New Mars Forums

Elon Musk was born and raised in South Africa. He has definite South Africa attitudes and values. His mother Maye Musk was born in Canada in April 1948, her parents left Canada in 1950 when she was 2 years old. They moved to South Africa in 1952. Elon left South Africa at age 17 in 1989 because South Africa had mandatory military service aka The Draft. He started college at age 19 at Queen's University in Kingston, Ontario. In 1992 transferred to University of Pennsylvania in Philadelphia, age 21. Completed his bachelor degree in 1997, then moved to California. So he lived in Canada for 4 years in his late teens, and his mother left Canada age 2. Does that make him Canadian?

Canada/US relations have oscillated throughout history. Several times certain US politicians have attempted to use economic coersion to force Canada to join the US. This is not the first time. I could write a long-post with history, but you would find it boring. Let's just say others have tried to do what Donald Trump is doing. Every time, relations become strained, Canadians pull together to say we will never join the US. One of those times, not the first or the last, was in 1866, right after the US Civil War ended. Canada's response was to form a country. Canada became a country on July 1, 1867.

Yes, Canada has a problem with Quebec. The separatists in Quebec are a minority, but a loud-mouth minority. In October 1970, the FLQ attempted to use terrorism to force separation. They murdered a Quebec provincial minister. And they were planning to use explosives to bring down the tallest office building in Montreal. Police caught them with the plans and the explosives. They were arrested before they could place the explosives. Look what happened to the World Trade Center on 9/11; that's what the FLQ wanted to do in Montreal in 1970. But after the Minister was murdered, the War Measures Act was activated. Police got aggressive. Arrested them before the building could be bombed.

Yes, we have a problem that Alberta oil is taxed, and "equalization payments" are going to Quebec but not Alberta. That's an internal problem in Canada. After the Quebec referendum of 1995, the federal government decided to basically bribe Quebec to stay. But it's 31 years, it's got to stop. Equalization is supposed to be for young provinces that don't yet have the resources to provided the same services  as the rest of Canada. But the two big provinces when Canada was founded were Ontario and Quebec. They were still big when I was a child in the 1960s and '70s. But after the first referendum in 1980, many companies left Quebec. They left out of fear that Quebec would separate, so companies moved to other provinces to ensure they remain in Canada. That happened again during and after the second referendum in 1995. Quebec has no one but themselves to blame. They're still the 2nd largest province in population, after Ontario. There's no way anyone could justify equalization payments for any of the 4 largest provinces: Ontario, Quebec, BC, and Alberta. Quebec is getting the largest sum by far, and it's got to stop! Saskatchewan, Alberta, and BC get nothing, Ontario gets a tiny bit. Payments to Ontario also must stop. Those provinces can collect their own taxes to pay their own bills.

As for Globalism, Democracy and coordination is a good thing. But some corporations want to run the world. Corporate executives want national governments to become irrelevant. That isn't going to happen.

Elon Musk is someone who was born and grew up in South Africa. His mother grew up in South Africa. He moved to California.

I need a generic technology thread. Or do I create a new thread?

Latest video by Felix of What About It: SpaceX Starship Flight 14 Lost Way More Parts Than We Thought!
Second half of the video mentions Elon Musk wants to build an AI data centre in space. Satellites with greater solar panel area than the International Space Station. Each satellite would house a few blades of circuits, each with multiple processors. Each satellite would also have a large radiator to use space itself for cooling. These satellites would be in polar orbit, so receive sunlight 24/7. And the orbits would adjust, precessing as Earth orbits the Sun. So the orbit would complete one complete precess rotation per year. This means it tracks the Sun so all satellites are in direct sunlight 24/7. Satellites would communicate with Starlink, not direct communication with Earth.

I think this is dumb. Launching such a satellite requires a lot of money. Even with reduces launch cost of Starship. Service is not possible. When (not if) processors fail, they cannot be replaced. In space, the electronics must be hardened to withstand space radiation. Not necessary on Earth. Building a data centre on Earth appears a much MUCH more practical design.

One issue with a data centre on Earth, is cooling. These data centres have been designed to use evaporative cooling. That is, cooled with air conditioners. Like any air conditioner, there are two parts. The evaporator inside the building, where a coolant such as freon evaporates from liquid to gas absorbing heat as it does so. The other part is outside: the condenser. There the gas condenses to a liquid, releasing heat as it does so. Pressure of coolant in the condenser is higher, causing it to increase in temperature due to thermodynamics. That releases the heat. Pressure of coolant in the evaporator inside the building is lower, causing cooling due to thermodynamics. But data centres do not consider this to be enough. Clean water is sprayed on the outside of the cooling fins of the condenser outside the building. This water sprayed on hot aluminum alloy fins will evaporate into the environment. As that water evaporates, it cools, taking heat away. This consumes a lot of water.

