Showing posts with label systems. Show all posts
Showing posts with label systems. Show all posts

Wednesday, October 29, 2014

SpaceX landing on floating platform in the ocean CONFIRMED! =P && Antares (by Orbital Sciences) explodes. =( UPDATED

SpaceX landing on floating platform in the ocean CONFIRMED! =P


So Elon Musk has confirmed SpaceX will attempt to land on a floating platform 90m*50m for their next Falcon 9 flight. YAY!

Florida Today:
He estimated no better than a 50-50 chance of a successful platform landing on the first attempt.
[...]
"So I think we're quite close," he said.
SpaceNews.com:
“We actually have a huge platform that’s being constructed in a shipyard in Louisiana right now,” Musk said in the interview, which was webcast live. He described the platform as about 90 meters long by 50 meters wide. “We’re going to try and land on that on the next flight.
“There’s at least a dozen launches that will occur over the next 12 months,” Musk said. “I think it’s quite likely — probably 80 to 90 percent likely — that one of those flights will be able to land and refly.”

Space.com coverage: www.space.com/27538-spacex-reusable-rocket-test.html


_________________________________________________________


Antares (by Orbital Sciences) explodes. =(


Sadly, the Antares rocket has exploded seconds after launch for their CRS-3 mission:


From simple visual inspection of this video, it seems like an engine failure for one or both engines - from the small engine flash and subsequent smoke; (or other failure causing engine flame-out or automatic shutdown). Subsequently, with no thrust, the rocket fell back to earth and the impact caused the big explosion.

Cargo carried included normal CRS foodstuffs, experiments, etc, and the Arkyd3 - a testbed for Planetary Resources' telescope satellites, including one which was Kickstarted (including a contribution from yours truly). 

According to Wikipedia, Orbital Sciences hasn't had that many failures, with successes in all Minotaur vehicles and the Pegasus with 3 outright failures and 2 partial successes (on primary payload) early in their launch history.Note that this is in line with many launch vehicle companies including SpaceX, whose Falcon 1 platform was their testbed which failed 3 out of 5 times.

However, SpaceX's Falcon 9 has had no failures except 1 secondary payload failure over its 13 total launches these last 4 years. Add to that: NASA pays $1.6 billion to SpaceX to deliver cargo over 12 missions while they have to pay $1.9 billion to Orbital Sciences for only 8 missions. Add to that each mission delivering more cargo in the Dragon by mass than Orbital Sciences' Cygnus spacecraft. The Cygnus is only superior in delivering more pressurized volume than the Dragon, and can't return cargo to Earth like the Dragon does.

Talk about SpaceX cutting costs and competition savings! (The citizens of USA should be glad SpaceX is now able to compete for USAF/DoD/NRO contracts.) I'm glad it's SpaceX that won the CCDev contract, and has a really cool Dragon V2 crewed spacecraft coming out, designed for reusability and landing by thruster functionality. Hurray for SpaceX reusability - providing sustainable and cheaper access to space!

Dailymail's coverage, in-depth: http://www.dailymail.co.uk/news/article-2810128/Ready-liftoff-Nighttime-rocket-launch-International-Space-Station-visible-East-Coast.html
Space.com's coverage, editor promises updates on this page:
www.space.com/27576-private-orbital-sciences-rocket-explosion.html

UPDATE:


More about the accident and engine here: http://www.spacenews.com/article/launch-report/42340antares-rocket-explodes-after-liftoff includes spectator video and sound of explosion:


Press conference at 9pm ET (12pm AEDT, now) on Nasa TV

Broadcast live streaming video on Ustream

More updates to follow probably; from conference:
Emphasis of launch/rocket science and space science being hard (obviously). No injuries, only lost hardware.
Orbital Sciences - will investigate what's wrong, solve the problem and prevent future problems.
Investigation includes evaluating debris (like aircraft investigations [like ATSB]!) - don't touch debris, could be dangerous.
Wallops - good range clearing of safety area.
ISS - incl crew in good shape, have supplies for at least 4-6months always, Progress spacecraft tomorrow, SpaceX on Dec 9, quite a bit of research hardware lost and spare hardware.

