Jump to content

[WIP][1.8.x] SSTULabs - Low Part Count Solutions (Orbiters, Landers, Lifters) - Dev Thread [11-18-18]

Shadowmage

By Shadowmage
in KSP1 Mod Development

 Share

Recommended Posts

Shadowmage

Shadowmage

Posted
  • Author
Posted
  On 6/2/2017 at 5:49 PM, CatastrophicFailure said:

Erm... were there supposed to be images there? I'm not seeing anything...

Expand  

Erhmm.. yeah... was some copy/pasted excel tables.  Which looked like proper tables in my post (and preview)... but apparently the forum software is not actually WYSIWYG (more like You get Random STUff, No What you See Or Posted).

Link to comment
Share on other sites
Shadowmage

Shadowmage

Posted
  • Author
Posted

Oh, wow... what wonderfully (terrible) forum software we have.  (this text was properly tab aligned in the code spoiler/preview) 

Constant DV target (5k) for constant payload (10t) (note Total vessel mass)
Volume (m3)	Density(t)	Mass	MassFrac	TankMass	payload	Dry			Total		ISP		dV	
54.5		1			54.5	0.12		6.54		10		16.54		71.04		350		5004.193748		KLOX
160.5		0.0708		11.3634	0.24		2.727216	10		12.727216	24.090616	800		5007.649442		LH2
148			0.29		42.92	0.24		10.3008		10		20.3008		63.2208		450		5014.753103		HLOX
										
Constant fuel volume with constant payload
Volume (m3)	Density(t)	Mass	MassFrac	TankMass	payload	Dry			Total		ISP		dV	
100			1			100		0.12		12			10		22			122			350		5881.511993		KLOX
100			0.0708		7.08	0.24		1.6992		10		11.6992		18.7792		800		3713.904417		LH2
100			0.29		29		0.24		6.96		10		16.96		45.96		450		4400.87607		HLOX
										
Constatn fuel volume, no payload
Volume (m3)	Density(t)	Mass	MassFrac	TankMass	payload	Dry			Total		ISP		dV	
100			1			100		0.12		12			0		12			112			350		7669.038893	KLOX
100			0.0708		7.08	0.24		1.6992		0		1.6992		8.7792		800		12888.20327	LH2
100			0.29		29		0.24		6.96		0		6.96		35.96		450		7249.614337	HLOX


or, in image format, as apparently that is the only thing this forum can handle

1 = Constant target DV with constant payload (5kDv / 10t)  (note total vessel mass from the 'Total' column)
2 = Constant volume with constant payload (100m3 / 10t)
3 = Constant volume with zero payload and massless engine (theoretical max)

qBvO3rd.png

All were done using 0.12 for the KLOX tank mass fraction, and 0.24 for the zero-boiloff mass fraction.  (stock uses 0.125 mass fraction on tanks; 8:1=1/8=0.125)
 

Link to comment
Share on other sites
RedParadize

RedParadize

Posted
Posted
  On 6/2/2017 at 5:28 PM, Shadowmage said:

Would you mind posting the stats for the craft that you are having problem(s) with?  As all the math points to LH2 based fuels being quite a bit better, even with the 'heavy' ZBO tanks)

Expand  

I will do that tonight or somewhere this weekend. I will also make sure I have nothing interfering, I have allot of custom stuff and patch...

Link to comment
Share on other sites
RedParadize

RedParadize

Posted
Posted

@Shadowmage Ok I ran some test without my personal patch. First of, sorry if I wasted your time, your graphic show you put much more trough into this than me. Sorry again (I am Canadian, can't resist to say sorry... sorry about that!) look like I was using lightweight tanks for the LFO. I didn't realize I changed it as that's what I use 95% of the time. Obviously that make a huge difference since Lightweight and ZeroBoiloff are the two extreme in term of weight.

