Advice on spacecraft design sequence

[Omnivore]

I'm in the middle of working on the Dark Stars ATU Spacecraft design sequence and could use a sanity check and/or advice/pointers from anyone who has done this type of thing before.

My goal is to have as simple a design sequence as possible while maintaining the desired level of realism (using reaction drives). I believe I can do this by using a detailed hull design sub-step ala FF&S, and then assuming a total mass based on hull volume to simplify the remainder of the design process.

For example, I design a 100Td hull using FF&S, assume a total mass of 1000 metric tons, and design the remainder with a process similar to High Guard. Granted the final total mass will be different than the assumed, but it should be simpler to adjust on a second pass I think.

Any advice, tips, pointers, etc appreciated, thanks

Omnivore

 

[Omnivore]

Ok, I'm looking at QSDS 1.5e as a starting point. Planning to simplify it along the lines of robject's ideas in another thread.

Now reaction drives through a bit of a wrench into it because of the changing G-ratings with varying ship mass. So... what if I would assume that all non-hull, non-cargo, non-fuel areas of the ship have an average mass of 7 metric tons per displacement ton? Add the hull mass to that (given from the hull design or in a table) and you have your dry mass.

Cargo mass would be assumed at 'standard freight density' of 10 metric tons per displacement ton (loaded). Fuel mass would be according to fuel type.

Does this sound workable, usable, and playable to you?

 

[robject]

QSDS with simplifications would work out fine, but as long as you're starting from scratch, I'd challenge you to find a minimal system that only designs exactly what's playable.

In short, I suggest starting even farther back, with Book 2, and adding on only what you need. Armor. Sensors. Landing gear. A Mass-based thrust table. I recently made some suggestions for Book 2 extensions and changes that make the system usable. By that I mean usable in a MegaTraveller world, or a Mongoose world, or a High Guard world, etc.

http://www.travellerrpg.com/CotI/Discuss/showthread.php?p=389734#post389734

http://www.travellerrpg.com/CotI/Discuss/showthread.php?t=25701

http://www.travellerrpg.com/CotI/Discuss/showthread.php?t=25889

But don't add detail needlessly, it will bog down. Ive been there.

 

[Omnivore]

Thanks robject, exactly the kind of hints I was searching for. (Side note: the process of searching the forum leaves much to be desired!)

The part I'm struggling with is that a hard scifi setting such as Dark Stars seems to encourage (at least for me) a tendency towards gearheadism. Although in reality it is nonsense since it is impossible/impractical to account for all the details without doing an actual real world design.

I suppose that what a spacecraft design sequence should resemble is a list of specifications you would give to an architect for a house you wanted built. Perhaps closer to the point, more like a list of specs you'd give to a Burt Rutan type for a custom private airplane.

Reaction drives do throw in the monkey wrench of acceleration based on actual mass and so changing over time as fuel is expended. Of course the spacecraft would not be designed to withstand the full acceleration capabilities of the engine(s) when the tanks are nearly empty, but rather some midpoint like 50% or 33% of fuel load. Another complication I find hard to ignore is that the simplest way I can find of treating fuel in the Dark Stars setting is to require three distinct fuel types.

Yet in the end, in order to be playable by non-gearheads (or people with real lives ) a usable design sequence has to make the complex simple. Much work ahead!

 

[robject]

Gearheads want to make meaningful tradeoffs. The casual designer wants to make something that works. I think you can have both.

For calculating acceleration given load, use a formula, and show it. However, also include a table of approximations. Make safe assumptions for calculating mass, and then show variations on that based on choices tailored for specific tradeoffs - tradeoffs that a typical purpose ship wouldn't make, but someone might in some edge cases, scientific, military, exploratory, or mercantile.

 

[robject]

I have a suggestion for realspace drives. Have drives logarithmically built to map to the most convenient units of space/time. This means you first need to figure out what the most convenient units are.

I'll assume that interplanetary travel in Dark Stars is measured in AU per month, i.e. "a long time". So acceleration is lower than the OTU. But if it reaches 1G, then effectively travel is pretty much the same -- i.e. it's slower, but not game-changingly slower. If you want the game to FEEL slower, then make realspace drives LARGE.

So, perhaps the convenient target is based on 1 AU. This drive, operated under constant acceleration and turnaround at the halfway point, will move a ship 1 AU in a month's time. Drive effectiveness likely depends on its strength relative to the ship's mass, so efficiency will range from a tenth or hundredth of an AU per month to, perhaps, 10 AU per month (that's about 0.1G).