Investigation data and everything locked down until investigation process goes through; wait for daytime for debris collecting. Solid propellant burning in area. Contract - provides for ability to eventually get hardware lost on this flight to orbit, Orbital compensates NASA if they don't reach orbit. Launchpad could be damaged, tank pressure sensor readings still available and holding pressure - extent of damage not known until fully investigated.
Engine failure observed in telemetry and visually ~T+10s and range then sent the destruct command ~T+20s, before it hit the ground.
Investigation: Orbital in command, supported very well by NASA and MARS (@Wallops), debris in morning, cameras, telemetry/data. Orbital obviously won't fly until they know and correct the problem.
Insurance - Orbital does have various/some amounts of insurance on their launches.
Engines - Orbital chose to go with this (Russian) engine as there weren't many choices for powerplants of this size, especially in USA, and it was extensively tested and has proved robust. They've also chosen another engine for future Antares, according to the link above providing the video of spectator view. Wonder if they considered using SpaceX's engines =P

Monday, August 11, 2014

Security and ROMs

http://www.engadget.com/2014/08/10/xiaomi-privacy-issue-cloud-messaging/
So apparently anyone could've just read all your messages and a lot of your contact details.
Earlier this week, Finland's F-Secure looked into claims that Xiaomi was secretly sending data from its MIUI-powered phones back to its servers, and it turned out to be true. Despite having not added any cloud accounts, F-Secure's brand new Redmi 1s -- Xiaomi's budget smartphone -- still beamed its carrier name, phone number, IMEI (the device identifier) plus numbers from the address book and text messages back to Beijing. Worse yet, the data was unencrypted, thus allowing F-Secure and potentially anyone to, well, get to know your Xiaomi phone very easily. Earlier this week, Finland's F-Secure looked into claims that Xiaomi was secretly sending data from its MIUI-powered phones back to its servers, and it turned out to be true. Despite having not added any cloud accounts, F-Secure's brand new Redmi 1s -- Xiaomi's budget smartphone -- still beamed its carrier name, phone number, IMEI (the device identifier) plus numbers from the address book and text messages back to Beijing. Worse yet, the data was unencrypted, thus allowing F-Secure and potentially anyone to, well, get to know your Xiaomi phone very easily.
This is partly the reason why I never touched MIUI.
But then in the end, it's pick NSA or MSS, or both. Like Skype, I'd think it highly probably Wechat and other similar apps reported to their home governments with juicy data. Lots and lots and lots of meta data.

____________________________________________________________________________


Also, I have been upgraded to a Sony Xperia Z1 from my Desire HD ("Karuu") this year, which I used after upgrading from the Dopod WinMo5-6 in 2010/2011. In light of my recent rooting and flashing exploits, and having experienced rooting and flashing custom ROMs on HTC (Desire HD), Samsung (S III, Tab), Sony (Z1), ASUS (TF300T), Yuandao (N90), Dopod/HTC (838 Pro), Zopo (ZP950) devices, and dealing with each manufacturer's quirks, here are a list of tips I should follow (while also following my flash guide/checklist):
  1. Do these even when you've done this many times, because a false (or true) sense of proficiency can be misleading and lead to mistakes being made. 
  2. Read everything, then re-read it! (Luckily most chefs/devs highlight potentially bricking hazards with warnings in red text. But still read everything in the OPs, and if more info is needed straight after OPs and latest posts in the thread and use search within thread/sub-forum).
  3. Don't assume each sub-forum (for each device on xda) is similar. Know your sub-forum by exploring. Also, each device handles differently.
  4. Plan the whole process - like using the waterfall methodology in SDLC. Know your aim, and define each step of how to get there. (Regain workable recovery: try flashing this 1st via fastboot, if that doesn't work, then ... etc.) 
This would save lots of headaches and having to do work-arounds which take literally hours longer than I'd have had to if I'd read (all the related threads!), understood, planned it out properly and followed the steps. This is one process where the Agile model way of doing things would work poorly, costing you more time. (And probably brick your device if you're not careful).
When you start trying the more labour-intensive/involved 'solutions', you start losing your shorter simple options because of wipes and incompatibilities between different partitions of your device. It's kind of like in cooking how you can always add more X but once it's mixed in, you can't take X out.

You can see why people operators like NASA, SpaceX, pilots, etc. have a pre-flight/launch checklist and the whole launch sequence timelines. One little mistake can have catastrophic consequences (not just bricking a device, more like death).