According to Emily's article, all the mission objectives will be met and
"...the Earth-to-Vesta cruise duration is shortened by two months with
the two-month delay to a September launch, resulting in an arrival date
at Vesta that is the same -- October 2011 -- regardless of whether Dawn
launches in July or September."

Thanks, missed that when I was reading the article.

Great -- but how? What precisely is the mission design / trajectory change that allows this to happen? Will there still be a Mars flyby/gravity assist? Will they burn more fuel? Less? Will it hamper extended mission capabilities? Why isn't this information available somewhere -- I've looked and can find nothing.

I think I read somewhere that the actual ion thrust period starts some 60-80 days after launch so there's plenty of margin there. Part of the additional margin might be supplied by the Delta II launch vehicle, too. Ion engines are weak but they sure provide greater flexibility than chemical ones.

ONCE they have had operational experience in space, they will know the REAL flight power output levels of the solar panels and the REAL specific inpulse <pounds (force) of thrust for so many seconds per pound (mass) of propellant> of the ion rockets.

You HAVE to be able to fly the mission with an unfavorible launch during the worst part of the launch window, large launch errors, poor solar panel performance, and poor ion drive performance.

If everything is nominal, you are going to have flight reserve capability. or be able to handle 'Uh-oh"'s that develop during the mission.

If things work generally better than nominal, ooooh!

The fact that those solar panels and ion engines need to be tested out in flight to determine their actual performace margins is the reason that the timelines, post-launch, are somewhat vague. The basic plan will be the same -- Mars flyby followed by Vesta and then Ceres -- regardless of the launch date, but I've been told that the Vesta arrival date, for instance, can change by as much as a couple of months from the nominal one if the engines perform significantly better than their design requirements. They'll just have to wait and see how fast a ship they've got once it's set sail! This is also the reason they're being cagy about whether they'll be able to do any other asteroid flybys, however distant. Until they know how well their engines are performing, it's kind of hard to figure out which asteroids will get close enough to the trajectory, and when, for them to be observed.

--Emily

OT here, but I find it intriguing that the engines provide so much flexibility in terms of arrival times despite their incredibly small delta-v in comparison to chemical propellants. It seems that even minor improvements in this technology will reap serious benefits for UMSF.

Pity that Xe has the highest atomic weight of the stable noble gases. I know I'd be shot for asking this in some circles, but wouldn't it be nice to use radon with its big, fat average atomic weight of 86 vs. Xe's 54...? (That stuff has to be good for something, after all... )

True. Pity, though, about that 3 day half-life...

Mercury (the element, not the planet) is nicely volatile, relatively easy to ionize and weights in
at a nice, fat 200 atomic mass units. Unfortunately, though, it's tendency to contaminate surfaces has
precluded its use in ion thrusters.

-Kevin

" incredibly small delta-v in comparison to chemical "

Ion rockets have enormously large delta-v capability because of their high specific impulse <bounce per ounce>. The killer is that they have horrendously low thrust... the force is small, but they difference is they keep going.. and going.. and going..........................

I wonder about the use <and cost per pound> of Krypton or even Argon. You get less bounce per ounce, but JIMO, for example, somebody said, was going to use more or less the entire world's supply of xenon. If simply lobbing something on an escape trajectory cost less, it might be cost effective to use more of a lower atomic weight ion. What are current tradeoffs on different gasses for ion rockets?

I think later on we found some more information that showed the magnitude was off by a factor of 10^4 or so. So there's plenty of Xenon gas for space exploration (at least for now).

I have been told hydrogen would actually be the most efficient ion fuel - If it could be stored more efficiently at high density. I think it has to do with the efficiency of acceleration mechanism - the amps per unit of thrust ratio. I'll try to find out more...

The way the rocket equation works, the most important thing in being efficient is the speed of the exhaust, not mass expelled per second. Just as chemical rockets try to achieve high temperatures and low molecular weight of their exhaust, the same would probably apply to ion engines. The trouble here would be as you say in the acceleration mechanism. Practical difficulties using a low molecular mass gas likely greatly offset the theoretical advantages it would have.

Hmm. Thanks for the enlightenment & corrections, all.

So, then, the ideal fuel would be an intermediate weight element that is particularly amenible to electromagnetic acceleration & magnetic vector control? Would something like Fe ions be the best (disregarding the horrendous difficulties of vaporizing the stuff)?

