787 Batteries and Chargers - Part 1
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes
on
0 Posts
HazelNuts39
Yes, there are wires in the space between cell #2 and cell#3. Also, it appears some space for separation was allotted for isolation, at least between the above two cells. The Heat Damage seems worst at areas near metallic components. The case survived; had it been bigger, would there have been less damage to the nearby components?
Your numbering scheme is appropriate, but NTSB still got the "sides" wrong. The front of the Battery should determine the relative orientation, imo.
Question? Why such relatively massive connector bars? Why such close proximity to the Stainless Case?
Why is the Battery case placed so near the deck? Why not a stand off base, to prevent the kind of heat damage to the floor we see in the NTSB photo of the technician in the E/E bay?
I think the elements of compliant design are all there, but the dimensions, volumes are cut too thin.... There was ample room to accomodate a stand off base and more room between Stainless case and cells.... Not to mention the proximity to the bulkhead that supports the power panels.
Hind sight, eh?
Yes, there are wires in the space between cell #2 and cell#3. Also, it appears some space for separation was allotted for isolation, at least between the above two cells. The Heat Damage seems worst at areas near metallic components. The case survived; had it been bigger, would there have been less damage to the nearby components?
Your numbering scheme is appropriate, but NTSB still got the "sides" wrong. The front of the Battery should determine the relative orientation, imo.
Question? Why such relatively massive connector bars? Why such close proximity to the Stainless Case?
Why is the Battery case placed so near the deck? Why not a stand off base, to prevent the kind of heat damage to the floor we see in the NTSB photo of the technician in the E/E bay?
I think the elements of compliant design are all there, but the dimensions, volumes are cut too thin.... There was ample room to accomodate a stand off base and more room between Stainless case and cells.... Not to mention the proximity to the bulkhead that supports the power panels.
Hind sight, eh?
Last edited by Lyman; 12th Feb 2013 at 12:06.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Each cell Temp monitoring (for SAFETY) is not essential for "management"
Hi,
HazelNuts39
I donīt believe the designers put independent temp sensors (8)
The connector, (well observed
) seems confirm it (with fewer wires).
Apparently they just relied on the "patented algorithms"
Another possible reason to not use CELL SEPARATION. A BIG ERROR!
itīs clear to me the designers (including Boeing) were overconfident in the State of the art cells. (no heat, low internal resistance, long MTBF, etc.)
HazelNuts39
A possible location for one or more temperature sensors...
I donīt believe the designers put independent temp sensors (8)
The connector, (well observed
) seems confirm it (with fewer wires). Apparently they just relied on the "patented algorithms"
Another possible reason to not use CELL SEPARATION. A BIG ERROR!
itīs clear to me the designers (including Boeing) were overconfident in the State of the art cells. (no heat, low internal resistance, long MTBF, etc.)
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
This will be very difficult to verify
Hi,
mm43 @ # 635
it seems, cell voltage history (JAL and ANA) were lost
PCBīs certainly are FUBAR and only option is what NTSB is doing now. By lack of other options (the "smoking guns" were certainly partially destroyed)
IMO Boeing obviously could innovate but:
1) The batteries replacement (VERY LOW MTTR) should prompt engineering attention.
2) Tracking on batteries performance (low performance) should started months ago. (could be just software inside charger)
Boeing high rocks were indeed concerned with what engineers said?
mm43 @ # 635
patented charging algorithms
PCBīs certainly are FUBAR and only option is what NTSB is doing now. By lack of other options (the "smoking guns" were certainly partially destroyed)
IMO Boeing obviously could innovate but:
1) The batteries replacement (VERY LOW MTTR) should prompt engineering attention.
2) Tracking on batteries performance (low performance) should started months ago. (could be just software inside charger)
Boeing high rocks were indeed concerned with what engineers said?
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Mother nature "fuse" protected POB on ANA 787 when descending to TAK
Hi,
Bear,
Didn't ANA report a drop to zero voltage concurrent with the smoke event?
Yes, battery voltage went to ZERO because mother nature opened the circuit of the BAD DESIGNED BATTERY as i showed in the equivalent circuit of the probable sequence in an earlier post.
Bear,
Didn't ANA report a drop to zero voltage concurrent with the smoke event?
