-
#160
by
jtrame
on 10 Aug, 2012 13:46
-
-
#161
by
Star One
on 10 Aug, 2012 19:54
-
Just a question about the RAD hardened processors they use on vehicles such as the MSL, is it true they are based on Sapphire rather than Silicon as I saw that quoted somewhere recently?
-
#162
by
mtakala24
on 10 Aug, 2012 20:11
-
heavy duty radiation-hardened processors are usually built on non-conductive substrate with silicon - compare that with conventional processors that are built on a silicon substrate from where the structures are built by etching away the unneeded stuff. Don't know about sapphire, though.
-
#163
by
Star One
on 10 Aug, 2012 20:18
-
heavy duty radiation-hardened processors are usually built on non-conductive substrate with silicon - compare that with conventional processors that are built on a silicon substrate from where the structures are built by etching away the unneeded stuff. Don't know about sapphire, though.
Isn't the architecture a lot simpler & build on a larger micron size as well, from what I read it was compared to being akin to what you would see in a home PC CPU about ten years ago?
-
#164
by
Retired Downrange
on 10 Aug, 2012 20:33
-
Just a question about the RAD hardened processors they use on vehicles such as the MSL, is it true they are based on Sapphire rather than Silicon as I saw that quoted somewhere recently?
http://en.wikipedia.org/wiki/Mars_Science_LaboratoryComputers: The two identical on-board rover computers, called "Rover Compute Element" (RCE), contain radiation-hardened memory to tolerate the extreme radiation from space and to safeguard against power-off cycles. Each computer's memory includes 256 KB of EEPROM, 256 MB of DRAM, and 2 GB of flash memory.[27] This compares to 3 MB of EEPROM, 128 MB of DRAM, and 256 MB of flash memory used in the Mars Exploration Rovers.[28]
The RAD750 is a radiation-hardened single board computer manufactured by BAE Systems Electronic Solutions.[1] The successor of the RAD6000, the RAD750 is for use in high radiation environments such as experienced on board satellites and spacecraft.[2] The RAD750 was released in 2001, with the first units launched into space in 2005.[1][3]
The CPU has 10.4 million transistors, nearly an order of magnitude more than the RAD6000 (which had 1.1 million).[3] It is manufactured using either 250 or 150 nm photolithography and has a die area of 130 mm².[1] It has a core clock of 110 to 200 MHz and can process at 266 MIPS or more.[1] The CPU can include an extended L2 cache to improve performance.[3] The CPU itself can withstand 200,000 to 1,000,000 rads (2,000 to 10,000 gray), temperature ranges between –55 °C and 125 °C and requires 5 watts of power.[1][3] The standard RAD750 single-board system (CPU and motherboard) can withstand 100,000 rads (1,000 gray), temperature ranges between –55 °C and 70 °C and requires 10 watts of power.[3]
The RCE computers use the RAD750 CPU (a successor to the RAD6000 CPU used in the Mars Exploration Rovers) operating at 200MHz.[29][30][31] The RAD750 CPU is capable of up to 400 MIPS
http://en.wikipedia.org/wiki/RAD750Searching for RAD750 and Sapphire gives NO results
-
#165
by
Robotbeat
on 10 Aug, 2012 20:35
-
Just a question about the RAD hardened processors they use on vehicles such as the MSL, is it true they are based on Sapphire rather than Silicon as I saw that quoted somewhere recently?
http://en.wikipedia.org/wiki/Mars_Science_Laboratory
Computers: The two identical on-board rover computers, called "Rover Compute Element" (RCE), contain radiation-hardened memory to tolerate the extreme radiation from space and to safeguard against power-off cycles. Each computer's memory includes 256 KB of EEPROM, 256 MB of DRAM, and 2 GB of flash memory.[27] This compares to 3 MB of EEPROM, 128 MB of DRAM, and 256 MB of flash memory used in the Mars Exploration Rovers.[28]
The RCE computers use the RAD750 CPU (a successor to the RAD6000 CPU used in the Mars Exploration Rovers) operating at 200MHz.[29][30][31] The RAD750 CPU is capable of up to 400 MIPS
That didn't answer the question.
-
#166
by
rds100
on 10 Aug, 2012 20:49
-
About rad hardened computers i wonder if you take a regular (OK, industrial) computer and wrap it with say 5mm of lead - how much rad hardening would that add, compared with the specially built rad hardened computers?
-
#167
by
Retired Downrange
on 10 Aug, 2012 20:51
-
Robotbeat....
I'm on an iPad and have to use the edit function to add additional copied material...
and you chimed in prior to my last edit
I see no specific reference to using Sapphire in the RAD750, looking several places, (including the manufacturers spec sheet) but with this info, one could search further to confirm an answer.
-
#168
by
thomson
on 10 Aug, 2012 21:39
-
MSL came with 6 balance weights that were discarded in the early EDL phase. Why are those weights made of tungsten?
One property that comes to mind when talking about tungsten is its very high melting temperature (3422 C), but why is that useful? After they are discarded, what problems would they caused if they melted?
