Russia's Latest Excuse for Phobos-Grunt Failure
Russia blames ‘cosmic rays’ for Mars probe failure, AFP
“Carrying out such a large-scale, lengthy job, they should have taken into consideration the effect of outer space on the equipment of an interplanetary station.” He also blamed sub-standard or fake foreign-made microchips used in the probe, saying more than 60 percent were not designed for use in space. “This is imported equipment and of course this is probably a reason,” he said. Russia had previously speculated that radiation from US radar could have put the probe out of action.”
Earlier posts
How does he know that more than sixty percent weren’t designed for use in space? If there was some portion of counterfeit parts then well that is something that happens but they couldn’t possibly put a number on it now unless they caught them at some point or knew what they were installing.
And the idea that it couldn’t handle the environment in LEO seems… Off.
This sounds like obvious junk to me. Am I missing anything as an outsider?
I suspect the comment about 60% of the parts not designed for use in space just means that they hadn’t been qualified. It’s not the same as counterfeit parts. That likely means they hadn’t been tested under high radiation and cosmic particle conditions. The parts might work but they just don’t know. They would be likely to work in LEO but could have problems on the trip to Mars.
It could also mean that 60% of the parts were not rad hard; which is by no means unusual.
I read another article where they quoted the person as saying counterfeit parts. If that’s the case, then they did a lousy job of buying the parts (ie getting them from disreputable suppliers).
The main thing is that rad hard parts would be required for the long trip outside of the Van Allen belts, not so much for LEO.
I also read where both processors rebooted at the same time and they suspected cosmic rays. No location was given but usually that’s likely to be over the south Atlantic ocean where the radiation is highest. For both processors to reboot is usually an indication of a software problem not discovered until flight (aka Ariane 4 I think where both computers rebooted shortly after launch causing the rocket to be destroyed) or a single event upset problem with the control circuits. That is a design issue and can easily be fixed with competent design engineers.
That makes more sense.
I’m curious why that wouldn’t be unusual; I had the impression that everything would have to be rad hard.
It’s very hard to fault Jim Oberg’s analysis in this week’s Space Review. Reading the statement from Russia suggests that they are still in the dark as to what happened. Maybe they should read this:
http://www.thespacereview.c…
I would think that the Russians understand that these “excuses of the week” don’t build much confidence in their customers. A customer of course understands there is a risk of their payload being lost, but customer’s also expect that the launch provider or people designing the hardware will work in a step-wise, objective process, rooted in scientific method, to diagnose issues or failures. We haven’t just heard crazy and numerous “explanations” for the loss of Phobos-Grunt from the media–we expect that sort of thing–but they’ve also come from people within Russia’s space organization and government.
It’s unbelievable that parts supply line wouldn’t be well vetted and that said parts wouldn’t be tested properly. How the hell do “fake, foreign-made microchips” get into the supply line? I was willing to give the Russians the benefit of doubt for their series of mishaps over the last couple of years, but with all the claims, posturing, and wildly fantastic explanations coming out of their government and related organizations, it seems they’ve lost their way…..to put it lightly.
PsiSquared – Counterfeiting or providing substandard parts (electrical, structural, etc) actually happens a lot more then you might think. This doesn’t just happen for Roscosmos. It happens for NASA, DOD, and private industry. In the US NASA participates with GIDEP (Government – Industry Data Exchange Program) http://www.gidep.org. For integrated circuits (chips), often batches are made that don’t meet the spec. They chips are thrown out only to be picked up by counter fitters who in turn sell them as if they are the real part.
Nox is right. By the way, there is a way around defective parts. At one time both the vendor and the user engaged in thorough testing and inspection, including stress testing and the famous Joy test. Over time, solid-state components developed such a reputation for reliability that end users offloaded the inspection role (QA) to the vendor. As time passed, many vendors, with an eye on the bottom line, went to sample testing a few random components, and some started skipping testing entirely. Today, you should probably assume that unless you specify it, only a few of the components you receive may have been tested, and it’s possible that none have. It seems to me that Russia may want to revive 100% incoming test on all components and not just functional testing, but stress testing as well.
It probably won’t make you feel any better, but parts counterfeiting and substitution is a huge issue in aviation as well.
This is sad. Apple products (ipods, ipads, & laptops) work just fine in LEO, but the Russians cannot design a deep space vehicle to handle the same environment.
http://nottotallyrad.blogsp…