Yes, battery voltage went to ZERO because mother nature opened the circuit of the BAD DESIGNED BATTERY as i showed in the equivalent circuit of the probable sequence in an earlier post.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Software algorithms and mandatory (airliner use) "per cell temp monitoring"
Hi,
mm43,
1) Per cell temp. seems was not considered necessary
2) As we can see Software is now playing a role in batteries. There are THREE algorithms capable to stress this BAD BATTERY. Both JAL and ANA incidents could be related to this fact (use of patented algorithms)
But, how to verify? This can make impossible for NTSB answer (conclusively) WHY, the current effort.
As i mentioned before, we would need EACH CELL VOLTAGE HISTORY. The comments from JTSB and from IIRC NTSB on 32 V is ABSOLUTELY IRRELEVANT. Were certainly based in a quick look to FDR.
Questions:
1) Battery voltage (BEFORE relay contactor) is showed to crew? Is recorded in FDR?
2) Battery voltage (BEFORE diode module) is recorded?
IMO FDR config. for 787 should be a little bit different.
mm43,
1) Per cell temp. seems was not considered necessary
2) As we can see Software is now playing a role in batteries.
There are THREE algorithms capable to stress this BAD BATTERY. Both JAL and ANA incidents could be related to this fact (use of patented algorithms)But, how to verify? This can make impossible for NTSB answer (conclusively) WHY, the current effort.
As i mentioned before, we would need EACH CELL VOLTAGE HISTORY. The comments from JTSB and from IIRC NTSB on 32 V is ABSOLUTELY IRRELEVANT. Were certainly based in a quick look to FDR.
Questions:
1) Battery voltage (BEFORE relay contactor) is showed to crew? Is recorded in FDR?
2) Battery voltage (BEFORE diode module) is recorded?
IMO FDR config. for 787 should be a little bit different.
Join Date: Jul 2002
Location: Ireland
Posts: 596
Likes: 0
Received 0 Likes
on
0 Posts
areobat
there appears to be no attempt to provide a thermal barrier between the individual cells. This would seem to virtually guarantee that if ones goes, they all go - kind of like storing eight Roman candles right next to each other.
4-6 layers of steel and an air gap looks like a thermal barrier to me!
I'm not saying that it is necessarily enough, but there is clearly an attempt to separate the cells.
Join Date: Aug 2011
Location: Grassy Valley
Posts: 2,074
Likes: 0
Received 0 Likes
on
0 Posts
Bear,
Didn't ANA report a drop to zero voltage concurrent with the smoke event?
Yes, battery voltage went to ZERO because mother nature opened the circuit of the BAD DESIGNED BATTERY as i showed in the equivalent circuit of the probable sequence in an earlier post.
Just because they got caught out in their assumptions about propagation, does not infer that they let a high temp fail safe system out of the design.
I'm still looking at the "WhiteWad"....mesmerized.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Previous posts "briefing"
Hi,
Machaca @ # 655
We can understand most of the reasons why posters are faster to post than to diligently read previous posts. This simply happens. Itīs easier and immediate. 
I would like to emphasize additional benefits in looking previous post:
1) We can learn from different views. Plural and rich.
2) Comments will be much more complete. Will take into account everything.
Worth the effort.
I remember i made this error in one of AF 447 thread on commenting something on "Remote FDR", before F-GZCP wreck was located. Anxiety explains.
Machaca @ # 655
I suggest a re-read of the thread as several expert contributors have previously addressed these supposed new concerns.
I would like to emphasize additional benefits in looking previous post:
1) We can learn from different views. Plural and rich.
2) Comments will be much more complete. Will take into account everything.
Worth the effort.
I remember i made this error in one of AF 447 thread on commenting something on "Remote FDR", before F-GZCP wreck was located. Anxiety explains.
RR_NDB:
Agreed and comes back to what I was saying about inadequate logging in
terms of the number of parameters stored. At a minimum, each cell's voltage,
cell temperature and charge / load current need to be logged to have enough
audit trail to trace back fault conditions to source.
From the info we have, it looks like the charger's led display provides the
equivalent of car's dashboard "idiot lights", which won't necessarily be
that helpfull to investigators.
FX: Message from Charger, "battery has failed". Fireman: Yes, thanks for that,
we just finished putting the fire out
I can see why they might only log a subset of the data, since logging all
at say once per second would take quite a bit of flash memory, but there
are ways round that:
Classify messages as to severity level: Informational, Warning and Critical
1) Informational messages re-use storage, deleting the least recent entries
2) Warning messages are saved until manually cleared, or out of storage,
then as per 1) above.