This is not necessary. It's a design for a dry hot environment like Arizona. Some place that has a flowing river, you can have river water flow across an aluminum alloy heat exchanger for the condenser. The freon or whatever coolant is used for the air conditioner is contained. Instead of spraying clean water on the fins, river water flows across the fins. River water need not be clean, it can be river water. The river water will be warmed, but not boil. This will warm the river, but it means clean potable water is not consumed. The river is not consumed, it is just warmed.

If there is no convenient river, then the data centre could be built on the shore of a lake. The lake water could be pumped across the aluminum alloy cooling fins. This would warm the lake water. Or coast of an ocean. Ocean water can be pumped across the cooling fins. Again, ocean water is not consumed, it is just warmed. Nuclear power plants have been doing this for decades. It's not new. A simple way of building a data centre without consuming water.

By the way, ammonia is a more efficient coolant than freon. The first refrigerators were built with ammonia. The problem was one woman was asleep in her bed when her house refrigerator developed a leak. While she was a sleep, her house filled with so much ammonia that it killed her. She died in her sleep. Freon is clear, colourless, odourless, and non-toxic. If there's a leak, it's not dangerous. A refrigerator using freon consumes a little more electricity than one using ammonia, but it won't kill you. Then the fear that freon will destroy the ozone layer. Freon was replaced by something even less efficient, but will not kill you, and will not destroy ozone. I don't know what they're using now. But expect a data centre will use ammonia. Ammonia will not destroy ozone, just ensure no one sleeps there.

As the arctic thaws, more and more of the arctic will be ice free in summer. And for an ever longer period of time each summer. That will make all the resources of the arctic ocean accessible. And large ships can traverse the Northwest Passage.

For reference, here is a record of global temperature calculated from ice cores from Antarctica. Blue is the EPICA ice sheet, green is the Lake Vostok. Peak during the last interglacial roughly 125,000 years ago is 6°C above the pre-industrial temperature. 260,000 years ago was two interglacial periods ago; for some reason the interglacial was cut short so was shorter time and didn't reach the same peak as other interglacials. About 370,000 years ago was 3 interglacials ago; it reached a peak of 5°C above the pre-industrial temperature. They found an ice deposit even deeper, extending 800,000 years ago. There's an ice age roughly every 100,000 years. Roughly 80,000 years of ice age, and 20,000 years of interglacial. So the 800,000 year ice core records 8 ice ages and 8 interglacial periods. The long term trend is each peak is higher than the last. The point is just due to nature, the peak temperature will be warmer than today. Do not expect the peak to be limitied to 1.5°C above the pre-industrial temperature. More likely it will be 6°C, the same as the last interglacial period. Good news is that peak is not due for another 2,000 years. That does not mean it will happen all at once. It means slow but steady global warming for the next 2,000 years, with each summer setting new record highs.

During the last interglacial period, the north polar ice cap completely melted during the summer. At the warmest period. It will again.
(Image from Wikipedia. Click image for Wikipedia page.)

YouTube: How US aircraft carriers resupply in the middle of the ocean (the hidden dangers will shock you)
Video of at sea resupply of USS Gerald R. Ford, the latest US aircraft carrier. Video is 9 months old. 14 minutes long. Narration is dramatic. Shows how to transfer a pallet at a time of supplies from a dedicated purpose-built US Navy supply ship to the carrier. Includes sending a fuel hose to transfer aircraft fuel for aircraft that operate off the ship. The Navy has been doing this for multiple decades. Again, there's no need for the carrier to sail to Diego Garcia.

Again, narration is overly dramatic. Doing this near the Strait of Hormuz would not require the ships moving at maximum speed. They could slow to a crawl. Just barely moving enough so the ships could manoeuvre, to maintain correct distance between ships.

Maybe I shouldn't weigh in on Iran. But I was thinking, why is resupply so difficult? Last month there was concern that the USS Abraham Lincoln was short of food while deployed in the Strait of Hormuz. The base in Bahrain was damaged. But every Nimitz class aircraft carrier has 2 C-2 Greyhound cargo aircraft. Yes, limited range. But a C-130 Hercules, or C-17 Globemaster III, or some other cargo aircraft could carry supplies to a base in the region. Then transfer supplies to a C-2 Greyhound, to ferry those supplies to the ship. Isn't that what the C-2 are for? The US has multiple bases on the region. There are 3 air bases plus an airport in Oman where this could be done. If nearest supplies are Diego Garcia, that's 4,136km (2,233 nautical miles) one way. The C-2 Greyhound can only fly 1,300 nmi (2,400km) while carrying 10,000 pounds (4,500kg) of cargo. One way. But distance from Strait of Hormuz to Muscat International Airport is 398km, depending where in the Strait the carrier is located. Definitely can be reached. If not that airport, there are others.

