There was a bit of discussion about dry launch in the space blogosphere in the last day or two. It seems to have started with Jon Goff's piece at Selenian Boondocks, which Clark Lindsey picked up and expanded on (see the "Fueling a Space Town" post), and was followed up with a post on agile space development by Dan Scrimpsher.

This is an important topic, and I wish that there was some sign that the new management at NASA is paying attention to it.

I would also add, as a response to the commenter who asks in Jon's comments section, why deliver propellant that has to be transferred as a fluid on orbit, rather than easier-to-handle propellant tanks? It's because delivering tanks doesn't offer the possibility of refueling them on orbit, so they'd only be single use. And in-space refueling is a critical technology in becoming a truly space-faring civilization, and the sooner we get on with developing and becoming comfortable with it, the sooner we'll reach that desirable (at least to me) destination.

[Update at 9 AM EDT]

I was imprecise above. As Paul Dietz points out in comments, delivering tanks doesn't preclude the possibility of refueling them later, but that wasn't what the the commenter was suggesting. What I should have said is that it doesn't advance us toward that (in my opinion) worthy goal, and it was clear from the commenter's question that he didn't have in mind tanks designed to be refueled (and it is a significant design issue).

[One more update]

I should have written "...preclude us from refueling from them later," to respond to Paul's most recent comment about mischaracterizing what he said.

Delivering propellant in disposable tanks doesn't preclude the possibility of transfering that propellant to other tanks.

No, it doesn't preclude that possibility, Paul, but if you read the comment, it's clear that that wasn't what was intended. The commenter was trying to avoid propellant transfer.

In our pioneering days here on earth the primary impetus in getting the Conestoga wagons moving was a promise of a better life i.e. “You’ll get rich boy.” Now most of those promises were false and convincing a group of investors to put down their money on any given venture will require some structure just as was needed to get the airline industry off the ground. I see this as the only role NASA should be playing in space until that infrastructure has been established. Of course that means including politicians. They are a necessary evil in this endeavor. Since they only do things that serve the public good with no though of aggrandizement, the problem is guiding them to the cause. A bureaucracy like NASD does not turn and move in a different direction on its own. It is prodded by those in power. An inefficient system, but it’s what we’ve got.
The business plan needs to be laid out for the players to evaluate. Once that is established choosing the politicians would be easy.
There’s money to be made boys. It’s time you sat down and talked. If you keep waiting for a bureaucracy to lead the way we’ll still be looking up three generations from now.

Your addendum is misrepresenting what I said. I wasn't proposed reusing the 'disposable' tanks later; I was saying that disposable tanks can be used to carry up propellant for transfer to a different set of reusable tanks (for example, those inside a space tug or lunar landing vehicle.)

We might even imagine launching the propellant in a different chemical form and converting it to fuels in space (water converted to hydrogen and oxygen, for example); the launch tank wouldn't need to be capable of holding cryogens.

Paul, we can do any and all of those things, but they have nothing to do with Jon's commenter's suggestion, which is what I was critiquing.

My one concern would be as to the location of the depot... that means that every launch that wants to use it has to go from the dirt to a particular orbit, stop there, and gas up. Is that always more efficient for interplanetary travel than direct trajectory insertion? Where's the optimal point to park a gas station--GEO, L1, somewhere lower?

BTW, I do like the above idea of shipping water to the station and cracking it with solar/nuc power in orbit. How efficient is that from a mass-lifted perspective?

I don't think there would be much savings in the fuel as you are launching the same mass just combined chemically.

There may be a small savings in not having to cope with as much boil off.

There might not be any advantage in tank weight either as a cryogenic storage tank would still have to be lifted to be able to store your LOX and LH2.

However water is as every bit as easy to handle as dihydrogen monoxide, with which we have some experience. :) It could be launched and stored as H2O and cracked as needed. There might be considerable savings in groundside handling araingements, which, except for the rockets themselves would involve.....plumbers.

Additionally one could use a HARP cannon to launch water at ~100 lbs a time as it is pretty insensitive. A big gun could fire(in theory) every few minutes, making it both constant and responsive. This would be an interesting area for a private entity to get into, buy some surplus 16" barrels and reproduce that wannabe Columbiad down in Barbados. Ship water to the station...or any paying customer..for later cracking to fuel or any other uses water may have....kinky though some may be.

