[0.23.5] Spherical and Toroidal Tank Pack (Updated 05/02/14) (New download link)

[Tallinu]

Alright, like I suspected, you definitely have a better handle on the math involved than I do. Interesting information, certainly.

the mass of a tank is not proportional to its surface area, but rather to its volume. As you increase the volume of the tank, you must also increase the thickness of its walls in order to contain the same internal pressure.

That certainly makes more sense in relation to the stock tank volumes and masses - like the same-length tanks of different diameter having the same change to mass and volume, instead of what I had expected.

cylinder width (torus minor radius) certainly does, but a quick glance at the equations indicates that wall thickness being proportional to just the minor radius would keep the tank mass proportional to the tank volume.

From Wikipedia (http://en.wikipedia.org/wiki/Pressure_vessel): "No matter what shape it takes, the minimum mass of a pressure vessel scales with the pressure and volume it contains and is inversely proportional to the strength to weight ratio of the construction material (minimum mass decreases as strength increases[5])." It sounds like the length doesn't matter, maybe? There's also a bunch of math I don't have the context to compare between for various shapes. But KSP certainly treats it like cylinder height doesn't affect the mass ratio, so it seems reasonable to follow that rule in this case even if it doesn't always hold fast in reality.

So, comparing the "large" torus (the one with a 1m minor diameter or 0.5m minor radius) with the stock 1.25m diameter tanks (radius of .625m), and assuming that they're built with similar materials and methods, wouldn't this actually imply that the toroidal tank's mass ratio would be similar to that of the stock tank? And that a toroidal tank with a minor radius of 0.625m would definitely have exactly the same mass ratio as the stock 1.25m diameter tanks?

As an aside, with the values I plugged into my config files giving them the stock mass ratios, I was able to make this beautiful (if undoubtedly a bit fuel-inefficient) 'lander' which turned out to be capable not only of landing but also of SSTO takeoff from Kerbin.

Yes, Talisar, Taniwha, this is one reason I want tori with larger minor diameters. A single XL variant with capacity similar to the two of these would cut out the KW pancake tank in the middle as well as looking better and wasting less space between the 3.75m engine and the inner edge of the toroidal tank.

[taniwha]

Well, since I included the the struts in both the mass and the volume, larger volume but smaller major radius toroids would probably make for slightly higher efficiency.

Here are the posts I did for skin thickness and the actual tank masses.

[Tallinu]

Thanks. 4.128 eh? So, slightly under 1/4 dead weight compared to 1/9 stock. "Not as bad as 1/3 dead weight" is, unfortunately, not a stellar review (no offense). I'll just keep using the stock mass ratios for now, so that I can actually effectively employ these parts in functional ships and not just "concept craft" aka futuristic art projects.

And out of curiosity, RE: "Also, it turns out that the jumbo-64 is about right at 4t assuming a skin thickness of 9mm and being made of iron (and a few other not too unreasonable assumptions)." Does this hold true for the 1.25m tanks as well, or the smaller 2.5m tanks, since they all have the same mass ratio?

[taniwha]

Any shorter tanks in the same series should have lower mass ratios, but the FL-T800 is a scaled version of the Jumbo-64 (1/2 scale, 1/8 mass/volume), so it is right.

Those other assumptions are mainly in reference to the tank's internal structure (hemispherical ends on the internal tanks).

Titanium alloy should get you pretty close to the same mass ratio, so I can't say you're cheating, but do keep in mind that I was aiming at keeping things consistent.

[Tallinu]

The way I see it, choosing to use a mod part with such a poor mass ratio is about the same as shooting yourself in the foot. The same effect (having extra fuel mounted out around the center) could be achieved using purely stock parts without resulting in nearly that much dead weight - just radially attach three Rockomax X200-32 tanks and slap on fuel pipes as needed. You could even hook them up as drop tanks for a measly 0.05 ton decoupler apiece. Presto, similar fuel to an XL torus and no excess dead weight.

But that's ugly.

These tanks look much nicer, as well as being an interesting change of pace, but I don't want to be forced to severely nerf my craft performance-wise just for aesthetic purposes.

