Saturday, November 12, 2016

Building a Spaceship in Dungeons & Dragons, Part 2.5

This is an update to Spaceships in D&D Parts I and II. I had hoped to give you an update to Rock to Hide Me instead, based on a paper I found on a government server. But unfortunately, after I read it, I concluded the paper in question was so mercilessly boring that I don't think I can turn it into an article that anyone would want to read. So, on to the next item on the itinerary instead!

I've made some significant modifications and updates to Spaceships in D&D Part I and II, responding to and incorporating some suggestions and criticisms by commenters on them – suggestions and criticisms more than a year old (egg on face), but better late than never. I'd particularly like to thank Brian Ballsun-Stanton, who made a number of very helpful comments. So as not to make everyone reread those articles, I will be discussing the changes separately here. Part III in the series, on mission planning, will follow, hopefully in the not too distant future. There may or may not be a Part IV, on things to do in space, depending on what I come up with.

A Big Mistake

While prepping this article, I discovered a very serious error in my geometric calculations that led to me underestimating the volume of the combustion chamber by approximately a factor of four.   Since the volume of the combustion chamber determines the thrust produced by a given blast of flame spell, this led to a significant overestimation of how much work we needed to do to get a given acceleration.   This issue has been corrected.

In the process, I also made the following diagram of our spaceship:

What can I say, I'm no artist.

Throttling System

This is something that was supposed to be in the Part II, but which I forgot to include. To recap: our propulsion system consists of a mithril combustion chamber reinforced with a lyre of building; a collection of decanters of endless water to provide matter; and an energy transformation field loaded with blast of flame to turn the matter into hot gas to serve as our exhaust. The energy transformation field is powered by a collection of spell traps loaded with first-level spells, with the traps activated by a button in the cockpit.

However, there's a problem here: there's nothing to turn the decanters off when we're not thrusting. We might actually be okay leaving them running, given that the water should eventually boil away in the vacuum of space, but that means we'll have a continuous residual thrust that we need to account for, and it means we have to worry about things clogging with water ice. In particular, our combustion chamber is only invulnerable for two thirty-minute periods a day; I'm worried about the nozzle clogging up with ice and then pressure building up inside the chamber until it ruptures. Best to have a way to turn them off when not in use.

According to the SRD, a decanter is controlled by command words. There's no actual ruling that I can find about, e.g., how close the command word needs to be spoken to the item for it to work. My guess is that the writers implicitly assumed that you would be holding the item when you spoke the command word, but there's nothing explicitly saying that you can't activate the item from a world away. Fortunately, this is easily determined via Science: during the Propulsion Research phase, measure how far away you can be from the decanter to control it, and particularly whether the mithril sphere we used in measuring the thermodynamics of our thrusting options blocks the effect. For the purpose of the rest of this article, I'm going to assume that the command word needs to be audible to the magic item with a DC 0 Listen check - which means we can't just say the command word in the cockpit to shut the decanters off. Fortunately, there are other options. Unfortunately, those options are a bit complicated, not to mention pricey.

The simplest method I've found - and please comment if you can suggest a better one - is an antimagic field. We use an energy transformation field loaded with antimagic field, and powered by 1st-level spell traps. Subcontracting the energy transformation field to Bo-Wing the Aeromancer costs 5,190 gp, and the 1st-level spell trap costs 24.05 gp, for a total of 5,214.05 gp.

Now, the geometry of this is going to be a little complicated. Our antimagic field is a 10' emanation.   An emanation is a sphere, but it's blocked by anything that blocks line of sight. In other words: if I can draw a straight line from a point to the origin of the antimagic field that doesn't pass through any physical objects, and which is 10' long or less, then that point is effected by the antimagic field. We also have two energy transformation fields, one powering our blast of flame, and one powering our antimagic field. Energy transformation fields are 40' radius spreads: a spread is like an emanation, except it can turn corners. In other words: if I can find a path from a point to the origin of the energy transformation field that is 40' long or less, where that path can bend and twist, then that point is effected by the field.