All other things being equal, the specific impulse of an ion drive using Xenon is about 11.5 times worse than one using hydrogen, BUT the thrust is 11.5 times better. Since even with Xenon (worse Isp), such a thruster has 10x the Isp of a conventional rocket BUT even with Xenon (better thrust) the thrust is so low as to barely be usable, it seems clear why it's currently the fuel of choice for ion drives.

To work this out, note that for a given energy (and assuming just one charge on the ion), thrust varies with the square-root of the mass of the ion, while specific impulse varies inversely with the thrust. Xenon has an atomic mass of 132 (not 54 -- that's the atomic number), and sqrt(132) = 11.5.

Since the mass ratio varies EXPONENTIALLY with the specific impulse, and exp(11.5) = 100,000, this does say that we could do fantastic things if we could just get enough energy to make a hydrogen ion engine practical. (And solve some other technical problems, of course.) :-)

--Greg

"...I think later on we found some more information that showed the magnitude was off by a factor of 10^4 or so..."

What's 4 decimal places between friends...?

You beat me to it. But can we assume one charge per ion? Surely not: it's easier to knock a few electrons off a Xenon atom than it is to ionise Hydrogen.

Andy G

Actually not.
It is easier to knock one e from Xe than H (first ionization potential for Xe is 12.1 < 13.6 for H) , but you need ~21 and ~32 eV to kick the second and the third e from Xe+ and Xe2+.

Thanks, TheChemist - you're quite right. I withdraw the remark I made earlier!

Andy

The last update by Marc Rayman, apparently, first on Space Daily:
Stand Down At Dawn Launch Pad

As seen on TPS's blog yesterday.


Doug

Chemist: Perhaps you can confirm something for me. I was surprised that the ionization energy for hydrogen was so close to the energy for Xenon, but then I thought that perpaps a larger problem would be breaking the H2 molecular bond. However, the enthalpy of H2 seems to be only about 217 KJ/mole -- just a fraction of the 1312 KJ/mole for ionization.

http://www.stanford.edu/~cantwell/AA283_Co...mochemistry.pdf

(See equation A1.33).

Am I reading this correctly? That seems awfully small somehow.

Now, contrast that with the energy required to accelerate 1 mole of Xenon to 30 kps. I come up with just under 60,000 KJ/mole. So the ionization cost seems to be just a couple of percent -- asssuming 100% efficiency, of course.

--Greg

Greg,

Xe is a monoatomic molecule, while hydrogen is a diatomic molecule H2.
The value of 1312 KJ/mol is for the ionization potential of a hydrogen atom, H.

H --> H+ plus e-

To use hydrogen for fuel, I guess you would have first to break H2 to atomic H (I have not looked into the specifics of ion engines so this is off the top of my head).
I hope this helps.

The ionization potential for molecular hydrogen is 15.4 eV/molecule (pay for ref here):

(Handy website for energy unit conversions: http://www.volker-quaschning.de/datserv/fa...n/index_e.html)

H2 --> [H2]+ + e-

15.4 eV/molecule x 6.022E23 molecules/mol x 1.602E-22 kJ/eV = 1486 kJ/mol.

Your typical MS system does the "ping and fling" thing to ionize molecules, then accelerate them. You wouldn't necessarily need to break apart molecular hydrogen in order to accelerate them for thrust.

-Mike

Thanks guys. I suppose I'm being greedy and wanting to ionize both hydrogen atoms, but it's clearly enough to just ionize the molecule.

Other than the thrust problem, of course, I suspect there are erosion issues. I note that some earlier systems used Cesium (VERY easy to ionize) but the Cesium corroded the equipment too fast. However the thrust problem appears to be serious enough that they don't consider using Argon or even Krypton as a fuel.

--Greg

You know, I've heard about this equation since reading Heinlein, but for some reason have never seen it. Would you be so kind as to post it, UG? Tired of being ignorant, here....

I have enjoyed the dicussion on ion engines and different fuel choices and impulse outcomes. The use of hydrogen gas as a fuel would certainly kick the impulse of an ion engine through the roof and the ionisation of hydrogen is not really a problem.

In chemical engines hydrogen gives high impulse for two reasons
1. the energy content (combustion)/ unit of mass is very high, 5X that of petroleum fuels
2. transforming heat from combustion to kinetic energy is very efficient with very small molecules

This is also the reason that hydrogen engines run fuel rich, the addition of so many small molecules (H2)improves the impulse even though they are not burnt

Unfortunatetly neither of these can help us with ion engines. We can not extract the energy of combustion that can potentially be extracted in chemical engines and they do not rely on this mechanism of transforming heat into kinetic energy either.

In ion thrusters the limiting factor is the amount of energy available. This electrical energy is converted into kinetic energy, so here is where the difficult decisions need to be made.