In reference to Dr. Simbergs goal of a spacefaring society. We'll go far if we learn vacum fuel cracking of one of the commonest elements in space.

I don't understand why you can't just transfer a new tank, rather than doing fluid transfer to the spaceship. The tank is just packaging. It could still be refilled in space, but if you transfer the entire tank you won't have to build every spaceship with the ability to transfer fluids and pressurize tanks. You use a specialized orbital facility to do that, and it produces full tanks. You swap a full tank for an empty one, and ship the empty back to the orbital refilling station. SCUBA divers have been using this technique forever. The mass of the tank stays in orbit, so you're still only shipping up fuel. The tank just becomes your transfer medium instead of a bunch of hoses, fittings, pumps and valves.

Another advantage of transferring the entire tank:

If you only need half the propellant for a certain mission, you can swap in a smaller, lighter tank. With in situ refuelling, a half-load of fuel still requires the big heavy tank designed to carry twice as much.

Or maybe you design your ship to carry between 1 and X identical tanks, depending on how much fuel you need. Now you only need to 'fill up' with the number of tanks required for the mission and save weight.

Carry only the fuel you need, and carry only the tank required for the fuel you need.

You can do that, Dan. The down side is the structural inefficiency and the more complex plumbing required for multiple, smaller tanks. In terms of tank weight, empty tankage doesn't mass that much, particularly for in-space vehicles.

That's all true, and I don't know where it would come out in the wash after everything is taken into account. Maybe when you add up the cost of the structural inefficiency (although that might not be that great for transfer vehicle that doesn't land), the multiple tank idea doesn't work. So maybe you have one tank, but it can be multiple sizes so you only carry the tankage you need. Or maybe it still just makes sense to do fuel transfer to the vehicle, and avoid the additional structure of a removable tank and fittings.

One other advantage to doing it off-vehicle, though - risk management. If you fill the tank at a 'filling station', then catastrophic problems don't affect your vehicle and people. Unless it turns out that the risks of transferring a full tank to the spacecraft are even greater than the risk of in-situ refuelling. I'm certainly not in a position to know.

I just wouldn't take anything off the table until the specific engineering analysis of the alternatives is done.

One advantage of launching water is that it's considerably denser than the equivalent amount of LOX and LH2 (although the amount of LH2 and LOX you get out isn't in the ideal ratio for a rocket engine, so you may want to launch some LH2 as well.)

Launching water with a HAARP cannon isn't very sensible (HAARP needed a three stage rocket in its projectile -- this is supposed to be an improvement?!). However, maybe someone could build a very small RLV capable of carrying 100 kg (or even less) to orbit, with short turnaround time. Air launch would be the obvious way to go.

You could launch powdered water...but you'd have to know what to add. ;-)

Paul, I'm also a strong proponent of launching water, storing it at a logistics depot in LEO, cracking it as needed and storing the results for a short term in pre-configured tanks for use by dry-launch vehicles. I've gone a little bit down the road of requirements for physical and electrical interfaces and "contracts" that would define the tank standards, and I think this could be a great opportunity for a breakpoint in the transportation architecture.

The big advantages seem to be: architecture flexibility, easier automation of tank transfer, non-cryo fluids are easier to store long term, standardization (and therefore commoditization) of vehicle fuel systems. This would be a great anchor point for a bunch of Value Network branches (in a "Mike Mealling" economic universe)

Major cons: large capital investment, "what if dogs don't eat the dogfood?", that is, the "chicken-n-egg" problem of convinicing vehicle designers to sign on to the standard (NIH?)

I don't yet know where the value break-points are, but I really think this approach is worth very serious investigation.

PS. I'm looking at tanks in this architecture not as disposable, but as re-usable. More like the model for Propane tank re-fills at gas stations: "You give me your empty tank and a little bit of money, and I'll give you a full one"

Ben: IMO, the launching of propellant (water or otherwise) becomes interesting if we can launch it in very small packets, using a launch system tailored to low cost at high rep rate. Laser launchers or micro-RLVs, say.

If you fill the tank at a 'filling station', then catastrophic problems don't affect your vehicle and people. Unless it turns out that the risks of transferring a full tank to the spacecraft are even greater than the risk of in-situ refuelling.