I didn't ran with massless engine as you did, nor did I calculated everything as you did, its definitively the way to go, but I preferred to use ingame engine in my test as its what I will end up building my rocket with after all. So I put a single RS-25 (5.75t) vs 5 Merlin-1d (5.5t total), Close in mass but not equal, but Merlin-1d is not top LFO ISP but have a huge TWR bonus. I adjusted dV using only diameter and a tad of V.scale:

LFO Lightweight                Merlin-1d (5.5t total) Isp=320

Playload Tankage mass Fuel mass Total Mass delta V
20t+5.5t 2.848t 27.318t 53.692t 2001
20t+5.5t 10.493t 93.279t 129.299t 4011
20t+5.5t 47.293t 420.383t 493.202t 6003
                                                                         

HLOX ZeroBoiloff             RS-25 (5.75t) Isp=452

Playload Tankage mass Fuel mass Total Mass delta V
20t+5.75t 5.352t 17.842t 48.970t 2008
20t+5.75t 20.320t 67.736t 113.832t 4007
20t+5.75t 162.565t 541.883t 730.224t 6007
                                                                                   

Most case scenario everything is as it should, but looks like at some point between 4k and 6k dV LFO become better. Now that I see this, I am starting to believe its actually the lightweight tank that is op. Note that with regular tank for LFO the total mass would be 1287t for 6000dV.  Problem is, lightweight tank actually exist and as you mention ZeroBoiloff do not.

The need for super high dV without staging is limited to reusable interplanetary tugs, so its not that much of a issue. On my personal patch, I will probably end up buffing Low/Zero Boiloff a bit and nerf lightweight tank max temp and collision resistance allot.

 

Here is my test rig craft text file. All tanks setup are included. Yellow=2000dV Orange=4000dV Red=6000dV and the blue tank is the test weight.

 

  Reveal hidden contents

 

Link to comment
Share on other sites
Jimbodiah

Jimbodiah

Posted
Posted

BTW... The new craft repository thread can now be found here:

I've started adding stock career builds as well (which also work in SSRSS, which I am actually using). After I get the initial ships set up to stock scale, I will start adding some for RSS.

Link to comment
Share on other sites
Shadowmage

Shadowmage

Posted
  • Author
Posted

Updated release is available:

https://github.com/shadowmage45/SSTULabs/releases/tag/0.6.36.136

A few small bug fixes and updates to some config files.  Includes an updated version of KSPWheel, so the SC-E should be functional again.

WARNING - The SC-C-ICPS and SC-C-HUS parts are being deprecated and will be removed in a near-future release.  Please begin converting your craft now, as when the change happens existing craft using those parts will break; the updated MUS fuel tank should serve as an adequate replacement for the parts being removed (old models and configs might be made available for a short while upon request).

Link to comment
Share on other sites
Jimbodiah

Jimbodiah

Posted
Posted

That is the last 1.2.2 release, it works fine; I still run a 1.2.2 RSS copy with it.

I just installed the latest build and I discovered a hidden gem in there; the SSTU-HECS. I was wondering if you would ever make a satellite we could use on contract missions. Good to see you are looking at it :wink:

Link to comment
Share on other sites
Shadowmage

Shadowmage

Posted
  • Author
Posted
  On 6/5/2017 at 2:35 PM, Jimbodiah said:

I just installed the latest build and I discovered a hidden gem in there; the SSTU-HECS. I was wondering if you would ever make a satellite we could use on contract missions. Good to see you are looking at it

Expand  

Ahh, yeah, I've been playing with that part in my dev games for quite awhile now.  Wasn't quite sure how well the feature set would play out for that part, but it ended up working quite well.

It used to get removed from the releases by my build script; but I haven't updated the build script to work with the new repository setup, so have been manually packing things and apparently forgot to remove it.

I'm actually not sure if the part even works at the moment; I do remember it had some minor config errors that needed cleaned up originally, but that was before I started all the model/recoloring rework stuff, so it may well have been broken during the updates.

The concept though will probably stick around and be developed a bit more in the future.  Probably will come up with a few other probe body designs, some more oriented towards integrated solar panels, others using RTGs.  Really want to come up with a 'science package' that can be slapped on them, something that gradually unlocks more experiments as the tech tree is unlocked.  Will likely also come up with some integrated probe-style RCS blocks to include in the part(s).  And finally, I'm still undecided on if I should include engine functionality in some of the probe-core bodies; with them being a mostly integrated part, and final part of a 'payload', they are one of the best candidates for integrating engines.

If/when I get to the 'service module' parts, they will likely use mostly the same setup, only with larger bodies/tanks/engines, and possibly a few specialized features on each.

 

  On 6/3/2017 at 4:44 AM, RedParadize said:

@Shadowmage Ok I ran some test without my personal patch. First of, sorry if I wasted your time, your graphic show you put much more trough into this than me. Sorry again (I am Canadian, can't resist to say sorry... sorry about that!) look like I was using lightweight tanks for the LFO. I didn't realize I changed it as that's what I use 95% of the time. Obviously that make a huge difference since Lightweight and ZeroBoiloff are the two extreme in term of weight.