3) Critical switches to real time (1 entry per second ?) logging and saves all
parameters related to the fault.
The above is standard technique all over industry, so why not this charger ?...
Agreed and comes back to what I was saying about inadequate logging in
terms of the number of parameters stored. At a minimum, each cell's voltage,
cell temperature and charge / load current need to be logged to have enough
audit trail to trace back fault conditions to source.
From the info we have, it looks like the charger's led display provides the
equivalent of car's dashboard "idiot lights", which won't necessarily be
that helpfull to investigators.
FX: Message from Charger, "battery has failed". Fireman: Yes, thanks for that,
we just finished putting the fire out
I can see why they might only log a subset of the data, since logging all
at say once per second would take quite a bit of flash memory, but there
are ways round that:
Classify messages as to severity level: Informational, Warning and Critical
1) Informational messages re-use storage, deleting the least recent entries
2) Warning messages are saved until manually cleared, or out of storage,
then as per 1) above.
3) Critical switches to real time (1 entry per second ?) logging and saves all
parameters related to the fault.
The above is standard technique all over industry, so why not this charger ?...
Last edited by syseng68k; 12th Feb 2013 at 16:10.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Mother nature helped Boeing in TAK
Hi,
SoS:
A failed attempt that ultimately grounded a Dream...
The short circuit of cell # 3 (ANA TAK) was directly related of (improper) SEPARATION of the cells to battery case. I posted the equivalent circuit of this short that ultimately destroyed (fused) the ground green wire connecting the battery case to system ground.
Bear:
The short circuit was immediate! ("light speed" The temp monitoring was useless since then...
Mother nature action was INSTANTANEOUS. Certainly more than 1,000 Amperes from the cell # 3 plus terminal to ground. (A mili ohms load to the 12 Volts of Yuasa LVP 65 three cells)
SoS:
...but there is clearly an attempt to separate the cells.
The short circuit of cell # 3 (ANA TAK) was directly related of (improper) SEPARATION of the cells to battery case. I posted the equivalent circuit of this short that ultimately destroyed (fused) the ground green wire connecting the battery case to system ground.
Bear:
Mother Nature? I don't think so.
The temp monitoring was useless since then...Mother nature action was INSTANTANEOUS. Certainly more than 1,000 Amperes from the cell # 3 plus terminal to ground. (A mili ohms load to the 12 Volts of Yuasa LVP 65 three cells)
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
We technicians cannot (or should not) say IMMEDIATELY everything we learn?
Hi,
syseng68k 2 # 669
This question will be answered by a serious govt. investigation or by a Richard Feynman style work made after Rogers commission Challenger SRB failure (also temperature related )
The commission members were Chairman William P. Rogers, Vice Chairman Neil Armstrong, David Acheson, Eugene Covert, Richard Feynman, Robert Hotz, Donald Kutyna, Sally Ride, Robert Rummel, Joseph Sutter, Arthur Walker, Albert Wheelon, and Chuck Yeager.
syseng68k 2 # 669
From the info we have, it looks like the charger's led display provides the
equivalent of car's dashboard "idiot lights", which won't be necessarily be
that helpfull to investigators.
FX: Message from Charger, "battery has failed". Fireman: Yes, thanks for that,
we just finished putting the fire out
equivalent of car's dashboard "idiot lights", which won't be necessarily be
that helpfull to investigators.
FX: Message from Charger, "battery has failed". Fireman: Yes, thanks for that,
we just finished putting the fire out
The above is standard technique all over industry, so why not this charger ?...
)
Feynman's account reveals a disconnect between NASA's engineers and executives that was far more striking than he expected. His interviews of NASA's high-ranking managers revealed startling misunderstandings of elementary concepts. For instance, NASA managers claimed that there was a 1 in 100,000 chance of a catastrophic failure aboard the shuttle, but Feynman discovered that NASA's own engineers estimated the chance of a catastrophe at closer to 1 in 100
The commission members were Chairman William P. Rogers, Vice Chairman Neil Armstrong, David Acheson, Eugene Covert, Richard Feynman, Robert Hotz, Donald Kutyna, Sally Ride, Robert Rummel, Joseph Sutter, Arthur Walker, Albert Wheelon, and Chuck Yeager.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Effective management must use proper inputs
Hi,
mono:
Same problem when we put put finger in the soldering iron. When we feel it, too late.
mono:
The lag between a short which could cause a runaway event and the associated temperature increase makes it an ineffective protection.