So, if you say it would require better materials than what the stock tanks are made of to construct them... then let there be titanium alloy. Honestly, if the "realistic" mass ratio wasn't *quite* so bad in comparison to stock tanks, like around 7:1, I wouldn't bother messing with it. In fact I might set mine somewhere around there instead. Just because "these particular tank models" (which are admittedly not an expert's work, even if I do like how they look) have extraneous stuff like bicycle rims doesn't mean a more realistic design would. It's just pixels. So if it's theoretically possible to design a toroidal tank with a better mass ratio, then why not fiddle with it to make it so?

[taniwha]

You know, I just noticed a flaw in my calculations. I used 9mm walls on a 0.5m radius tank. That 9mm is from the 1.25m radius Jumbo-64. A quick pass through python (my calculator) indicates the mass ratio should be close to 8.8 (actually higher because I didn't adjust the capacity by the reduced wall thickness).

I now need to do another pass through blender to get accurate volumes :/. But not tonight: it's nearly midnight and I'm tired.

More simple calcs in python indicate 9.01...

>>> (14.8375+0.6082*0.6)/(0.6082*7.8*0.4)

8.011460889214918

0.4 is from 0.5/1.25 and 0.6 from 1-0.4. Other numbers from TO-L-1 in the second link.

BTW, I am not admitting defeat, but claiming victory... over the problem itself . It bugged me that the ratio was so bad when everything else I had done showed it shouldn't be, yet the math showed it was. The math was right, just the tanks were seriously over-engineered. Built like tanks, I guess.

[edit]add 1 to the above ratio to get the mass ratio.

[Tallinu]

Hah, I guess this whole crazy discussion had some unexpected benefits then!

Sleep well in your silly timezone where it's midnight instead of morning, and congrats on catching the problem. I look forward to seeing the revised stats.

[AntiMatter001]

I personally love MFT, but if you don't have it installed the tanks in this mod will default to containing only Liquidfuel/Oxidizer. You just won't be able to switch out the contents unless you change the config manually.

oh... but will there be rcs tanks or will i have to re-config them into life? (like copying the .cfg then renaming resources and such to get what i want)

[Talisar]

oh... but will there be rcs tanks or will i have to re-config them into life? (like copying the .cfg then renaming resources and such to get what i want)

You'll have to manually copy and configure a separate tank config file to make RCS versions if you don't use MFT.

[JTG]

Toroidal and sphere tanks alike look nice on stations and provide a handsome amount of storage. Nice idea!

[DivisionByZero]

I just want to double check something:

This mod includes "storage tanks" in the utility section of the VAB. However, I can't seem to do anything with them except change a texture. Is it the case that I need modular fuel tanks mod to change those tanks to hold whatever it is I need?

Also, is this able to handle new resources from other mods? I'm thinking specifically of the KSP Interstellar mod and making some alternative tanks for argon/etc.

Thanks on both counts. Love the mod, by the way.

Cheers,

[Talisar]

You will be able to fill those storage tanks with resources for TAC Life Support and/or Extraplanetary Launchpads as long as you have that mod and Modular Fuel Tanks installed. Otherwise they will be just dead weight.

You could easily set them up to hold resources for KSPI or any other mod, either through a direct config edit or a ModuleManager config. Or, if you'd like to be able to adjust the contents as you need them in-game, you could make up a custom tanktype for MFT that includes those resource definitions.

Glad you like the tanks

[DivisionByZero]

Well, I was trying to remake something like this ship and the spherical tanks are perfect.

[Aedile]

I think the Toroidal Tank Hub XLarge and Large need update for the node definition size. They seem to be size 1 which makes them somewhat floppy...

[JeBuSBrian]

They are very floppy, and my ships fall apart when using them. But I suppose if the node sizes increased, it'd be that much harder to actually place tanks on the nodes, since they're so close together as is.

[Talisar]

I think the Toroidal Tank Hub XLarge and Large need update for the node definition size. They seem to be size 1 which makes them somewhat floppy...

They are very floppy, and my ships fall apart when using them. But I suppose if the node sizes increased, it'd be that much harder to actually place tanks on the nodes, since they're so close together as is.

Yeah, the larger nodes do make attachment very difficult. Starwaster is using a plugin on his on his MTV Copernicus pack that lets the nodes be size 0 in the VAB and then resizes them to be the appropriate size when launching. I need to check that out and see if it would help with this.