 

[Omnivore]

Excellent idea on the drive ratings, I'll have to poke around with the numbers. The drive ratings in Dark Stars are quarter G (2.5m/s^2) rather than full G as the more fuel efficient drives are rather wimpy in the thrust department. This will probably mean about a month on average to and from Lares drive points (special jump points).

I went back to basics on the hull design but.. well.. I couldn't keep from getting gearheady. Reason being I want a rock solid reason for a small ship universe and that comes from considering G-ratings, Agility (turning) ratings, and required structural strength. However, I borrowed a bit from High Guard so the system is a lot simpler than FF&S overall. The nice thing is I can do a simple table of common hull sizes at TL12 and it becomes very easy to convert them to other tech levels.

Back to grinding numbers!
Brian

 

[robject]

Regarding mass tons, I might suggest starting with a simplifying assumption that non-dense parts of the ship (living space, empty fuel tankage, etc) are 1 mass ton per displacement ton, and dense (mechanical, fuel, armor, etc) parts of the ship are 10 mass tons per displacement ton.

If that doesn't work well enough, then edge the "empty mass" higher until it's where you want it.

For example, a 100t fuel tank, full, masses 1000 kkg. At half empty, it masses 550 kkg.

It is convenient if ships are, as you mention, by default, structurally designed so that the engines are too powerful for the ship when the fuel tanks are half empty. That means you can, under normal operations, assume a constant level of acceleration from half-tank to empty. I like that, it's a good simplifying assumption.

I suspect, then, it would also be possible to design a ship so that it can only withstand a certain acceleration ("1AU/Month") and install engines that can push it at that acceleration on a full tank. In that case, there is no extra math to do: the ship can move at its fully rated acceleration. Entities which can afford the absurd expense would build ships this way. Entities which are economical would build the other way - install the smallest engines possible, and make the ship crawl.

 

[robject]

I like your Lares points; makes for a nice Niven-esque feel to the setting.

Lares points are going to be close to the homeworld. I would guesstimate 1 AU (especially if Lares points are off the ecliptic as you hint). So once again I would suggest that 1 AU be your unit of range/time.

Hull design - just list the factors, and how they relate. I think you may have two kinds of structural parameters, one for forward acceleration, and one for angular stress in rotation. These are bracings, not armor, in that they keep the engines from tearing your ship apart. What are the tradeoffs?

For armor, I would suggest taking a page from T5. T5 simplified FFS2's armor scheme by assuming one layer takes 4% of the hull volume for internal bracing and coatings, and has an Armor Value equal to the ship's TL. From there, some modifiers can creep in, but that's the basic. So you don't need to specifically identify the material being used (Superdense at TL14, Hullmetal at TL16, CrystalIron at TL10...), but just note the strength. Additional elements can be added to change its nature a bit, but automatically knowing the AV goes a long way (number of layers x TL = AV).

By the way, 4% (more or less) is a pretty good number, because the tradeoff in volume for an armor value of appx 10 to 15 maps nicely to 3 or 4 dice of damage. IF you want to get gearheaded there, you can increase the percentage with the layer number. I.E. % = (3 + Layer number). It's reasonably realistic, and penalizes excessive armor.

 

[far-trader]

Regarding mass tons, I might suggest starting with a simplifying assumption that non-dense parts of the ship (living space, empty fuel tankage, etc) are 1 mass ton per displacement ton, and dense (mechanical, fuel, armor, etc) parts of the ship are 10 mass tons per displacement ton.

Were it me, I offer this as an even simpler imo alternative, I'd skip the whole mass tons v displacement tons and treat the design system as rated for mass tons. This is imo the way CT originally reads and goes to some lengths to enforce artificially. The big bonus this way, all those grossly over sized deckplans are no longer quite so broken. That 500 to 600 dton Far-Trader? So what? It is still 200 mtons. It just has a lot of low density construction.

Example:

The 100ton Scout/Courier

Jump Drive A2 is 10tons (mass)
Maneuver A2 is 1ton (mass)
Power Plant A2 is 4tons (mass)

etc. etc.

So the 15mton total of drives may only require 16 deckplan squares, and allow 4 more deckplan squares for access between (10dtons)... just like in the official deckplans for the Type S in Supplement 7. Poof, no issues with the deckplans.