Same problem when we put put finger in the soldering iron. When we feel it, too late.
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Battery from Thales. Also the charger?
Hi,
mm43:
The smoking gun was two batteries labeled Thales. The label in the charger shows Thales responsibility is also in the "surrounding electric environment"
mm43:
The smoking gun was two batteries labeled Thales. The label in the charger shows Thales responsibility is also in the "surrounding electric environment"
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Hi,
HazelNuts39,
Machaca posted:
Unless there is a similar connector in other PCB point i donīt think the wires are coming from temp. probes for each 8 cells. This (ideally) would require more wires. I would require 16 wires or a minimum of 9. The connector mentioned in Machaca post 221 and showed in the picture of your post may be for fewer temp sensor(s).
HazelNuts39,
Machaca posted:
The lower harness (narrower connector on lower left) likely runs to a temp sensor under each cell.
Unless there is a similar connector in other PCB point i donīt think the wires are coming from temp. probes for each 8 cells. This (ideally) would require more wires. I would require 16 wires or a minimum of 9. The connector mentioned in Machaca post 221 and showed in the picture of your post may be for fewer temp sensor(s).
Join Date: May 2008
Location: USA
Posts: 44
Likes: 0
Received 0 Likes
on
0 Posts
SoS -
There is a physical barrier between cells 2 & 3 and cells 6 & 7, however, only a small air gap separates cells 1 & 2, cells 3 & 4, cells 5 & 6, and cells 7 & 8. I guess that could limit damage to just half of the battery, but I suspect that four burning cells on one side of the physical barrier will easily supply enough energy to ignite the four on the other side of the barrier. In fact, I think Boeing has proven it - twice even.
There is a physical barrier between cells 2 & 3 and cells 6 & 7, however, only a small air gap separates cells 1 & 2, cells 3 & 4, cells 5 & 6, and cells 7 & 8. I guess that could limit damage to just half of the battery, but I suspect that four burning cells on one side of the physical barrier will easily supply enough energy to ignite the four on the other side of the barrier. In fact, I think Boeing has proven it - twice even.
Join Date: Jul 2009
Location: France - mostly
Age: 84
Posts: 1,682
Likes: 0
Received 0 Likes
on
0 Posts
Originally Posted by RR_NDB
The connector ... may be for fewer temp sensor(s)
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
Redundancy and voting
Hi,
HazelNuts39:
Yes, may be. The designers (of the battery) clearly assumed temperature of the cells would not be a problem. "Brief" APU use, MAIN batt. use only in emergencies (remember for the first time in an airliner MAIN battery was put COMPLETELY out of the bus) and light use on ground (not so frequent use when not AC present, and towing)
So, the "patented algorithms" comes always to my mind. . i remember many years ago when working and living in a Skydive area i opened a li ion batt of a friend cameramen and found inside: Surprise! Electronics! I was shocked! All my k.I.S.S. principles came to my mind.
During all this days since Jan 7 i am designing a SAFEST (commercially speaking viable solution) and my charger WILL NOT USE ANY SOFTWARE for the critical cell charging. Just hardware (OBVIOUSLY REDUNDANT AND PROTECTED)
IMO algorithms should be made for help: Safety, etc. When relying on it for critical tasks you must have all means to FIRE them IMMEDIATELY when necessary.
The use of algorithms to optimize a DANGEROUS battery with ANY MECHANICAL PROTECTION against positive feedbacks (the so called here, thermal runaway) is :
I would like to analyze at least the schematic diagram of the harness on top of the cells in order to imagine what my colleagues had in their mind. Very probably they put redundancy on "balancing and sensing" wiring.
Certainly thermal aspects were not adequately managed. Temperature measurements, even in three different locations would not help too much. As it seems could be the case. Remember that:
In BOS AND ALSO in TAK the batteries were not supplying energy. (only if something like a diode module (TAK) or short circuit (BOS) but this possibilities seem were discarded by JTSB and NTSB.
Even a relay/contactor failure would not be sufficient for TAK case.
On redundancy i can sadly comment: We are observing redundant errors.
And the problems seems to come from "higher hierarchies" than just engineering/technical level.
HazelNuts39:
...just like there are three wires for each voltage.
So, the "patented algorithms" comes always to my mind.