[Tallinu]

It would be nice if there was a way to 'squash' the nodes from spheres into... erm... flattened spheres? The shape of some pills, like advil. And not just visually, but to make them have a smaller grab range along that axis, so that closely spaced nodes wouldn't be as difficult to attach things to, without losing the visual representation of the node size. But if all else fails, making all nodes appear size 1 or even 0 would be far preferable to size 2 and 3 nodes overlapping so much that they're a pain to connect things to like they are currently. I wonder if that could be done with a VAB editor plugin with a toggle function, so you could enable or disable it at will...

[Justin Kerbice]

A tips for nodes attachment issue in the VAB: put the camera in order to have node side in front of the point of view like:

_______________

/-|\

\-|/ part

|-----------------

node

put the part to attach the same way and be easy with snapping.

this way you avoid to become =>

[Tallinu]

If you're talking about parts that can surface-mount and having trouble getting them to attach to a node instead of a surface, like docking ports, I figured that trick out ages ago. Doesn't hurt to restate it, of course. I was talking about parts like the hubs with three large nodes that overlap significantly though. It's still possible to get parts like the toroidal tanks to attach to the center node, but it's very, very tricky unless you attach parts to the top and bottom nodes first (or at least whichever of them is taking higher priority over the center node).

[Justin Kerbice]

If you're talking about parts that can surface-mount and having trouble getting them to attach to a node instead of a surface, like docking ports, I figured that trick out ages ago. Doesn't hurt to restate it, of course. I was talking about parts like the hubs with three large nodes that overlap significantly though. It's still possible to get parts like the toroidal tanks to attach to the center node, but it's very, very tricky unless you attach parts to the top and bottom nodes first (or at least whichever of them is taking higher priority over the center node).

It's not only for surface mount part (don't know what's wrong but with my testbed 0.23 install, I have plenty of node connection issues either in SPH or VAB. For engines, fuel tanks or batteries for example. Most of the time, the zealous idiot editor/node snap do a real mess and I have often two batteries merged together which are impossible to disassemble.

It mostly occurred with large node which have an "OP" gravitational snapping SOI I might say, the big nodes attract everything in sight like a black-hole.

We should have a key to switch from every nodes on target part, like the [ ] for switching vessels in view.

To go back to your issue, I see what you mean and, from my point of view, it's the same node snap issue as I describe above (maybe I was not clear in the first place). The node select by key may be the easiest feature tu add by plug-in if node already done by editor extension plug-in.

[JeBuSBrian]

Yeah, the larger nodes do make attachment very difficult. Starwaster is using a plugin on his on his MTV Copernicus pack that lets the nodes be size 0 in the VAB and then resizes them to be the appropriate size when launching. I need to check that out and see if it would help with this.

This would be great. I simply cannot get the toroidal tanks into space, currently. No matter how many struts I apply, at some point during the launch, they just go floppy. Still attached to the ship, but wobbling around like crazy, throwing off the center of mass, making it impossible to fly.

[taniwha]

Talisar: I've redone the toroidal tank balancing. I used the same prodedure as here (more or less: I took some shortcuts as I needed only the new thicknesses, thus I reused the values that wouldn't change). The new values are as follows:

Name    Capcity   Shell  Dry Mass
TO-S-1   0.2374  0.0039   0.0307
TO-M-1   3.7093  0.0616   0.4806
TO-L-1  15.1932  0.2524   1.9687
TO-XL-1 51.2771  0.8519   6.6444

Capacity and Shell volumes are in m³, Dry mass is in tons.

This gives mass ratios of 8.717. Not quite as good as cylindrical tanks, but I suspect that's the extra details causing the inefficiency. Still, much better than the previous 4.128

[Talisar]

Thanks for that taniwha! I'll adjust the values and push an update. I'll also adjust the node sizes on the hubs and toroidal tanks at the same time.

[Tallinu]

This gives mass ratios of 8.717. Not quite as good as cylindrical tanks, but I suspect that's the extra details causing the inefficiency. Still, much better than the previous 4.128

Hooray!

[Talisar]

Minor update HERE. Main post updated as well.

Changelog:

v3.12

- Updated masses for toroidal tanks, they are much lighter now (taniwha again to the rescue!)

- Increased node sizes for toroidal tanks and hubs to take advantages to changes in joints introduced by the KSP v0.23.5 update.