Neither of our energy transformation fields can overlap the decanters of endless water, because otherwise they'll suck up the decanters' energy and they won't work. Our antimagic field, on the other hand, must overlap the decanters. However, the antimagic field can have its origin anywhere inside the energy transformation field powering it.   By placing the origin of the antimagic field-generating energy transformation field inside a box connected to a 39'-long coil of pipe, we can position the origin of the antimagic field so that it covers the decanters but the energy transformation field does not.

Unfortunately, because antimagic field has a minimum 110 minute duration, this means we'll need to plan our course changes well ahead of time. To thrust, we set the lever controlling the spell traps at point B to off, and wait 110 minutes for the antimagic field to dissipate. At this time the decanters begin to generate water, and we play the lyre of building and set the lever controlling the spell traps at point A to on. The blast of flame field charges and fires, and we begin to thrust. When we want to stop thrusting, we set the lever controlling A to off and the lever controlling B to on. The blast of flame field stops firing, while the antimagic field field charges and fires, shutting off the decanters.

(Since antimagic field can be shut off by its caster at will, it would be a reasonable DM call to say that we can program our energy transformation field to create an antimagic field that lasts a shorter period of time, but that's definitely not supported by the rules.)

Do Decanters Count as Construction?

This is a rules interpretation issue which I neglected to raise in the original draft. Our decanters are inside the combustion chamber, which, during thrust, is full of high-temperature, high-pressure gas. The lyre of building protects the chamber walls from damage. Does it also protect the decanters? If not, they're going to be destroyed pretty quickly.

The text in the SRD says that the lyre "negates any attacks made against all inanimate construction (walls, roof, floor, and so on)". My personal ruling would be that, if we build the decanters into our chamber wall, then they are part of the construction. However, that's just a ruling. We need to add a step into our Propulsion Research section to determine if this is the case - fortunately, we already have all the materials ready to hand.

If the lyre does not protect the decanters, then we're in trouble. There are no readily-available ways to protect them that I can see, and almost all propulsion methods require decanters or some analogous magic item to serve as reaction mass. I can think of a few ways that might work, like make whole spell traps to repair damage faster than it's done, but the only way to determine if they will actually work is by trying - there's no way to determine how much damage the decanters are subject to per round from exposure to the combustion chamber from the rules as written.

Alternative Approaches to Propulsion

Several alternative propulsion methods to the decanter + blast of flame approach have been suggested. Some of them are definitely inferior to my current method, some of them we can't say without doing in-game research. Since we can't say for sure that any of them are actually superior I've decided to stick with my current approach, but I discuss them all below, and I've added appropriate sections to the Propulsion Research section.

Using decanters of endless water as a water rocket: The idea here is to use decanters on geyser mode, exploiting the thrust of the water itself without trying to heat it. The data we have is insufficient to determine what the thrust produced by a decanter is. We know that, on geyser mode, it produces 30 gallons per round, or 5 gallons per second, but we don't know what speed that's produced at. But, if we know the size of the decanter's mouth, we can calculate how fast it's moving. The SRD describes the decanter only as an “ordinary-looking flask”. If we assume the mouth of the flask is 1 inch wide, or 2.54 cm, then the mouth has an area of 5.06 square centimeters. Five gallons is the same as 18,927 cubic centimeters, so the speed of the water is 18,927 / 5.06 = 3,740 cm/sec, or 37.4 m/sec. The density of water is 1,000 kilograms per cubic meter, or 0.001 kilograms per cubic centimeter, so the decanter produces 18.9 kg/sec of water, for a thrust of 706.86 N per decanter.

Not counting the combusion chamber - which the water rocket would not need - our vehicle weighs about 30 tons, so to get 1.5 G acceleration, we'd need 450,000 N / 706.86 N = 637 decanters to replicate our current thrust. With our various cost reduction modifiers, it costs us 2,164.22 gp per decanter, for a total price of
1,378,608.14 gp... Somewhat in excess of our current approach.

Steam rockets: The basic idea here is to combine decanters of endless water with a source of heat, such as matter agitation or heat metal, to produce a steam rocket.   Unfortunately, the descriptions of these spells and effects are extremely vague about what specific temperatures they produce and so on.   For example: matter agitation effects 2 square feet of material; what is that in volume?   How hot does heat metal make the metal it effects?   And so on.

