View Full Version : 787's with GPS rollover issue grounded
hoss183
11th Apr 2019, 14:05
15 Boeing 787s were grounded in China due to GPS rollover issue. ? News In Flight (http://newsinflight.com/2019/04/08/15-boeing-787s-grounded-in-china-due-to-gps-rollover-issue/)
India Four Two
12th Apr 2019, 02:07
And a KLM 777:
https://arstechnica.com/information-technology/2019/04/gps-rollover-apparently-cause-of-multiple-flight-delays-groundings/
tdracer
12th Apr 2019, 03:03
Honeywell provides the flight management and navigation software for the Boeing 777 and 787. The company issued two service information letters (https://aerospace.honeywell.com/en/news-listing/2019/march/service-information-letter-briefing-gps-week-rollover) in March warning of the GPS week-counter rollover and providing technical support data.
On another thread I made a post about how some operators often ignore airframer service bulletins - even those that have safety of flight implications.
You found two operators that proved my point...
Smythe
12th Apr 2019, 04:06
However, EASA said, tests of some GPS devices revealed that not all manufacturer implementations will correctly handle the April 6 rollover. The EASA bulletin is intended to help operators understand the implications of the week-number rollover and recommend steps to rectify the issue.
https://ad.easa.europa.eu/ad/2019-01R1
So far, the only reported issues have been with Honeywell GPS systems...so far.
Add: BOM weather balloons grounded due to rollover. BOM said its equipment supplier advised the bureau of the fault after the rollover on April 7, and balloon launches ceased on April 8.
A Shanghai Airlines 787 was also reportedly affected.
fox niner
12th Apr 2019, 06:31
And a KLM 777:
I can guarantee that there has never been a 777 from klm, that was scheduled to fly to Bogota. Never. So that part of the story does not hold.
DaveReidUK
12th Apr 2019, 06:48
I can guarantee that there has never been a 777 from klm, that was scheduled to fly to Bogota. Never. So that part of the story does not hold.
Last weekend's KL741 was a B772.
Herod
12th Apr 2019, 08:04
A delayed Y2K problem?
kristofera
12th Apr 2019, 08:12
15 Boeing 787s were grounded in China due to GPS rollover issue. ? News In Flight (http://newsinflight.com/2019/04/08/15-boeing-787s-grounded-in-china-due-to-gps-rollover-issue/)
Sounds like fake news that is based on a single tweet. The tweet (and article linked to above) include a photo from an aircraft with tail number B-1113. The same aircraft has been operating flights every day this month...
https://www.flightradar24.com/data/aircraft/b-1113
DaveReidUK
12th Apr 2019, 08:53
Sounds like fake news that is based on a single tweet. The tweet (and article linked to above) include a photo from an aircraft with tail number B-1113. The same aircraft has been operating flights every day this month...
https://www.flightradar24.com/data/aircraft/b-1113
The aircraft would have grounded to allow embodiment of the SB that has been applied to pretty well every other 787 in the world equipped with a Honeywell GPS in order to mitigate the issue.
Presumably it takes rather less than 24 hours to accomplish. :O
GordonR_Cape
12th Apr 2019, 08:58
A delayed Y2K problem?
More like an early version of the Unix year 2038 rollover problem: https://en.wikipedia.org/wiki/Year_2038_problem
The Year 2038 problem relates to representing time in many digital systems as the number of seconds passed since 1 January 1970 and storing it as a signed 32-bit binary integer.
More like an early version of the Unix year 2038 rollover problem: https://en.wikipedia.org/wiki/Year_2038_problem
GPS rollover is not directly related to Y2K or the Unix year 2038 problem, but the underlying concept is the same.
The GPS standard has 10 bits of data for the date which is only enough for around 20 years. The last rollover was in 1999.