This is basic logistics. Isn't that a basic part of the job for any officer of rank Colonel or Navy Captain or higher? Secretary of Defense shouldn't even have to address this himself, there should be staff to do this.

Good thing Starship deployed. Starship is going to have to earn revenue soon. So far it's paid by SpaceX investors. They got money from NASA for the Lunar lander, but no serious progress on that. And some profits from Falcon launches. But Starship is mostly funded by investors. StarLink gets its money from subscribers; more subscribers means more money. But more subscribers also means more traffic. StarLink is already slowing down due to traffic congestion. So they need more satellites. Starship just deployed so many satellites it's equivalent to several launches of Falcon 9. Subscribers revenue from StarLink will help.

One mission for Starship is Earth-to-Earth. If they want to carry passengers, they have to stop blowing up. Yes yes, I know. Product development. But Elon Musk announced his Interplanetary Transport System at the 67th International Astronautical Congress in Guadalajara, Mexico, on September 27, 2016. It's now 10 years. How many Starships have blown up? On paper Starship sounds good. And it is slowly making progress.

Saturn rockets were studied 1960-1962. Saturn C-5 was selected on January 25, 1962. First all-up unmanned test launch was Apollo 4, November 9, 1967. First crewed flight was Apollo 8, December 21, 1968. From selection to first manned flight 6 years and 11 months. If you include the early studies, then 8 years. First paper study of Saturn A-1 started 1959, but you can't count that because that was so much smaller than Saturn V it's like comparing Falcon 9 to Starship.

So yes, I do expect Starship to launch, deploy payload, and catch both stages without blowing up. Hear me Elon? Next launch! Double catch!

Thank you for the email, GW. I received your spreadsheet. Still haven't been able to calculate Mars approach speed. One orbital calculator assumes Hohmann transfer. This is an express trajectory. As Dr. Zubrin said, using roughly 10% more propellant to reduce transit time to 6 months. Increasing speed (reducing transit time) increases propellant exponentially, so faster than that is not practical. And a 6 month transit happens to be the free return trajectory for Mars. A pure Hohmann transfer takes 10.5 months. Reducing transit time to 8.5 months takes a trivial increase in propellant; basically within error margin. Reducing to 6 months takes more. Reducing still further becomes not practical.

But how to calculate Mars approach velocity with 6 month transit? Then descent into Mars gravity well to aerocapture altitude increases velocity further. Google AI can give me those numbers, but AI makes mistakes. Trying to calculate without AI is difficult.

Yeah, I remember my dad flip out when gas went above $1 per gallon. In Canadian dollars, and Imperial gallon. 1 i.perial gallon = 1.2 US gallons (1 and 1/5). If you want to get picky it's 1.20094993, but close enough. Still.

I remember filling my car when it was 23¢ per litre.

Sorry. Tomorrow did. I appreciate the work. Trying to get a design that works.

TMI is going to be another issue. I posted of the possibility of nuclear fusion, and not nuclear thermal, but direct fusion thrust. It's  a long LONG way away. But.if that works, it fundamentally changes a lot of things. Including orbital insertion.

And yes, Dr Zubrin has a Ph.D. in Nuclear Engineering, only a Master of Science in Aeronautics and Astronautics.

So the next step is to verify the arrival velocity. I used Google AI. NASA has tables with Mars trajectories. For an express trajectory that takes 6-months one-way from Earth to Mars, what velocity does it arrive? And realize falling into the gravity well of Mars will add velocity. So the question is velocity when it first hits the Mars upper atmosphere.

Thanks Gary. I started by using Google to look up periapsis velocity for a highly elliptical high Mars orbit. The numbers I got were 5 to 6 km/s. So I used 6 for minimum deceleration. But when working out exact apoapsis altitude and orbital period, I discovered that velocity is hyperbolic. Mars escape velocity is 4.97 km/s. Rather than use 4.95 as barely in orbit at all, is used 4.85 as a small safety margin. So we have to be careful when using Google AI; it makes mistakes. Slowing from. 7.5 to 4.85 is more deceleration than I expected.

I'll work on a more precise mass estimate. It'll help. If we can get mass down, it'll help with orbital insertion.

I asked again, but with effective nose radius of 38 metres.
This time it used Sutton-Graves Constant for Mars (Csg): Approximately 1.89 x 10⁻⁴ kg^(0.5)/m
(Sorry, no superscript decimal point.)

After changing exit velocity from 6.0 to 4.85 km/s, the question was:
"mars aerobraking from 7.5 km/s to 4.85 km/s at altitude of 80 km would produce how much heating? With effective nose radius 38 metres. Expected deceleration profile for this pass."

It then offered expected deceleration profile for this pass.

Again edited to reduce final velocity from 6.0 km/s to 4.85 km/s.