Let's talk lower estimates: what is the absolute minimum that a steam rocket could cost?  We can figure this out by ignoring the actual turn-it-into-steam parts of the rocket, and just considering the cost of the decanters.   Now, take a look at part II of this series, and specifically the section labeled Engine, and specifically specifically the equations therein.   Now, the specific heat ratio of steam is about 1.3, the exhaust temperature is 373.2 K, aka the boiling point of water, and the average molecular weight of water is 18 kg/kMol.   Plugging those in, we determine that the exhaust velocity of a steam rocket with an ideal nozzle is 1,222.27 m/sec.   Since our decanter produces 18.9 kg/sec of water, this amounts to a total thrust of 23,100.8 N per decanter.   To replicate our current thrust of 2,078,574 N, we will therefore need 90 decanters.   At a cost of 2,164.22 gp per decanter, that's 194,779.8 gp in decanters alone, before we add in the cost of the systems to actually boil the water - more than the cost of our current engine.

Nonetheless, such a system could still have a role to play.   In particular, unlike blast of flame, it's plausible that our steam rocket might not need a lyre of building to keep the combustion chamber and nozzle from evaporating - we can make it out of the same stuff they make tea kettles out of.   That means the engine could thrust continuously, instead of for two half-hour periods per day.   Also, if you're willing to accept a lower acceleration, the steam rocket can have a smaller minimum cost than a blast of flame rocket.   I think the steam rocket has a role to play as a system for getting around once you're out in space, the magical equivalent of the real world's ion drive.

Using pocket dimensions as ideal combustion chambers: This is a very cool idea, but I don't think it actually gets us anything – because we already have an ideal thrust chamber. Since our “combustion” chamber has infinite strength, thanks to the lyres of building, we can let the chamber pressure be (effectively) infinite, and at a much lower price than using this approach.   A pocket dimension could thrust all day, but I don't know of any way to buy a suitable pocket dimension for less than the cost of 48 lyres, which would also let us thrust all day.

Additional Cost Reductions

I've found some ways to eke out our budget a bit further.

First, in the original draft, I gave the astronaut the Landlord feat... And then forgot to make use of it! At 11th level, the Landlord feat gives you an allowance of 75,000 gp to be spent on a stronghold, plus one-to-one matching funds for any money spent beyond that. Unfortunately, the big expense of the ship is the engine, which is not statted as a stronghold, but this can still be applied to the rest of the ship.

Second, since writing the original, I found the Complete Cost Reduction Handbook, which has a bunch of ways to reduce the price of the items we're using. Unfortunately, many of them are from books I don't own (Bind Elemental applied to conventional magic items, membership of Dark Spire college). But we can at least use Favored in Guild (Arcane), which reduces costs by 5%, and Apprentice (Craftsman), which reduces costs by 10%.

Other Minor Corrections

I forgot to include the price of the spell trap powering the navigation system (24.05 gp).

We can't actually cast energy transformation field - I thought it was 6th level, but it's actually 7th - so we'll need to subcontract that. Each casting costs 910 gp, in addition to the material component costs that were already included.

We need to include a small closet in the cockpit to hold the energy transformation field powering the navigation system. Otherwise, the field will cover the whole cockpit, making it impossible to use other magic items or cast spells.


  1. Some notes in no particular order.

    For control of the decanters, I suspect Ventriloquism (Sor/Wis 1) is better suited. Or pipes. Having the decanters in the cockpit, or surrounded by an Amplify (Bard 1, Magic of Faerun (ew)) effect could be useful. Stormvoice (Champions of Valor (seriously?)) can also be a useful sound control thingo. Or Control Sound (Psi 2) Command words are poorly defined. As a (fairly) bad idea, making them intelligent items with the 10 ranks in listen is not the worst idea ever (but it's quite bad).

    Do we have a formal amount of damage that the flasks would need to withstand? If you make them (and the combustion chamber) out of "Fyrite (Dragon 347 p49) which is completely immune to heat and is costed at 50gp/lb"

    If the whole chamber is made out of Fyrite (which does not have a hardness) or adamantium, or has the hardening spell applied (depending on damage), or use a *remarkable* amount of quintessence to isolate the heating chamber from time because physics is cute?

