer service life and less prone to
catastrophic failure. Running components well within ratings improves
reliability, all else being equal, but these batteries are being pushed hard.
High current to start the apu, then fast charging within short timescales,
both of which will cause significant temperature rise and perhaps hot spots
due to varying resistive paths within the cell.

Let's hope they get it right, but the proposed solution seems light on
detail in terms of what changes will be made to battery management and
charging. Afaics, there is still only a single current sensor. There's
no mention of improved temp sensing, non volatile logging for the cell
voltages, contactor fail safe or other items, though some of this was
hinted at online a couple of weeks ago.

As for the strengthened enclosure, reminds me of the old black joke about
fusing: If the fuse blows, put in thicker wire and if that also blows, put
in a paperclip ...
syseng68k is offline  
Old 17th Mar 2013, 16:38
  #1013 (permalink)  
Join Date: Jul 2002
Location: 40N, 80W
Posts: 233
Syseng68k Post #1012

Total energy for APU starts does not seem to be the problem, because even the Boeing 18kW spec. only needs about 30% of the energy stored an a fully charged battery (though not all that energy is available energy).

Boeing requirement was 18kW x 45 seconds x 3 = 0.675 kWh
Battery stored energy is 75Ah x 32V = 2.4kWh.

Current is the problem.

I should have mentioned that the FAA docket said the APU took only 35 seconds to start, which is less than the 45 seconds in the Boeing SCD.

Just to be sure that I understand your post, are you suggesting that the inverter may have been limited to taking something like 375A max. and was able to start the APU in 35 seconds with this limitation? (I know nothing about APUs.)
PickyPerkins is offline  
Old 17th Mar 2013, 18:12
  #1014 (permalink)  
Join Date: Jul 2002
Location: 40N, 80W
Posts: 233
A detail in the description of the new Boeing battery box

If you have a partially full barrel of oil and keep it though several summer/winter temperature/humidity cycles you will usually find a layer of water at the bottom. During the summer humid air enters the air space above the oil and in the winter this moisture condenses and descends to the bottom of the barrel, where it remains trapped.

Now, bear with me, because there is a point to all this.

I have had a similar experience with a steel pipe with a plate welded to the bottom used to support a girder over my garage. Over a period of 30 years condensation within the pipe rusted the inside of the bottom of the pipe to the point where it fell over onto the floor!

This accumulation of water may happen in any nearly sealed enclosure subject to temperature/humidity cycles where the condensate cannot drain away.

I was therefore very interested to see a detail of the new Boeing battery box, as follows, “Finally, a set of changes is being made to the battery case that contains the battery cells and the battery management unit. Small holes at the bottom will allow moisture to drain away from the battery and larger holes on the sides will allow a failed battery to vent with less impact to other parts of the battery.”
Boeing Provides Details on 787 Battery Improvements - Mar 14, 2013

So maybe they think condensation had a role in the failures, and now there is to be no attempt to seal the battery box. Condensation will be allowed to drain out of the box. The stainless steel enclosure will be vented to the outside but otherwise sealed, and I suppose that enclosure will accumulate water just like my barrel of oil unless some provision is made for a drain. But at least the condensate will be in the enclosure and not in the battery box.

The Boston event occurred on Jan. 7, and I seem recall that the aircraft arrived from Narita, Japan, in the early morning when the temperature was around freezing. I Googled for “What is the average temperature in January in Narita Japan?”, and got a web-page which said, among other things, “Narita has a warm humid temperate climate with hot summers and no dry season.“ and "The month of January is characterized by gradually falling daily high temperatures (to about 30°F), with daily highs around 48°F throughout the month, exceeding 56°F or dropping below 41°F only one day in ten".

So was this aircraft regularly going to and fro from an area “with no dry season” to one with freezing conditions?

Last edited by PickyPerkins; 17th Mar 2013 at 18:22.
PickyPerkins is offline  
Old 17th Mar 2013, 19:22
  #1015 (permalink)  
Join Date: Jun 2009
Location: Oxford, England
Posts: 297
Current is the problem.
Exactly, but yet another to add to the list . Looking at the figures,
a cell with greater ampere hour capacity might have been a better
choice. I wonder how many discharge / charge cycles these batteries
would experience per week and in particular, how many cycles for the
replaced batteries and those that failed ?. Fast discharge / charge
cycles do have an affect on cell aging and overall lifetime.