And those 6 square staterooms with 26 squares of common and access space (25dtons total), massing a total of 16mtons? Poof again, all the issues disappear.

etc. etc.

Draw your deckplans (just about) however you want. Make the staterooms huge if you find 4 squares too small, it's mostly empty space (no mass) anyway. They still require the same 4mtons per for the actually installation (including basic hull mass). Make you cargo decks 6m tall so you can easily drive on/drive off and stack crates to allow room to walk around them, you're still limited to and by the total designed mass tons.

You could even go so far as to say something like...

...you need to make a 2 parsec jump with the 400mton designed J1 Merchant? Well sure you can if you run with the entire cargo hold (200mtons) empty reducing the overall ship to 200mtons (and it only takes 40mtons of fuel, the same as you already have for the normal 400mton J1). What's more you can do 2G with the maneuver drive when empty!

 

[Omnivore]

A constant density would simplify things assuredly. It might not be a horrible idea but I am concerned that military craft would have a significantly different average density than a cargo hauler. How important is that? Not entirely sure yet.

I've 'finished' (for certain values of finished) the Hull design portion of the Dark Stars spacecraft design sequence. It sidesteps the mass issues for now by assuming the structural supports and hull armor use the same material - or at least a material with the same general density and cost.

The hull design sequence, with table of hulls from 10Td to 10000Td is here.

Next up I figure I'd better take a good look at the 'landscape' to see what Isp's and times are required for common travel. That'll give me a better idea of what I need to get out of the engines and what the relative importance of average density would be.

Thanks
Brian

 

[far-trader]

A constant density would simplify things assuredly.

That's not directed to my offering of designed mass tons suggestion is it? Because I didn't mean to imply it was constant density at all, but may have explained it poorly or insufficiently.

It is quite the opposite in fact. You get to decide the density by somewhat arbitrarily assigning whatever displacement you like to any specific object.

You might even have different manufacturers creating the same (effectively) element but quite different in displacement. For example Yard A might make the 10mton Jump Drive A in a form that requires a mere 5dtons including access, while Yard B takes a different form factor and uses 10dtons not including access space. Obviously while both drives perform the same, cost the same, and mass the same they cannot be easily swapped. Such as the players need to replace their Yard A Drive that was destroyed... but they are half way across the sector and the Yard B Drive simply won't fit unless they cut a big hole into the cargo deck...

...speaking of which, cargo is a constant density, 1 to 1, in CT. This is due to packing and handling issues. That need not be the case with the design mass tons suggestion. You still load your cargo in mass tons per what you are designed to carry, but now that 1mton of feather mattresses eats up a lot more cargo hold than that 1mton of heavy metal ore.

 

[Omnivore]

That's not directed to my offering of designed mass tons suggestion is it? Because I didn't mean to imply it was constant density at all, but may have explained it poorly or insufficiently.

It is quite the opposite in fact. You get to decide the density by somewhat arbitrarily assigning whatever displacement you like to any specific object.

Oops! I plead excessive time spent staring at spreadsheets! Now that I understand it, yes it does make more sense. In essence it becomes a purely mass based system. I'll have to ponder it over more coffee - or maybe after some sleep

I remember looking at a pure mass based idea awhile back but what threw me was armor. You have to assume some relationship between mass and volume in order to wrap an armored hull around a collection of parts - at least I think you do... maybe I'm still missing something.

 

[Omnivore]

By the way, 4% (more or less) is a pretty good number, because the tradeoff in volume for an armor value of appx 10 to 15 maps nicely to 3 or 4 dice of damage. IF you want to get gearheaded there, you can increase the percentage with the layer number. I.E. % = (3 + Layer number). It's reasonably realistic, and penalizes excessive armor.

I ended up using the base 4% figure too but in a different way - more ala High Guard. I split the 4% (at TL7-9) in two, with one half coming from structure, the other from shell for the assumptive 'run of the mill' ship. This made the base armor value equivalent to Striker AV32, or High Guard -1 (if HG had negative armor values).

I'm assuming non-standard materials based on various forms of CNT and similar but I can see I missed a few things in that and need to rethink it a bit. (Pricing is off for one thing and mass/density requires more thought).

I also might need to refactor for tenth G based ratings rather than quarter G, since the constant acceleration month long 1 AU trajectory doesn't need anywhere near a quarter G of thrust.