    A portable hole is 20kGP, a lyre is 13kgp. Since, when spread on a surface, it doesn't exist, the whole "exhaust gasses are hot" problem isn't a problem (Or line the exhaust port with something approprirately durable -- though that makes controlling stuff inside the whole much harder. On the other hand, psionic items are command thought activated. Worst case, you can pay the extra to upgrade them to use-activated and have the "use" be some sort of mechanistic interaction with the environment that's environment-safe.

    1. On ventriloquism and its cousins: the problem is that the decanters are going to be in vacuum much of the time. They're built into the vehicle structure, so theoretically if you shout loud enough they could hear, but I don't see any way to model that in the rules, and I suspect you'd end up blowing out your own eardrums if you managed to yell that loud.

      On damage: I did consider fyrite, not just for the decanters but also to build the combustion chamber itself out of. But heat isn't the only problem: we also need to worry about the pressure in the combustion chamber, which - since it's an ideal chamber with asymptotically high pressure - is going to be absurdly high. Unfortunately, there's no way I can find to put numbers on that, since D&D - for some reason - doesn't have any rules on pressure damage that I can find.

      On quintessence: That's an interesting idea. I don't think we can use it to protect the decanters, because there always needs to be a path from the decanters to the combustion chamber when they're in operation, and the exhaust gas could flow back along the path. I do wonder, though, if there might be a way to submerge the decanters in quintessence when not in use, as a throttling mechanism... But that does require a Shaper-7, to make the stuff in bulk. Would need to restat our astronaut to make this work - which I'm not necessarily averse to doing; in the original draft of this, way back when, the astronaut was a named character, Mazin the Maker, and he was a shaper, not a wizard. Let me think about this...

      On command thoughts: that's also very interesting; I wasn't aware that psionic items used command thoughts instead of command words. I'm not sure we can use it here, but it's something worth knowing, so thank you. I think the argument that a command thought should be able to control the decanters from the cockpit is stronger than the argument that a command word should be able to, but, of course, the SRD doesn't give us any real sense of the range of these things... I dunno, if I was DMing, I would rule that you have to be holding an item for a command thought to work, but that's a very questionable ruling. I'll consult the RAW questions thread at and see if anyone knows.

      On the portable hole: there are three problems I see here: first, the hole is only 10' deep, and our blast of flame is 60' long. The thrust/cost ratio of these spells drops precipitously as the volume of the combustion chamber declines. Second, you still need to use lyres of building to protect your decanters and spell traps even if they're inside the hole. And third, now you need to worry about control systems that can cross planar boundaries (a lot of divination effects explicitly call out that they can't do that). In particular, I would be very hesitant to say that you could have a spell trap where the trigger is on one side of the hole and the spell is on the other.

      I just don't think it's worth it, unless we can find a variant portable hole with a bigger volume. The only real advantage - aside from the fact that it would look cooler - is that it would save us the cost of the combustion chamber, about 30,000 gp.

      Final note: is there an actual listed price somewhere for upgrading a magic item to use-activated? That would make a lot of things much easier.

    2. It's here:

      Normally some rather broken stuff can be done with custom magic items, but since we're already into spell-trap territory that doesn't matter.

      This build actually sounds like it could be productively done by an erudite (with the variant in (though that does beg the question of very high level spells.)

      With regards to blow-back, have you worked out how much damage the chamber is taking? And is there any piping/pressure arrangement which can have the water piped into the chamber?

      Also, as an aside (since the prior conversation was many years ago) have you costed using wall of flame (with permanency cast on it) as your energy source? (Perhaps with ejecting molten iron (wall of iron)?

    3. On use-activated items: Thanks, I should have thought of that. That's definitely got potential, but I need to think about how to make this work. We're way into DM interpretation territory here - well, we really always have been - but I would personally rule that, for a use-activated item, we would still need to do something to the item to turn it on and off. That could be something as simple as screwing a lid onto the decanters, but we would still need to do something. Unfortunately, the existing methods of keeping the system from blowing apart won't apply to animated objects and other creatures. Fortunately, we could probably arrange some kind of trap door arrangement. Let me cogitate for a bit on the details of this, but I think it will work, and it should at least be cleaner than mucking about with antimagic fields.

