Curiosity rover gives Mars the middle finger, prepares to get drilling After an extensive summer of long, long, long-distance remote support by NASA engineers on Earth, it looks as though Curiosity on Mars is back in the drilling game. At the end of 2016, the rover's drill bit, which it has used more than a dozen times to explore the bedrock of the Red Planet, started playing up. The problem was …
Lest anyone think that Drilling or Digging on Mars was an easy Topic imagine this - you have to select the type of drill your going to use based on a guess of what material you will find on Mars. A few years later when you actually get to Mars, you have to guess how hard the rock is that you want to drill and what it's made from based on Pictures taken of the outside surface. You then have to set appropriate speeds and feeds such that a) you can drill in the rock, and b) if you're wrong about the consistency or that directly below the surface there is some harder impurity or the like that you wont destroy your bit. And you have to program this with enough feedback loops to do it autonmously, because you cant manually stop the drill if something goes wrong. At least not for 28 minutes (14 minutes for notification of the failure to get to Earth and 14 minutes for the big Stop Signal to be sent back).
To put that in Earth Terms - imagine you have to select a drill bit to drill a hole in the wall of a house that you havent seen and the best you can say is that the house is in, for example, Belgium. Then you have to select the right feeds and speeds for use of that drill based on a photo of the wall. Do you take the masonry bit knowing that most walls in Belgium are masonry, but also knowing that if you hit a metal stanchion you're screwed? Or do you take the carbon steel bit knowing that you can probably cut through anything, but that you are likely to damage the surrounding areas and damage the science you want to conduct in those areas?
The Boffins working on these things deserve a huge round of applause (and a pint) for their great work making these operations so successful!
Just to be clear, while the right type of high speed carbon steel bit will cut through most metals, it won't effectively go through rock. Masonry bits are tungsten carbide tipped are a lot of metal milling bits, the difference is in the angle of attack but with rock/masonry it is also better to use a hammer action which I doubt is used on a Mars Rover as it can be quite destructive of the tool. In this case I suspect the bit is quite exotic and combines diamond and tungsten carbide. If I were designing a very remote robot , feedback from the rate of bit feed would be essential and I am nowhere near as smart as the guys at NASA, so they probably have that covered.
They make such appliances. For home use, not just commercial. But they are not cheap. Basically, you can shell out big money and make a major appliance purchase once & be done with it, or you can make the same purchase every couple of years for the rest of your life.
The expensive version usually costs three-four times what the cheap models cost. Do the math(s) and follow your bliss.
"They make such appliances. For home use, not just commercial. But they are not cheap. Basically, you can shell out big money and make a major appliance purchase once & be done with it ..."
Yeah, tried that. AEG washing machine, three times the price of a Hoover. After two years it packed up. Not one to chuck away seven hundred quid lightly, I got into the electronics. After a month of on-and-off debugging of solid-state stuff floating at line voltage, I found that a crappy 100k carbon film resistor had gone open circuit.
I discovered the controller was Italian -- I'd never have bought the machine if I'd known that. That resistor must have been a 0.5 penny component. Replaced it with a metal film type, and four years later it's still washing.
AEG are mostly the same as every other Electrolux Group product these days, sadly. Might as well buy a Zanussi. I recall seeing the difference in old pre-Electrolux AEG and new stuff with one image - the bearing set of a washing machine drum : the smaller, rear bearing from the old set was bigger than the front bearing from the new set.
Not sure if it's still the case, but Miele might be one of only a couple of home kitchen appliance makers that's still doing it properly. Ebac is making a small range, although I'm not sure if they buy their internal bits in from one or other of the big two groups.
Sad demise, really - names you could trust are gradually being diminished.
I have a Bosch dishwasher which is over 10 years old and I am sure will reach 13. It was assembled by me as I had to change its failed water heating element and discovered that Bosch start with the element and assemble the whole machine around it, so it was completely dismantled on my kitchen floor one Christmas.
However, I doubt my dishwasher would last long on Mars. Huge respect for the Curiosity Rover project, I wish I was part of it.
I asked the tour guide at JPL why NASA seemed consistently to massively underestimate project lifetimes, including basing future expectations on past performance, but since I was with a load of grade school kids, he preferred to answer a question about whether the ISS had weapons in case aliens attacked.
I asked the tour guide at JPL why NASA seemed consistently to massively underestimate project lifetimes, including basing future expectations on past performance,
It's hard to make comparisons to past performance when most NASA probes are not duplicates of prior hardware. Each probe, or pair of probes, usually has a lot of new hardware that hasn't operated in the novel situations NASA throws them into. You couldn't use (for example) Voyager's performance to predict anything about the Curiosity rover, which differs in everything from RTG to computers to operating environment.
An example of the risk in estimating lifetime is found with Opportunity and Spirit. NASA estimated, based on Sojourner's quick demise due to electrostatic accumulation of dust on solar panels, that 90 days would be a reasonable lifetime for Spirit and Opportunity. They had clear past performance to guide this calculation. However, the Mars Exploration Rovers did not accumulate dust as fast as expected because they were taller than Sojourner. Further, Sojourner's short life did not give NASA the chance to experience the "cleaning events" that blow dust off MERs' solar panels and helped extend their lives by years.
In the opposite direction, Curiosity based its wheel design on the proven engineering lineage of Sojourner's and MERs' aluminum wheels. However, this design proved inadequate in the case of the much larger Curiosity, which is battering its wheels badly. While NASA had worried about dust wear at the hubs (addressing that concern delayed Curiosity's flight by two years), it didn't consider the rapid demolition of the wheels themselves. After all, they were well-proven by past performance.
I can make a guess what's going on with NASA's estimates for probe lifetimes: they're padding their estimates with factors of safety but only using the un-padded value. For example, I'm currently working on a team to develop an adaptor that will mate some underwing hardware to an aircraft that makes carrier landings. We have to certify the adaptor will endure a worst case carrier landing G-load of X G's, where X was given to us by the aircraft maker and US Navy. There's a minimum factor of safety in strength (x2.5) we could use if we put a prototype adaptor on a plane from the Boneyard and (in a quantified, scientific manner) tear it off the airplane. But that's a bloody expensive test, so instead we're using a larger safety factor (x3.75). There's also safety factors for the use of the large billet of metal (the interior properties of thick-sectioned metal drop compared to thinner material), the manner of assembly (e.g., welding knockdowns), and so on. By the time it's done, I'm fairly certain the adaptor could hold the hardware if the plane was flying in the clouds of a 75G brown dwarf star but we're only going to say, "Yes, we'll warranty this for X Gs."
NASA was probably hoping for years of service out of its MER rovers, but it had no prior examples of those exact models of rovers operating in Mars' environment. So, it was only going to guarantee and plan around 90 days of service, and set minimum success targets of 30 days. It got a little more hopeful with Curiosity (2-year target lifespan) but, again, it hasn't deployed anything like Curiosity before.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