Designers must always interprete data sheet values. Not everything is
specified in the data sheet, but if
the cell data only graphs discharge curves up to 250 amps, then they
are implying that that rate is the maximum they have envisaged for
the component, even if the absolute limits are higher. Any half way
decent design would work well within those limits, but it seems not
in this case, or am I missing something ?.
Just to be sure that I understand your post, are you suggesting that
the inverter may have been limited to taking something like 375A max.
and was able to start the APU in 35 seconds with this limitation?
(I know nothing about APUs.)
An apu is just an engine driven generator, with starter motor. In
this case, the starter motor is dual purpose and acts as a generator
once the engine is started. Iirc, it's a 3 phase ac machine and starting
works by applying ac to the windings. In starter mode, the AC supply
comes from the battery via a dc to ac inverter, which is software
controlled to provide a current limited ramp up of power to the
motor. I dont have the figures, but the inverter dc input current will
be limited to a safe value to protect the motor, inverter and battery
from overload...
syseng68k is offline  
Old 17th Mar 2013, 20:48
  #1016 (permalink)  
More bang for your buck
Join Date: Nov 2005
Location: land of the clanger
Age: 79
Posts: 3,513
I wonder how these people are getting on with their batteries:

The World's Largest Lithium-Ion Battery Farm Comes Online - Forbes
green granite is offline  
Old 17th Mar 2013, 21:05
  #1017 (permalink)  
Join Date: Apr 2005
Location: UK
Posts: 459
Just viewed the Boeing fix, very nice.

I know they are trying to cover all the bases and did a good job I think.

Two big points from my take on it.

1. Batts are not reqd during flight.

2. Safety box to be fitted anyways.

I think there are well covering the future when any batt fails that it's no problem at all. Apart from batt not avail for use.

I think they are doing a god job indeed.

I hope no 787 batt ever smokes again, but if they have a high failure rate, regulators will insist on more changes/ETOPS problems etc etc.

Batt or any problems with the 787 will make too much press over the next few years, that will have its own problems to be dealt with.

If they have any smoking batts in the box events, it will be interesting to see how the press is handled.

I guess they have learned a lot, toned down batts and made good changes to the overall system to keep it chilled and for it to be isolated when the computer says "not normal"

So two big items remain.

1. To get the batt fix certified.

2. Checking other systems on the 787 were certified using a correct process.

I wish the 787 good fortune...
Joetom is offline  
Old 17th Mar 2013, 23:05
  #1018 (permalink)  
Join Date: Feb 2013
Location: Placerville, CA
Posts: 71
Heat and smoke are the first signs of a thermal runaway, once one cell starts to overheat, if the battery is not cooled by quenching the other cells will quickly follow suit. The bigger the battery the more cells.
Remember, to avoid confusion and unnecessarily scaring the public, when discussing situations like yours, we must qualify the words "thermal runaway" with either "cell" or "battery" because Boeing has chosen to use "thermal runaway" ONLY to refer to "aircraft thermal runaway" in which the end result is destroying the plane.
When they say that thermal runaway can only result from overcharging, they are using their definition of thermal runaway.
So the two accidents involving 747 cargo craft were probably what Boeing would call thermal runaway, (although as I understand it that has not been proven.) The two 787 incidents were not.

Last edited by inetdog; 17th Mar 2013 at 23:08.
inetdog is offline  
Old 17th Mar 2013, 23:33
  #1019 (permalink)  
Join Date: Jun 2009
Location: Oxford, England
Posts: 297
we must qualify the words "thermal runaway"
Absolutely. Sounds like Orwell's Newspeak, where some words vanish from the
dictionary and others take on entirely new meanings .

If those batteries didn't suffer from thermal runaway, then what was it ?...
syseng68k is offline  
Old 18th Mar 2013, 00:14
  #1020 (permalink)  
Join Date: Feb 2013
Location: Placerville, CA
Posts: 71
If those batteries didn't suffer from thermal runaway, then what was it ?.
It was two sequential occurrences:
1. "Cell thermal runaway" in one cell, followed by
2. "Battery thermal runaway" as it spread to other cells.
Fortunately "aircraft thermal runaway" (otherwise to be known only as "thermal runaway") did not occur as a result.

PS: The above is what I feel to be an accurate paraphrase of the descriptions given during the 1 1/2 hour Boeing webcast from Japan.

You are clearly an ungood, or maybe even plus-ungood Newspeaker.

Last edited by inetdog; 18th Mar 2013 at 01:04.
inetdog is offline  

Thread Tools
Search this Thread

Contact Us - Archive - Advertising - Cookie Policy - Privacy Statement - Terms of Service - Do Not Sell My Personal Information -

Copyright © 2021 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.