      On erudites: Yeah, StP erudite probably is the optimal way to go, but I really don't like them.

      On blowback: I can put together a decent estimate of the damage from fire, but not the damage from pressure, since there's no modeling basis for it I can find. In theory, it should be possible to isolate the decanters from the combustion chamber - it's essentially the same as how, in real life, the exhaust flame doesn't back up into the fuel tank. In D&D, though, I don't know of any way to model this. Maybe pumps powered by animated objects? That might work.

      On Wall of Flame: I did cost using wall of flame as part of a steam rocket. Using the current assumptions for how blast of flame works, the blast approach outperforms any of the steam rocket approaches. However, that's dependent on a number of assumptions about what blast of flame actually does.

    4. (Sorry, just waking up here, it's 7am) In terms of levels and items and whatnot, it may be wroth going down the artificer or warlock routes, especially considering the artificer gets such wonderful bonuses. ( (of especial note is the "item creation" line: "For purposes of meeting item prerequisites, an artificer's effective caster level equals his artificer level +2.")

    5. I've been using an Artificer (main character's cohort) as Chief Project Engineer.

      I'm going to do some asking around on pumping systems. Between our absurdly high Knowledge (Architecture & Engineering) skill check, D&D super materials, and fabricate spells, we ought to be able to make a pumping system at least as good as what they use in the real world today. If we can physically isolate the decanters from the combustion chamber, that makes everything so much easier; in particular, it lets us use animated objects to control the decanters, rather than antimagic fields. But we're also assuming that the combustion chamber reaches asymptotically high pressure, and I'm not sure if we can actually pump water into that, or how much performance we lose if we limit the chamber pressure to something we can pump into.

  2. In terms of pressure and heat, let's see what we're looking at in terms of hardness requirements. Since we can make the decanters out of adamantine, if we're over hardness to start with, then we can do magically reinforcing nonsense. I haven't seen any obvious ways to make things one-way, though "skate" (psi1) does impressively reduce friction. Otherwise, (as is the case in reality) it may be time to accept that the chamber pressure is too high due to engineering constraints.

    Have you found any useful citations for anchoring walls of force or prismatic walls to a ship? For a prismatic wall, it differentiates between gasses and "everything else" If you can figure out how to anchor one to a moving vehicle, then destroy all colours except yellow and indigo, you get a "gas-proof" wall. That blocks the flame and the super-hot steam.

    1. Re: hardness requirements: the problem is that there's no way I've found to put numbers on what the actual requirements are. We can make stuff really hard, but there's no way to tell if we've made it hard enough.

      Re: force and prismatic walls: the rules are self-contradictory on this. At various points force walls are called out as immobile, then at other points there are instructions on how to build them into mobile fortresses. Ultimately I concluded that the most plausible explanation is that they are immobile relative to whatever they're adjacent to, so if we have a rigid framework of some material, then we can anchor the force walls to it. I initially planned for the combustion chamber to be made of force walls, to save weight, but ultimately concluded that the cost was prohibitive - all those permanency spells!

      But the idea of using yellow-indigo prismatic walls as a one-way membrane, to let the water through but keep the exhaust gas from leaking back, that had not occurred to me, and that's a very good idea. I need to do some math to determine how the costs break down - an animated mithral pump may ultimately be more cost-effective, even if it means a lower chamber pressure - but that's a very good idea.

      Thank you for your help!

    2. Here's a thing: the SRD says that you need to bring each layer in the prismatic wall down in order, with the last layer being destroy-everything violet, so the spell won't work. But it also says that "certain magic items can create prismatic effects one color at a time" - any idea what it's talking about? I ctrl+F'd the magic items section but didn't spot anything...

    3. Argh. threaded convos. Can we take this discussion to a different medium?

      Permanency spells cast by *yourself* are much much much much much cheaper.

