SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : Loral Space & Communications -- Ignore unavailable to you. Want to Upgrade?


To: Mr. Adrenaline who wrote (2371)4/16/1998 11:59:00 PM
From: Mr. Adrenaline  Read Replies (3) | Respond to of 10852
 
This thread has been busy! I can't even hope to go back and read everything I've missed..
It's been awhile since I've even read SI, let alone posted anything. I've been busy. In case anyone is interested, the first four satellites are fine and we are ready for the second launch.

One thing that I have wanted to write about is asteroids/space junk. Every once in awhile there are questions about that. Let me start be describing what is known as Ps. If I could write it here, I would write a capital P, with a small s subscript, and it would be read "P sub s". It stands for "probability of success". Any well-engineered part/device has had its Ps calculated. For example, while I have never worked in the automotive industry, I am certain that the car that you drive has a Ps. Let's assumes it's 0.95. That would mean that your car has a 95% chance of fulfilling it's intended life cycle, which is probably 15-ish years.

A GEO satellite has a design-life of usually 12 to 15 years, and a Ps of somewhere in the mid to upper 90's. Globalstar has intended lifecycle of 7 1/2 years. When the Ps is calculated, all of the life limiting factors are accounted for, and there is a budget for different sub-systems and parts. This budget is critically reviewed by an insurance underwriter, if the satellite is insured.

I bring this all up because no where in this budget is there a line item for "impact by asteroid". The reason is simple. It has never happened. This is one time we can be thankful for lawyers. (I feel the flames coming already.) If there was a finite probability that a satellite could be lost due to asteroid impact, the insurance companies would have found it by now, and require that it be accounted for.

There is one notable possible exception. That is the upcoming Leonid meteorite showers. It has enough people concerned, that the American Institute of Aeronautics and Astronautics has a conference dedicated to it. I must say here that I am not an astronomer, and that I haven't followed this closely enough to even call myself well informed on this subject. But. In August of 1993 the annual Perseids meteorite shower was supposed to be a big one. A lot of government and private agencies published warnings for satellite operators. Yours truly got to spend a sleepless week baby sitting a herd of satellites, "just in case", so that we could take immediate steps if a satellite happened to get hit. It was all in vain. And I suspect that the Leonids will be the same. But it does give those who worry about such things something to do.

But let's assume the worst. For sake of argument, let's assume that the Leonids took out 10% of the G* constellation. The law of averages would also dictate that we would have to take out 10% of other constellations as well. That means 10% of Intelsat, Comsat, etc., etc. Because of the nature of his thread, lets limit the discussion to G* and I*. It would hurt both constellations, but not equally. That's because technically G* could operate with one satellite. The coverage would be terrible, of course. You'd have to publish the times that the satellite would be overhead in a newspaper, and some areas would see outages for days a time. Of course this is not a viable scenario, but I exaggerate to make a point. Somewhere on this thread I published the probability of having a satellite(s) over one's head for a given latitude. If G* lost 10% of it's constellation, then that coverage would change, as would the entire system capability until such time as replacements were brought on line. But the capability erodes gracefully -- in other words roughly linearly. This is due to the "bent pipe architecture. I* would not be so fortunate.

Because I* relies on inter-satellite links, as a satellite fails the capability to route calls falls off rapidly. For those I* fans, I'll say save your flames. This is not "I* bashing", but just plain physics. In order to have a satellite link, the satellites must "see" each other. If a orbital plane is fully populated, a given satellite can route calls to either one of two neighbors in that plane. I am not an I* expert, but it is my understanding that calls are routed within a plane until the planes intersect over the poles. This was done for a couple of reasons, but (again my understanding is) the major one was the lack of traffic at the poles. At the poles a call may be switched from one orbital plane to another as a call is routed from say, California to Japan.

If one satellite is taken out of the orbital plane, calls can be routed only one way on its trip to the poles. That is, the chain has been broken because when one satellite is taken out, it's two neighbors can't see each other because the Earth's horizon blocks the view. If this was limited to one plane only, chances are no one would notice until a backup got on line. But if I* lost 10% of it's satellites in my hypothetical scenario system wide capability would suffer dramatically.

By how much, I can't say. Again, I'm not an I* expert. It is well known in the industry that I* will suffer more from losing satellites than G* would. (The G* program has it's share of I* defectors working for them, and visa-versa, which is where I get my information.) But I can't put numbers to it.

But all this is purely academic. Neither G* nor I* will loose 10% of it's satellites at any given time. Both systems will lose an occasional satellite, and both systems are equipped to deal with it.

Mr. A