 

[Omnivore]

FUpdate: after tossing around a bit, I came to the conclusion that the opportunities for simplification with a mass-centric design far outweighed the drawbacks. So I assumed an average wet mass density of 4 metric tons per displacement ton and reworked the hull design equations in terms of mass rather than volume.

Since the vast majority of volume and surface area concerns are contained in the hull design, the assumed average density allows a pure mass based design for the remainder of the space craft design sequence. So outside of the hull design itself, the sequence is back to a simple percent plus fixed basis overall that should be familiar to anyone who has used High Guard at all.

I've also changed from the quarter G rating system to a straight up 1 m/s^2 rating system so that part should be much clearer and easier to work with. Also, while just a name change it should result in better clarity, I renamed hard points to mount points in the hull design sequence as they aren't just for weapons. Any component part that requires any surface area will require one or more mount points.

Thanks,
Brian

LATE BREAKING NEWS: The first draft of the Engineering section has been added to the Dark Stars blog. I expect to need to do some more tweaking and fleshing out but the basics are there.

 

[drkem99]

Ken Pick's article in Freelance Traveller ( http://www.freelancetraveller.com/features/shipyard/tonnage.html ) gives a figure of 5 to 7 metric tons per Traveller displacement ton for surface warships with varying degrees of armor. Submarines (with considerably less open interior space and heavy, strong hulls) are designed to be close to 14 metric tons per displacement ton (the density of water. This minimizes the volume of ballast tanks the submarine needs.)

Depending on your design philosophy, you can pick one of these numbers for use. I think 4 metric tons per displacement ton may be a little light for your purposes.

 

[Omnivore]

Depending on your design philosophy, you can pick one of these numbers for use. I think 4 metric tons per displacement ton may be a little light for your purposes.

I'll have to take a look at the Freelance Traveller article. Initially I had thought 7 Tm per Td would be about right, but then I did the initial specs for the available reaction drive engines and... seems 50% or more of the mass is going to be in fuel tanks full of low density fuel. That basically cuts the density in half.

Realistic Isp reaction drives are a whole new ballgame in Traveller from what I've seen so far.

Thanks for the tip,
Brian

 

[drkem99]

If you have not already looked at Winchell Chung's Atomic Rockets web site, you should do so; there are pages of data and nomograms on various reaction drives you should find helpful at this page and others on the site: http://www.projectrho.com/rocket/enginelist.php

Your goal to use realistic Isp drives in Traveller is commendable but probably futile. The fictional fusion drives in Ken Burnside's Attack Vector: Tactical (see Chung's engine list) probably represent a bare minimum for the required technological performance to allow reasonable interplanetary travel times and to allow starship combats of interest to your players. Even so, the AV-T fusion drive's performance is well above what is technologically plausible without a major breakthrough in the handling of waste engine heat.

The required mass ratios (reaction mass requirements) for most interplanetary missions with any realistic Isp drive are well above 2 (i.e., well above 50% of the ship's mass is fuel.) Unless you are fusing it in the drives, L-hyd also makes a poor reaction mass choice due to its low density; water, for example, is far superior and easier to store.

Traveller: The New Era uses a technology called high efficiency plasma recombustion as a maneuver drive that might work for your design sequence. (HEPlaR) is a very high impulse and high thrust reaction drive with a reasonable reaction mass requirement. While just as magical in its own way as thruster plates, it is more subtle in its violations of the laws of physics; it also has the advantage of avoiding relativistic weaponry in your campaign.

 

[Omnivore]

Yeah Atomic Rocket rocks, as I've noted elsewhere on these forums a few times AV:T is interesting to a point but really blew it with those unnecessarily beyond the edge of the envelope engines, in my opinion at least. Don't confuse thrust with Isp. Beyond the interface to planetary bodies, raw thrust isn't worth much, Isp is king.

At higher tech levels, even in a realistic universe, an Isp in the 70k range should be obtainable at some point with advanced ICAN II or Daedalus style Orion Fusion aka IC-Pulse fusion, and with an engine capable of that, interplanetary voyages with times of a few months are quite possible. However the 50% case I was referring to was a boundary condition of a high tech but still reasonable drive traveling from a high port to a Lares point (similar to Niven's Alderson points) at a distance of 1 AU in a month or so of time allowed.

As a last resort should it prove necessary I'm much more likely to modify the one piece of handwavium in the setting (the FTL drive) than go to the extremes of AV:T's or TNE's impossibility reaction drives.

 

[mike wightman]

Use a denser fuel - ion engines using xenon give quite a high impulse IIRC.