Is it that they want it to hit the ground as solid to make bigger impact? My understanding is that the kinetic energy delivered would be mostly the same - solid or molten.
-
#169
by
randomly
on 10 Aug, 2012 22:15
-
Because Tungsten has a very high density, much higher than lead, for a 'relatively' cheap material. Which minimizes the size of the weights. It's been used as keel weights in some America Cup sailboats for that reason.
Also it's very hard so it's not going to deform under high G loads like lead could and cause any problems when the weights are released.
The high melting point was not a consideration.
-
#170
by
randomly
on 10 Aug, 2012 22:32
-
Just a question about the RAD hardened processors they use on vehicles such as the MSL, is it true they are based on Sapphire rather than Silicon as I saw that quoted somewhere recently?
The technology is called Silicon on Sapphire. It consists of a thin layer of monocrystalline silicon that is grown on a sapphire substrate. Sapphire is used because the atomic spacing of Sapphire is very close to the atomic spacing of silicon and makes it possible to grow a monocrystalline silicon layer.
The actual transistors are all still silicon, but the sapphire substrate is an extremely good insulator and eliminates substrate currents from flowing around that occur in a regular bulk silicon wafer. This is extremely important for RAD hard devices because radiation causes these substrate currents that can flip transistor states.
SOS also reduces inter-element capacitances and results in considerably faster transistors than you would get otherwise. This is also important since RAD hard processors and memory use very large transistors to reduce the effects of radiation hits, and large transistors are slower due in large part to parasitic capacitances. Current Intel processors are built with a 22nm feature size, the RAD750 is built with 150nm minimum feature size. So transistors are at least 50 times larger than typical desktop computers.
Edit - 8/11/2012
The MSL computers apparently operate at 133 Mhz, not the often quoted 200 Mhz. The slower clock speed might indicated that these are actually fabricated with 250 nm feature size instead of the smaller 150 nm feature size. Somebody out of JPL would probably have to answer the question of what the process size actually is for the MSL cpus since all I can find is BAE RAD750 product data sheets and requotes from that and not what is actually in the MSL.
-
#171
by
Star One
on 11 Aug, 2012 17:26
-
@Randomly thanks for that information.
This new article about the computers onboard the MSL also quotes the 200MHz clock speed again for them?
http://www.spaceflightnow.com/mars/msl/120810computer/It mentions in here that BAE are currently working on a quad-core CPU which will be order of magnitude faster & more powerful than the current ones in use.
-
#172
by
iamlucky13
on 11 Aug, 2012 19:11
-
MSL came with 6 balance weights that were discarded in the early EDL phase. Why are those weights made of tungsten?
One property that comes to mind when talking about tungsten is its very high melting temperature (3422 C), but why is that useful? After they are discarded, what problems would they caused if they melted?
Is it that they want it to hit the ground as solid to make bigger impact? My understanding is that the kinetic energy delivered would be mostly the same - solid or molten.
Because tungsten is very dense. That means the physical dimensions of the weight are small, which means you have more latitude in where you can place it, such as out towards the edge of the vehicle, where it has the greatest moment arm, and therefore you can minimize the mass required for balance.
About rad hardened computers i wonder if you take a regular (OK, industrial) computer and wrap it with say 5mm of lead - how much rad hardening would that add, compared with the specially built rad hardened computers?
I don't remember the source right now, but I was reading a paper the other day discussing the effect of different types of shielding on the radiation exposure at Mars. One of the points mentioned is that lead is effective against solar particles, but not so much galactic cosmic rays...or perhaps I have that backwards. The result being that it can help, but not as much as you'd like.
-
#173
by
thomson
on 11 Aug, 2012 22:14
-
@randomly, @imlucky13: Thanks for your detailed answers.
Another 3 questions:
1. MMRTG is powered by Plutionium-238, which produces various types of radiation when decaying. Does it affect RAD in any way? One possibility is that characteristics of MMRTG radiation and any obstacles between MMRTG and RAD are well known, so they can be corrected/taken into account during interpretation of RAD readings. The other is that MMRTG is so well shielded that it does not leak any measurable amounts of radiation. So, how does it work?
2. MMRTG used as power source has minimum lifetime of 14 years. MSL primary mission is envisaged to last around 2 Earth years. I assume that even under most optimistic assumptions, drive system will not last 14 years, so MSL will lose mobility well before 14 years. But is it reasonable to hope that at least some useful data will still be produced in 14 years?
3.
http://nuclear.gov/pdfFiles/MMRTG.pdf states that US launched 26 missions including 45 RTGs. That means that some missions had more than one RTG on-board. Which missions had more than one RTG?
-
#174
by
Will
on 11 Aug, 2012 23:12
-
Is Curiosity going to examine its own crashed descent stage? In addition to getting us some really cool crash pictures, it would also present the opportunity to study some of that big pile of dirt the stage kicked up. Maybe the stage will have even dug some distance down into the ground, allowing subsurface sampling.
Asked at the second press conference. Answer: no.
