Hacker News Reader

Comment by trhway | Hacker News Reader

napkin for ion drive powered by solar - tech available today (and nuclear as the power source for the ion drive instead of the solar - tomorrow for Mars and beyond) - looks much better than chemical even for Mars. As SpaceX already using ion drives on Starlinks, and the SpaceX will be the one going to Mars, i'd expect that it will be ion drive.

View on HN

Solar electric propulsion has great specific impulse [i.e., efficient use of propellant] but very low thrust. You're forever spiraling away from and then down into gravity wells.

It's terrific for cargo, but not viable for crews.

not forever. You can get to Mars faster than with chemical, something under 90 days one way in some reasonable configurations.

What does that look like? A ten ton ship including 100 square miles of solar panels?

You need about 10KWt per 100kg, i.e. 50m2 of panels. For 10 tons it would be under 5000m2, ie. like 70x70m. The panels do need to be relatively thin. Less than half the mass - 4 tons is the expelled mass (using say those NASA drives with 3500 Isp). And if you get panels at 0.5kg/m2, you'd get to 15km/s in under 2 months with about 3.5 tons for the rest of the ship including payload. More panels - either thinner ones or with less payload - faster time to Mars. Nuclear with ion drive - even better (and the only choice beyond Mars), can get you 35km/s+ (Voyager) or like 70km/s+ with 2 stages in under year.

You've definitely missed some decimals somewhere and/or are not considering the amount of thrusters needed. Ion thrusters total thrust is measured in the millinewtons and they have a very poor thrust:weight ratio (which does not account for the energy source mass). They're also relatively unreliable making them generally unfit for round trip missions without excessive redundancy which further increases the mass and thus the number of thrusters needed. Almost like a sort of microcosm of the tyranny of the rocket equation.

7KW NASA drive weight about 10kg. At 10KW it would expel about 8milligram/second for 0.3N thrust (even with paltry 50% efficiency). It will get ship of 100kg total starting mass to 15km/s in under 2 months spending under 40kg expelled mass. The arithmetic is very simple. (my favorite arithmetic here is 3 stage nuclear powered ion drive ship with increased voltage resulting in ~35000 Isp and thus getting 1000-1500km/s in 100 years, and thus 1000 years fly-by to Alpha Centauri, with the tech basically available today :)

>They're also relatively unreliable

Given their simplicity i think it is among the most reliable pieces of tech around. 5 years non-stop - no issues :

https://www1.grc.nasa.gov/space/sep/gridded-ion-thrusters-ne...

Note that all this is just NASA so far without SpaceX/Musk applying their engineering magic yet which i think would significantly improve the mass, efficiency, etc.

Mass = 10,000kg (per the initial topic)

Force = 0.236N (per your link)

Acceleration = 0.0000236m/s (acceleration = force / mass)

Seconds in 2 months = 5259600

Final Velocity after 2 months = 277 miles per hour (0.124km/s)

---

This is an extremely naive and optimistic calculation since I'm assuming the engine is weightless and does not require fuel or power. It's easier that way and still emphasizes the point perfectly fine. You're gonna need a lot more engines to get anywhere. And of course a 10 ton ship is not practical. The goal for Starship is to to have a payload in the hundreds of tons, let alone the entire vessel's mass.

You also only have half the distance to accelerate, because you then need to turn around and decelerate the remaining distance. And we're also ignoring landing/takeoff which obviously isn't happening.

>Mass = 10,000kg (per the initial topic)

>Force = 0.236N (per your link)

>I'm assuming the engine is weightless and does not require fuel or power.

10KW 10kg engine plus 40 kg expelled mass per 100kg total starting mass. Thus for 10000kg - 1000kg of 100 x 10kg 10KW engines, 4 tons expelled mass, 2.5 tons - solar panels providing power. 1 engine - 0.3N, 100 engines - 30N. 15km/s in under 2 months, plus 2 months deceleration. The same for 100ton ship or 1000ton.

>landing/takeoff

that is by local chemical Starships "shuttles" which would transfer people and cargo to/from the interplanetary ships. Using Starship shuttles to fuel the interplanetary Starship, though chemical one, has long been in the published SpaceX/Musk plans.

> Nuclear with ion drive - even better (and the only choice beyond Mars)

I'm sure you'll be interested in this concept.

https://iopscience.iop.org/article/10.1088/1538-3873/ac4812

And this related one.

https://www.frontiersin.org/journals/space-technologies/arti...

i'd say i hope for the hybrid - nuclear (especially fusion once we get to it) generates power while something like in your second link collects the mass for expelling from the drive (though not exactly "collects" - more like drives it through the electrostatic field, ie. using the tech like in your link in the role of the ion drive).

The nuclear can work far beyond Solar system while the running out of the expelling mass is the main limiting factor here. If we can collect the mass along the way i think we can get to Alpha Centauri with 2-5% of "c" instead of 0.3-0.5% i mentioned in the other comment here (and we can accelerate to even higher speeds on the distances larger than the one to Alpha Centauri)

?????? You definitely gotta try to explain that one

I expect the best way to use ion thrusters would be to combine them with chemical engines.

Use chemical engines for the high thrust trans-Mars injection burn so you can exploit the Oberth effect, then use ion thrust to continuously accelerate the ship for a (modest but still helpful) reduction in travel time.

https://en.wikipedia.org/wiki/Oberth_effect

Reply on news.ycombinator.com