Lets assume we have a choice of accelerating 1 unit of mass of fuel to 30000m/s typical in todays ion thrusters or to to 100,000m/s in some engine in our dreams. Great we have just increased our impulse from 3000 to 10,000. To infinity and beyond as Buzz would say.

Problem, you need 10X the energy to do that, (remember Ek= 1/2 mv*2) or with the same energy supply we can only do that to 1/10 unit of fuel. Unfortunately 1/10 of the fuel accelerated to 3X to velocity will only give you 1/3 of the thrust. Remember that thrust comes from consevation of momentum P=m.v (so 1/10 * 3= 1/3)

Damn that inconvenient maths.
We have just gone from a mouse's fart in thrust to a crickets fart but the fuel will last almost for ever

So Xenon is chosen quite deliberately to give the maximum thrust from the available power, same reason that solid rockets give the majority of the thrust at launch in chemical rockets


So don't forget that famous rocket engineers saying without the impulse you wont get anywhere but in the end its the thrust that makes the bacon
PS I don't know if its famous rocket engineers saying, I just made that up

We all have to remember that the use of rocket propulsion for spacecraft falls into two very distinct categories.

To launch from a planetary surface or out from a low orbit requires sufficient thrust to overcome the relentless pull of the body deep into whose gravity well you are located. In this situation, impulse takes a back seat to thrust; you need to have enough thrust to counterbalance gravity, or else you're never going anywhere. The Saturn V needed to develop more than seven million pounds of thrust not primarily because of the required specific impulse, but because the entire fueled rocket weighed more than six million pounds. It needed the short-term thrust required to counter-balance the extreme weight of the vehicle and get it moving out of the bottom of Earth's gravity well.

However, once you are in a trajectory that does not intersect any other solar system body, specific impulse reigns supreme. Now you're in the world of Newtonian physics, in which your spacecraft continues to move along a vector, losing little speed, and doesn't need to fight to keep from falling back down into a gravity well. As long as you work *with* gravity and not against it, the thrust needed to overwhelm the effects of a gravity well just isn't needed. Now is the time when you want to max out the specific impulse of your engine, since even a very low thrust engine can apply a considerable amount of acceleration to your spacecraft if you let it continue to run for weeks or months.

So, the decision becomes pretty simple -- you need high-thrust engines while working in and close to gravity wells, and you don't end up caring as much about the efficiency (i.e., specific impulse) of your engines. But once you get out into space, you're better off using low-thrust, high-impulse engines. It's not necessarily intuitive, but that's how it works.

-the other Doug

No, wrong. The specific impulse can be 1,000,000 but that will not get you to your destination unless the thrust is enough in the time you have available. Noone is going to wait 10 years to get to Vesta. It is a compromise between thrust and specific impulse for the powers source available. The principle is exactly the same for ion engines as for chemical engines, the only difference is a few orders of magnitude.

If specific impulse reigns supreme then why do they use Xenon instead of hydrogen

It reins supreme in terms of efficiency. I never said it got you there faster, it just gets you there with minimum spacecraft weight per m/sec of delta-V. And I dispute that no one will wait ten years to get to Vesta -- though this isn't a method you'd use when you transport people. For people, who need to dash from one radiation-safe spot to the next, you need to opt for speed and deal with the incredible hit you take in performance.

And it has been discussed in detail how much more energy it takes to ionize hydrogen than xenon... we're also talking about what can be done feasibly with our current technology.

-the other Doug

Try this: Tsiolkovsky rocket equation. Notice the delta V is proportional to exhaust velocity with fixed mass ratios, there is no time variable in the equation.

Simple yet powerful; thanks, UG!

The choice of xenon over hydrogen has very little to do with the ionisation energy, 15.4 eV vs 12eV is inconsequential when compared to the energy needed to accelerate the ions. It also needs to be remembered that if one looks at ionisation energies/ mole and lets assume they are roughly equal for H2 and Xe. Then the ionisation energy/unit of mass is 65X as high for H2 as Xe because of their different atomic weight. The ionisation energy / gram if far more important in calculating efficient use of power than ionisation energies/ mole


The limit of current technology has as much to to with the power source as anything else, that was the point I was trying to make.

All the early work with ion thrust was done with caesium and mercury before they switched to xenon. So what have they all got in common? High atomic mass and the ability to be handled as fluids at a convenient temperature. They were after maximum thrust per kW. Cesium and mercury were abandoned because of electrode erosion. Maybe hydogen has a similar electrode erosion problem as well.