 

Expand  

Thanks for taking the time to put that info together, I'll play around with that a bit to see where it all sits in my calc sheets.

Two quick things going on that would explain part of the discrepancy / imbalance:

1.)  Comparing light-weight tanks vs. zero-boil-off.  That would be like comparing an ironclad hull to a cheap aluminum hull; there really can be no comparison made other than 'one is far heavier than the other'.  As you point out, there currently can be no real-world-based comparison between the two.  Lightweight tanks exist, and have flown (and are what I based the stats on).  ZBO tanks are... speculative at best, and mostly fictional in practice.

2.)  Using HLOX in the zero-boiloff tank.  The way the code works (and KSP), is that the zero-boil-off penalty is applied to the volume/resource-mass for the entire tank, not just the cryogenic portion.  So you are paying the 'zero-boil-off' penalty on your OX tank as well (and as it is more dense, it is quite a big penalty, as the ZBO mass is a percentage of resource mass).  The best (and only efficient) use of the ZBO tanks is for pure LH2.  This is actually quite a large mass penalty being paid for insulating the LOX, and would make a far bigger difference than you might think (about the same magnitude of difference between the 'standard' and 'lightweight' tanks).

 

Going forward... yes, the lightweight tanks are a bit OP for stock KSP... they are geared more towards late-game RO use (or other large rescaled systems with higher dV budgets), I might 'nerf' them a slight bit for stock KSP use.  The zero-boiloff mechanics/calculations will not be changing however (doing so would require that 'VolumeContainer' be aware what a cryogenic fuel is, and what 'tank type' to assign to non-cryo fuels in a mixed tank setup... currently it doesn't care, all boiloff/cryo functions are handled by a separate module), so the best setup for a zero-boiloff-HLOX type setup would be to use a separate 'lightweight' tank for the OX, and a ZBO tank for the LH2 (yes, two separate tanks, one for each fuel); this would ensure you are only paying the ZBO penalty on for the tank/fuel that actually needs it.

Either way, I'll put together a bit of a better comparison sheet using more solid data for its base (the stuff I did the other day I guessed/estimated quite a bit regarding densities), and will do a good examination of the fuel tank setups and balances for a near-future update.  At the very least there is some room for some of the tanks to have a better volume use (lower volume loss, e.g. spherical tanks), and depending on code might be room for some additional mass and/or boil-off adjustments.

 

Link to comment
Share on other sites
Jimbodiah

Jimbodiah

Posted
Posted
  On 6/5/2017 at 3:19 PM, Shadowmage said:

Really want to come up with a 'science package' that can be slapped on them, something that gradually unlocks more experiments as the tech tree is unlocked.

Expand  

I made apart using your radial materials bay that holds all the other stock (and BDB/DMagic) experiments and even tried using the upgrade system to unlock the new experiments with the techtree, but never got that to work. I don't think the upgrade system allows you to add modules, only change them? But it would be super if you got that working, it would solve a lot of esthetic problems with all the fugly stock parts strapped onto ships.

A satellite for contracts and communication relays would be on my short list for SSTU parts, together with an ATV style resupply ship :wink: 

Link to comment
Share on other sites
tater

tater

Posted
Posted

An ATV would basically be a COS part with the adapters set appropriately, plus a modular service module, no? Seems like it would need a new adapter part to get the "beak" look of the docking port. Perhaps a new "station" part that is a cylinder with blankets instead of metal texture? 

The COS parts are very slightly smaller than the ATVs to scale (which would be 2.88m in kerbal scale). Anyway, with a new adapter and a SM... Maybe the upper stage tanks could get integral solar as an option?

Link to comment
Share on other sites
tater

tater

Posted
Posted

Low part count matters, but part of the fun of KSP (to me, anyway) is creating novel craft. So I'll take a few extra parts in the name of flexibility.

I would much rather have the option to use a tank of parts with an adapter at the top (Perhaps MFT-P, where P is for "pressurized" and it only exists to hold parts, science, LS, etc) and a series of scalable SMs. This allows making all sorts of logistical crafts, but the adapters and variants are designed such that it's possible to make the ESA ATVs (cloth variant for both tank and SM, the ATV style top with small docking port, integral solar, etc.

RCS in general needs some options, the stock parts are unattractive, and the stand-alone SSTU ones right now are too large.