. i remember many years ago when working and living in a Skydive area i opened a li ion batt of a friend cameramen and found inside: Surprise! Electronics! I was shocked! All my k.I.S.S. principles came to my mind. During all this days since Jan 7 i am designing a SAFEST (commercially speaking viable solution) and my charger WILL NOT USE ANY SOFTWARE for the critical cell charging. Just hardware (OBVIOUSLY REDUNDANT AND PROTECTED)
IMO algorithms should be made for help: Safety, etc. When relying on it for critical tasks you must have all means to FIRE them IMMEDIATELY when necessary.
The use of algorithms to optimize a DANGEROUS battery with ANY MECHANICAL PROTECTION against positive feedbacks (the so called here, thermal runaway) is :
I would like to analyze at least the schematic diagram of the harness on top of the cells in order to imagine what my colleagues had in their mind. Very probably they put redundancy on "balancing and sensing" wiring.
Certainly thermal aspects were not adequately managed. Temperature measurements, even in three different locations would not help too much. As it seems could be the case. Remember that:
In BOS AND ALSO in TAK the batteries were not supplying energy. (only if something like a diode module (TAK) or short circuit (BOS) but this possibilities seem were discarded by JTSB and NTSB.
Even a relay/contactor failure would not be sufficient for TAK case.
On redundancy i can sadly comment: We are observing redundant errors.
And the problems seems to come from "higher hierarchies" than just engineering/technical level.
Join Date: Mar 2011
Location: engineer at large
Posts: 1,409
Likes: 0
Received 0 Likes
on
0 Posts
ICAO is moving to remove the exemption that allowed Li batteries on aircraft as cargo...
"A U.N. agency that sets global aviation safety standards is moving to prevent aircraft batteries like the one that caught fire on a Boeing 787 last month from being shipped as cargo on passenger planes, people familiar with the effort said.
Over the past few days, the members of the International Civil Aviation Organization's dangerous goods committee have proposed revoking an exemption that permitted lithium ion aircraft batteries as heavy as 77 pounds to be shipped on passenger planes, the sources told The Associated Press. All other lithium ion battery shipments on passenger planes are limited to 11 pounds or less because of the batteries' susceptibility to short-circuit and ignite."
"A U.N. agency that sets global aviation safety standards is moving to prevent aircraft batteries like the one that caught fire on a Boeing 787 last month from being shipped as cargo on passenger planes, people familiar with the effort said.
Over the past few days, the members of the International Civil Aviation Organization's dangerous goods committee have proposed revoking an exemption that permitted lithium ion aircraft batteries as heavy as 77 pounds to be shipped on passenger planes, the sources told The Associated Press. All other lithium ion battery shipments on passenger planes are limited to 11 pounds or less because of the batteries' susceptibility to short-circuit and ignite."
Join Date: Feb 2011
Location: Nearby SBBR and SDAM
Posts: 875
Likes: 0
Received 0 Likes
on
0 Posts
An inflight fire inside FWD EE bay was going to happen
Hi,
A simple analysis (equivalent circuit) reveal ANA at TAK case was benign because battery (very probably) destructed an aircraft cable and was (due heat in cell # 3 plus terminal region) destroyed internally (opening) preventing it to continue the thermal runaway like happened in BOS.
The (probable) short circuit was:
Battery was bound to caught fire (like in BOS) but:
1) A short to ground created generated a thousand Amps current to ground from cell # 3 plus terminal, region.
2) This huge current "opened" the circuit.
PS
There is one point in TAK different from BOS:
JAL APU battery probably was at maximum charge level or approaching it.
ANA MAIN battery certainly lost energy during the short circuit.
Question:
This could explain why a fire (external) didnīt happen in TAK? Lower charge after intense discharge?
A simple analysis (equivalent circuit) reveal ANA at TAK case was benign because battery (very probably) destructed an aircraft cable and was (due heat in cell # 3 plus terminal region) destroyed internally (opening) preventing it to continue the thermal runaway like happened in BOS.
The (probable) short circuit was:
Battery was bound to caught fire (like in BOS) but:
1) A short to ground created generated a thousand Amps current to ground from cell # 3 plus terminal, region.
2) This huge current "opened" the circuit.
PS
There is one point in TAK different from BOS:
JAL APU battery probably was at maximum charge level or approaching it.
ANA MAIN battery certainly lost energy during the short circuit.
Question:
This could explain why a fire (external) didnīt happen in TAK? Lower charge after intense discharge?
Last edited by RR_NDB; 12th Feb 2013 at 19:25. Reason: Text tweak