SoylentNews is people
SoylentNews
It takes chutzpah to tweet “rockets are tricky” shortly after one you have just launched has deliberately blown itself up. But Elon Musk, founder and boss of SpaceX, is not a man who lacks self-confidence, and he did just that on August 22nd after the terminal malfunction of one of his company’s Falcon 9 vehicles. That Arianespace, a French rival of SpaceX, announced on the same day that two satellites it had tried to launch to join the European Space Agency’s Galileo constellation (intended to rival America’s Global Positioning System), had entered a “non-nominal injection orbit”—in other words, gone wrong—shows just how difficult the commercialization of space can be.
If spacecraft are so precarious, then perhaps investors should lower their sights. But not in terms of innovation; rather in altitude. Airbus, a European aerospace company, thinks that developing satellite-like capabilities without satellites is the answer. Hence the firm’s recent trial, at an undisclosed location (but one subject to Brazilian airspace regulations) of Zephyr 7, a high-altitude “pseudo-satellite”, or HAPS for short.
Zephyr (named after the Greek god of the west wind) is actually an unmanned, ultralight, solar-powered, propeller-driven aircraft. But it is designed, just as some satellites are, to hover indefinitely over the same part of the world. With a 23-metre wingspan and a weight of only 50kg, it is fragile and must remain above the ravages of the weather and the jet stream both by day and by night. It therefore flies at an altitude of around 21km (70,000 feet) during daylight hours, and then glides slowly down to around 15km when the sun is unavailable to keep it aloft.
(Score: 2) by evilviper on Friday September 05 2014, @01:16PM
No, thoughtless rundancy doesn't ever work. The obstacles (eg. weather) that prevent one craft from doing it's job, will prevent both. And without seamless hand-off, you are decreasing uptime every time you swap them out.
Just throwing more craft at the job, is like putting more redundant servers in the same rack. No matter how many you add, it still won't improve reliability.
Hydrogen cyanide is a delicious and necessary part of the human diet.
(Score: 2) by sjames on Friday September 05 2014, @03:05PM
Nobody said anything about thoughtless redundancy though. Yes, you have to watch out for conditions that will take all redundant parts offline at once (black swan events, for example), but that doesn't detract from the principle.
Of course you also have to consider that if the satellite fails, it could be a months long national outage rather than a days long local outage.
(Score: 2) by evilviper on Saturday September 06 2014, @06:51AM
Yes, that's exactly what I was responding to:
"If you have two 9s of reliability with one vehicle, you'll have 4 9s of reliability by operating two vehicles simultaneously"
Yes, it does. You have to design things from the start to be able to recognize faults, and handle fail-over. Just having extra units doesn't do any of that, and instead will increase your downtime.
A month-long outage after a decade or two of operation, is far higher availability than a day-long outage every couple months. And the more frequent outages will have people scrambling for alternatives, while the one-off outage can be more easily worked-around one time rather than ongoing day-to-day.
Hydrogen cyanide is a delicious and necessary part of the human diet.
(Score: 2) by sjames on Saturday September 06 2014, @10:34AM
I suspect OP was simplifying. Surely you don't expect someone to spend a year designing the perfect failover system before replying! Pull the stick out!
In many cases, a day long outage from time to time is much more tolerable than a months long one where months will be a lot more than 1 when you have to launch a new satellite, even if you have one ready to go.
If you can't tolerate a day long outage, then you need to have a spare in each location ready to take off.
(Score: 2) by evilviper on Saturday September 06 2014, @10:48AM
You're still calling me a liar for simply responding to what the GP said.
No, but I do expect some acknowledgment that such fail-over systems have costs (not just "Build two!"), and consideration that there might not even be practical options in some cases.
There's a big difference between a day-long outage every decade, and a day-long outage every couple weeks. The later would be completely unacceptable for TV viewers, most internet access, etc. etc. Fine for bulk data transmission, but not much else.
It works in the other direction, too. It's far cheaper and easier to depend on a satellite that works great for a decade, and when it goes out, have everyone aim their dish for the next satellite and keep it there for the next decade...
Hydrogen cyanide is a delicious and necessary part of the human diet.