When unpatched GPS units rolled over to 000000000 on 8th April, the date showed as 1999. The Honeywell service bulletin below explains it well. The standard has now been updated to 14 bits of data so we won't have to worry about this again for a few hundred years.
https://aerospace.honeywell.com/en/news-listing/2019/march/service-information-letter-briefing-gps-week-rollover
B787register
12th Apr 2019, 15:59
Does anyone know which registrations are grounded I know of a few
Smythe
12th Apr 2019, 16:00
The CNAV GPS week number is now represented as 13 bits, or 8192 weeks, and only repeats every 157.0 years, meaning the next return to zero won't occur until the year 2137
Interesting that the 787 has issues with this. GPS rollover was basically fixed the last time it happened in 1999 :bored:
ooops:
At 7:59 p.m. E.D.T. on Saturday, the New York City Wireless Network, or NYCWiN, went dark, waylaying numerous city tasks and functions, including the collection and transmission of information from some Police Department license plate readers.
The shutdown also interrupted the ability of the Department of Transportation to program traffic lights, and prevented agencies such as the sanitation and parks departments from staying connected with far-flung offices and work sites.
The culprit was a long-anticipated calendar reset of the centralized Global Positioning System, which connects to devices and computer networks around the world (https://archive.nytimes.com/www.nytimes.com/library/tech/99/08/biztech/articles/23gps.html?module=inline).
A Squared
12th Apr 2019, 21:02
Meanwhile my 20+ year old Garmin portable powered up with the correct position and date on the morning after the rollover.
Smythe
12th Apr 2019, 22:03
Meanwhile my 20+ year old Garmin portable powered up with the correct position and date on the morning after the rollover.
As will all of these GPS systems. The almanac reference really affects first acquisition time, and the number of sats that have been put into place since the last almanac in 1999, ie if it is not in the almanac, it is not used.
A Squared
12th Apr 2019, 23:15
As will all of these GPS systems.
So what's your working theory here, that the articles about the 787's not making the switch as well an the picture of the PFD with the date of 1999 are all fake?
So what's your working theory here, that the articles about the 787's not making the switch as well an the picture of the PFD with the date of 1999 are all fake?
Smythe is partially correct. The bug is an initialisation problem only, however the affected GPS systems will not start up with the correct date. Without the correct date/almanac loaded the GPS systems will effectively be doing a cold start and it takes up to 15 minutes to download the full almanac from the GPS satellites (time to first fix), at which point they will show the accurate time and date for as long as they are online.
The reason the aircraft are grounded due to this problem is there a risk that if for any reason the GPS system had to be rebooted during flight it would take the GPS an excessively long time to get a first fix and come online.
https://en.m.wikipedia.org/wiki/Time_to_first_fix
Smythe
13th Apr 2019, 00:53
So what's your working theory here, that the articles about the 787's not making the switch as well an the picture of the PFD with the date of 1999 are all fake?
Why would it be a fake??? As in my post, and as eppy also noted, the effect is the acquisition time of first fix. Your older system will still boot up, as will all the others, even the ones that show the date as 1990. It will just take a while to acquire the sats and database will be short..
GordonR_Cape
13th Apr 2019, 05:02
Non aviation answer about GPS rollover by mathematical comedian Matt Parker:
https://youtu.be/iyz7dSnZItw
bill fly
23rd Apr 2019, 07:04
Finally the Millennium Bug...
It’s been quite a wait.
:)
JetMender
23rd Apr 2019, 19:35
Many 787s world-wide were affected by this issue, despite Boeing / Honeywell saying beforehand there should not be a problem.
It caused multiple issues & the fix was not straightforward.
GordonR_Cape
23rd Apr 2019, 20:15
Many 787s world-wide were affected by this issue, despite Boeing / Honeywell saying beforehand there should not be a problem.
It caused multiple issues & the fix was not straightforward.
Some details would be interesting, if you have any:
Was the fault with the date only, or the actual GPS function?
Is the GPS date used by the aircraft flight systems?
Does the aircraft flight system closely couple to the GPS, or is it read-only?
Do the problems extend to the inflight entertainment, and was that part of the grounding?
What was the difference between B787s that were grounded, and those that were not?