      In terms of hardness, let's start by talking about what damage we must withstand to see if it's worth checking. Assume that every part of the wall takes the full damage from your final flame product. (I just don't know what that number is) If it's too far above normal hardness, we can ignore the rest of the exercise.

      Enchanting a magic item with just one colour of a prismatic wall (at full price plus the cost of all the other negation spells) could... be... a DM ruling? Ew)

      Another route is to figure out how to get class features into magical items, as the initiate of the sevenfold veil's yellow warding is exactly what's indicated.

      Bleh. I can't find anything about magical items bypassing (we'll ignore artefacts). Perhaps figure out how to extend metamagic or up caster level? The wall from initate works for 10 min / level, and there are plenty of ways to boost caster level.

    4. 4 walls of force (since we'll need to leave one end open and one end with an adamantine cap for our pressure) come to 10,000 xp. 40k GP. (I assume the pressure will be greater than 1,000 pounds per... unit. (Dark way measures these units by creature. so perhaps psi just to be... non-silly?) On the other hand, this does get you a torchship.

      Hrmm. With control currents (Druid 4) we can induce a remarkable amount of water flow. Can you just show that providing 30ft/round extra movement to a 20 foot radius section of our chamber (assuming that it's evenly filled with decanters) provides less energy than the steam or firetrap boilers? Either way, I suspect it would make an absolutely wonderful first stage or primary pump.

    5. Is it cheaper to power 4 walls of force with an energy transformation field or 4?

  3. I just looked something up and found one of my answers some years ago:

    Apparently the stronghold builder's guidebook has a "zone of elemental immunity" that could be quite handy for this sort of thing. And, in the same vein, asserts that the guide has a prismatic screen effect.

  4. On alternate means of communications: My e-mail is in the upper right hand corner; drop me a line. :)

    On permanency: While our astronaut can cast permanency, he's not high enough level to cast it on a wall of force, so we need to contract it out.

    On hardness: The problem is that we don't know what the final damage is. The heat can be approximated by just adding up all the blast of flame spells, but the pressure damage can't be calculated in any way I can find.

    On Initiate of the Seven-Fold Veil: Let me get back to you on this, I'm AFB at the moment.

    On Walls of Force: One problem here is that we still need the lyres - the wall of force needs to be anchored in something, which will then be exposed to the exhaust gases. The main advantage I see there is that it roughly halves the mass of the vehicle - which is a pretty significant advantage, since it means we can use a weaker, and therefore cheaper, engine.

    On control currents: sadly, the velocities involved are trivial compared to what we've got now, and what we need.

    On EFTs and walls of force: To capture the entirety of the blast of flame, we need each of our side walls to be 60' tall and 60' wide, for a total area of 1800 square feet. A wall of force covers CL * (10 foot squares), so we need the wall to be cast by an 18th-level caster. The base cost of the EFT is 5,910, plus 18x50 = 900 gp for the wall, and we'd need four, so that's 27,240 gp total. Plus spell traps and the mithral framework, but that probably won't be a huge amount of money, and even if it takes us over the price of the existing approach, the system will be much lighter. But I feel like I considered this approach in the past and rejected it for some reason, but cannot recall what it was... Let me think about this and see if I can remember.

    On zone of elemental immunity: only protects against heat, not pressure, and we already have ways to protect the system against heat.

    Still digging on pumping and prismatic walls, we'll post more later.

  5. Hi,
    Could you provide us a link to that "mercilessly boring" paper? I must admit that I find the Manhattan Shelter Study and related projects so fascinating as to outweigh the boringness, and in any case, having written a fair number of like papers (though on less fascinating matters), I have a soft spot for them :)
    Thanks in advance!

    1. Sure!

      I must warn you, though, it is pretty darn boring. Given the timing, I'd hoped it was a followup to the original Manhattan Shelter Study. Unfortunately, it turned out to be a massive compilation of statistics and other information about tunnel excavation in urban areas, with no mention of designs or anything - just how to price the tunnels. It's possible that it was prompted by the MSS, but there's no evidence of it in the paper, and the depths it considers are much shallower (typically about 100 to 200 feet).