If that were our primary consideration then we should design a thruster that runs on nuclear energy and uses a high efficiency light source to expel photons at the speed of light. Our impulse goes up to 3x 10*7. Maybe that will be possible one day but to get any meaningful kind of thrust we are going to need an awful lot of power.


Time constraints are however a practical consideration as I have outline above, no scientist wants to expire before his experiment in complete

The Kennedy Media Archive now has photos of Dawn being de-stacked and returned indoors.

--Emily

Agreed, but this has nothing to do with what the rocket equation says and yet you were replying to my remark on it. Yes, thrust is important in certain regimes, but in other regimes the actual efficiency of the engine is more important. That's the gist of what the other Doug is trying to say I think.

The unfortunate sight of seeing Dawn back in its processing facility is made up for by the photos of Phoenix now in place at pad 17-A.

I did wrap the reply to your post with that to Doug's and I suppose it should have been kept separate.

I cant resist a comment on your last post and that it to say that "efficiency" is not the same as "impulse". I would venture to say that the hydrogen Shuttle main engines may in fact be more efficient than ion thrusters. Efficiency in engineering terms is the fraction of the input energy converted to the desired output energy. For the Shuttle main engines this is approx 50%. I cant find any accurate figures for ion thrusters but it would surprise me if it were much higher than this. Again the barrier we come to is the limitations on the power sources used today.

Ion thrusters have a couple of efficiency parameters. The electrical efficiency (beam power, or
half mdot v squared, divided by power in) is often 90% or more. You waste a little bit in ionizing
the stuff in the first place. There is also a propellant utilization efficiency - again typically over 90% -
some propellant leaks out un-ionized, and some must be expended in a neutralizer to balance the
charge flux.

Of course, there is the question of propulsive efficiency - ion thrusters are great at putting power
efficiently into the kinetic energy of the beam, which is not the same as adding kinetic energy to
the vehicle. That efficiency is maximized when the vehicle speed approaches the exhaust
velocity....

Found some information.
Thrust efficiency appears to range from about 40% to 75% depending on Impulse with about 60-65% at Isp 2500
http://www.engin.umich.edu/dept/aero/space...IAA-96-2973.pdf

Now that Phoenix is safely on its way, September 26 is being eyed as the new launch date for Dawn...

Slightly OT, but I have reprocessed Hubble's 1994 Vesta FOC image (well, a series of images, but only in one band and of one face). While there are certainly equally good (if not better) images, most of them are in the infrared or visible spectrum, while this is in the near UV. And I don't care what the silly scientists say, if you look at this one closely, it is obvious that Vesta is really a giant sculpture of a smiling pig.

Leaning Jowler?

http://en.wikipedia.org/wiki/Pass_the_Pigs

Nice, Ted! I haven't seen much done with FOC. Also I forgot to tell you how great the Miranda dark side pics were.
Phil

Thanks. I also made this composite of the "Big-3"



Ceres and Vesta are HST FOC images, Pallas is from WFPC/2 (It is a super-resolution view, but the images are not from the PC chip).

Nice one of Ceres. I'll have to compare it with the color set that appeared recently.

Phil

Really nice composite of the "Big-3".
I've never seen a resolved picture of Pallas before. It seems quite spherical (another dwarf planet ?)

Marc.

That Ceres image is beautiful!

Looking at those pictures of "the big 3" reminds me of this Wikipedia illustration showing the outlines of the first ten asteroids against the outline of the Moon:

http://en.wikipedia.org/wiki/Image:Moon_an...ids_1_to_10.svg

I remember being told that an extended mission to 2 Pallas was infeasible due to the difference in orbital inclinations, but I wonder if one to 10 Hygiea (the 4th-largest asteroid) has been discussed. Anyone know?

--Greg

Hygiea is certainly much closer to the ecliptic (4 deg. inclination vs. 35 deg. for Pallas). However, it also lies farther from the Sun than Ceres, in an orbit that is particularly eccentric for a large asteroid, so unless there is the great chance that a trajectory would take Dawn to the near-perihelion node of Hygiea's orbit at the time that Hygiea happens to be there, this option would seem to be a delta-v "hog", too. It would also mean that a relatively quick flyby would be the payoff if such a trajectory were possible at all.

It would be nice to see up-close, but orbitally speaking, it's a tough target as asteroids go. I'm sure that any trajectory there would be in exchange for more than one flyby of easier targets.

Launch update:

Date: Sept. 26 (No earlier than)
Launch Time: 7:24 a.m. EDT

At least the afternoon T-storms are obviously not an issue...