Slashdot Mirror
Mars Camera's Worsening Eye Problems
Mr_Foo writes "According to a Nature article, the Mars Reconnaissance Orbiter's HiRISE imager is suffering from a loss of peripheral vision. The problem surfaced less than a month after the orbiter reached Mars. One the camera's four color detectors has completely stopped working, and it is feared that the problems are spreading. Currently seven of the fourteen HiRISE's detectors are sending back corrupted data and although the issue is only creating a 2% loss of signal at this time it is expected to worsen. The lead investigator for the mission is quoted as saying the problem is systemic: 'In the broken detectors, extra peaks and troughs are somehow being introduced, causing... a "ringing" in the signal. "We don't know where the ringing is coming from," [the investigator] says.' Warming the electronics before taking images seems to help the problem. This effect might be one reason why the detectors on the cold periphery of the array were the first to pack up."
I don't know why anyone is every surprised when stuff goes wrong on missions, or equipment breaks down. Nasa is a governmental agency and as such has a big beaurocratic morass, and often different divisions don't know what the others are doing. I used to date the daughter of the Vice President of Operations at Nasa. The first time I ever got to talk to him, I was so excited; I asked him about an upcoming mission that was going to be taking off. It might actually have been this one, it was a year or two ago when it left. He actually hadn't heard about the mission; I had to describe it to him.
This is probably the first time not a single story about a defect in an image sensor had pictures showing the effects of the defect.
I read that as ringing in the electrical signal. Possible caused by a ground on a coax cable working loose
or maybe a component failing due to thermal cycling or cold stress.
The boards are built at room temperature, it's pretty cold in space if the sun isnt shining on something.
Parts contract and if whatever they're attached to doesnt contract at the same rate, if can loosen things
or even crack them over time. Qualifying parts for that sort of thermal stress is what makes things cost so
much for Space Grade parts. It's also where they try to save money, only test 10 parts instead of 100 or
only test for 10 day/night thermal cycles instead of 1000. It's always easy to say after the fact that
they should use better, more expensive parts, but sometimes if you do, the mission goes over budget and
doesnt get done at all.
Starman97@Gmail.com (bring it on spammers)
Seem to recall the Hirise sensors have 256 cells for every column. Each cell samples the same point of light. The 256 samples are averaged to give a final result. Motion of the camera repeats the process for the next row. If 4 cells in a column die, you should get a 2% reduction in dynamic range, but far better than any consumer camera.
Deciphering the scant information, it sounds like if we could see the defects, they would have the same type of ghosting you used to have with old SVGA cables. It's probably restricted to columns and looks like a double image in certain columns.
Wouldn't the use of a ThermoElectric Generator negate most of the issue because of the heat generated?
My feeling is that this is a radiation caused problem. Radiation can "move" particles
on the electronic circuits. This will affect timing (faster or slower depending on
the circuit), and this is a reason why the temperature change has an impact.
The only thing is that I will not expect to have very new technology (350nm or older).
If this is the case, there should be HUGE amounts of radiation to have such impact.
Well, I guess that it is space after all.
btw, if the error is much bigger than the radiation models could predict, the satelite
may be going through some "unexpected" radiation source (great) or there was some
problems during design (sucks).