And the main reason: risk of contamination from Hydrazine fuel
There's another reason that hasn't been mentioned yet. There seems to have been a sensible effort to ask where Curiosity intended to go for maximum scientific payoff, and try to put the discarded bits of spacecraft jettisoned on the way to the surface as far away from there as practical. So looking at the descent stage would also take Curiosity away from the most interesting local real estate.
-
#175
by
jnc
on 11 Aug, 2012 23:20
-
lead is effective against solar particles, but not so much galactic cosmic rays...or perhaps I have that backwards.
My
guess (but a high-confidence one) would be 'more effective against solar'. Solar stuff is produced in less-energetic reactions, and
the highest-energy cosmic rays are from extra-solar sources. So assuming the shielding works better against lower-energy stuff, that gives the 'more effective against solar'.
Noel
-
#176
by
cleonard
on 11 Aug, 2012 23:23
-
1) Pu-238 is the optimal fuel for RTG's. It requires the minimum shielding of any of the other possibilities. Usually the structure of the RTG is enough and no additional shielding is needed. Now some radiation does escape and I am sure that is taken into account as far as the RAD is concerned.
2) 14 years is a long time so who knows. I would be surprised if the electronics last that long. Actually the RTG's are likely one of the most reliable parts. It's not that they stop after 14 years. The power would have decayed enough that normal operations is no longer possible. The most important of those operations is running the heaters at night. No heaters and the electronics will die relatively quickly from temperature cycling.
3) Most deep space missions have more than 1
Pioneer 10 & 11 had 4 each
The Voyages have three each.
The viking landers had two each
Cassini has three
Galileo had 2
New horizons only has 1
-
#177
by
randomly
on 12 Aug, 2012 19:52
-
@Randomly thanks for that information.
This new article about the computers onboard the MSL also quotes the 200MHz clock speed again for them?
The BAE RAD750 product sheet says they can operate UP TO 200 mhz. It covers a number of different versions with speeds of 110 mhz to 200 mhz.
Ben Cichy of JPL says the MSL RAD750s are running at 133 mhz.
-
#178
by
randomly
on 12 Aug, 2012 20:27
-
About rad hardened computers i wonder if you take a regular (OK, industrial) computer and wrap it with say 5mm of lead - how much rad hardening would that add, compared with the specially built rad hardened computers?
It's not very effective unfortunately. Some Galactic cosmic rays can penetrate through Meters of lead. Radiation causes damage to the actual devices that is cumulative. It's essentially knocking atoms out of the crystal structure which among other things causes increases in leakage currents in the transistors.
As the radiation damage accumulates the CPU is going to fail at some point. So you want to design it so it lasts as long as possible. You want larger transistors so they can take more damage before they are compromised. You want larger threshold voltage differences and larger voltage swings so it's more tolerant of leakage currents and shifts in transistor threshold voltages from radiation damage. You want lower operating frequencies and so on.
All the technology trends that make current computers so fast, complex, and energy efficient are the same trends that make them even less radiation resistant. For space applications you really need to design computers that are radiation hardened from the ground up. Unfortunately that means you are not going to get state of the art processing power. It's an unfortunate tradeoff that has to be made. They are also very expensive since there is a lot of intensive engineering involved and so few units sold.
-
#179
by
randomly
on 13 Aug, 2012 00:48
-
@randomly, @imlucky13: Thanks for your detailed answers.
Another 3 questions:
1. MMRTG is powered by Plutionium-238, which produces various types of radiation when decaying. Does it affect RAD in any way? One possibility is that characteristics of MMRTG radiation and any obstacles between MMRTG and RAD are well known, so they can be corrected/taken into account during interpretation of RAD readings. The other is that MMRTG is so well shielded that it does not leak any measurable amounts of radiation. So, how does it work?
Pu238 decay produces almost exclusively alpha particles which are very easy to shield. A small fraction of the decays produce neutrons from spontaneous fission events, and there are some gamma rays produced further down the decay chain. Both neutrons and gamma are not so easily shielded against but the intensity is fairly low. The radiation from the MMRTG is a small fraction of the background radiation from GCR, probably less than 10%. Also the type of radiation and associated energy levels is known and can be accounted for in the detector.
2. MMRTG used as power source has minimum lifetime of 14 years. MSL primary mission is envisaged to last around 2 Earth years. I assume that even under most optimistic assumptions, drive system will not last 14 years, so MSL will lose mobility well before 14 years. But is it reasonable to hope that at least some useful data will still be produced in 14 years?
It's certainly possible. half life of PU238 is about 88 years so thermal power will drop to half in 88 years, electrical power will drop faster than that due to degradation of the thermocouples, but these things are incredibly durable. The Voyager RTGs are still operating after 35 years in deep space.
The MMRTG on the MSL charges up a battery which is used for driving and running the instruments during the day. If nothing else fails and the battery capacity doesn't degrade to much there is no reason you couldn't still be driving around collecting data 20 years from now.
3. http://nuclear.gov/pdfFiles/MMRTG.pdf states that US launched 26 missions including 45 RTGs. That means that some missions had more than one RTG on-board. Which missions had more than one RTG?
I think you can find that info here
http://solarsystem.nasa.gov/multimedia/downloads/Standard_RPS_Report_Final_011205.pdf