Link to comment
Share on other sites
Shadowmage

Shadowmage

Posted
  • Author
Posted
  On 6/3/2017 at 4:44 AM, RedParadize said:

@Shadowmage Ok I ran some test without my personal patch.

 

Expand  

 

Bit more info.  Concerning ZBO tanks, and HLOX.... it is almost mandatory that you setup the LH2 in its own separate fuel tank if you want to see any of the advantages of the higher ISP from the engine.

Hopefully this table will tell the tale...   The total at the bottom of the 'Total Mass(wet)' column would be the full vessel mass.  Note the massive difference in vessel mass between the 'full ZBO' and 'split ZBO' tank setups (with split representing two actual fuel tank parts, one for LH2, one for OX).

xrT7Sss.png


Now, when running the numbers for pure LH2 setups (e.g. for nukes), there are no such oddities.  Nor when running the numbers for pure KLOX.

KLOX setup for comparison (using the 320ISP Merlin from above):

aQARTWT.png


Note the -massive- difference in vessel mass between the 'standard' and 'lightweight' tanks on the KLOX side (lightweight is far, far, better at that range).

Also note, that when using a proper 'split-ZBO' tank setup, the HLOX craft comes out far ahead of the KLOX setup (~150t vessel vs. ~305t vessel); but if using the standard 'full-ZBO' tank, HLOX loses out considerably to a 'lightweight' KLOX tank setup (716t vs 307t), but is still far better than a 'standard tank' KLOX setup (716t vs 1286t).

 

Honestly, aside from the quirk of the 'zero boiloff' penalty for the oxidizer in the tanks, I would say the 'zero boiloff' and 'lightweight' balancing is right about where they need to be.  Just remember that you might need separate LH2 tanks anytime you want ZBO for HLOX setups, and you should be good (or rather, just remember that ZBO tanks also penalize you for non-cryo fuels).

 

Edit:  Note -- I may have miscalculated the advantage of the 'lightweight' tank a bit, using 0.5x the standard mass fraction; when looking at my config files it is actually 0.75x, so the advantage of 'lightweight' is not as great as posted above.

Link to comment
Share on other sites
sjsharks39

sjsharks39

Posted
Posted
  On 6/5/2017 at 2:35 PM, Jimbodiah said:

That is the last 1.2.2 release, it works fine; I still run a 1.2.2 RSS copy with it.

I just installed the latest build and I discovered a hidden gem in there; the SSTU-HECS. I was wondering if you would ever make a satellite we could use on contract missions. Good to see you are looking at it :wink:

Expand  

I can't get it to load. I was getting errors on 1.2.2 with the current version of SSTU, so I came here looking for answers. Sounds like I have to revert... no problem. Found this link, downloaded, and installed it. I'm getting an infinite loading screen that no longer gets errors, but the wheels keep spinning at loading SSTU and no progress happens. Any ideas?

Link to comment
Share on other sites
Jimbodiah

Jimbodiah

Posted
Posted

@tater  There is an ATV .craft on the repository thread. It has a small tank for the SM part and I added a medium COS hab for the cargo ection (own patch to have different loadouts, mainly LS and some hypergolics).

sstu_atv1.jpg
here with a small cargo section.

Link to comment
Share on other sites
tater

tater

Posted
Posted (edited)

Still working on summer schedule for kids... so I can mess around a little during the day:

3ppW51m.png

MUS tank, a few stock MP engines (If I had a couple more mods in my SSTU 1.3 test build I could have made it better), and a probe core (does the upper stage have that? I forgot to check), the nose is another tank with an adaptor, and I clipped it, plus the docking port. So way too many parts at 12. Plus recolored white (dull for the lines between panels)

 

Edited by tater
Link to comment
Share on other sites
Jimbodiah

Jimbodiah

Posted
Posted (edited)
  On 6/5/2017 at 5:34 PM, sjsharks39 said:

I can't get it to load. I was getting errors on 1.2.2 with the current version of SSTU, so I came here looking for answers. Sounds like I have to revert... no problem. Found this link, downloaded, and installed it. I'm getting an infinite loading screen that no longer gets errors, but the wheels keep spinning at loading SSTU and no progress happens. Any ideas?

Expand  

You need to remove all the new SSTU and extra folders that came with it (kspwheels, community resource pack and module manager 2.8.0 file). The re-install the old SSTU and module manager 2.7.5. Module Manager uses the latest version, so make sure you don't have 2.8.0 installed as well.

Edited by Jimbodiah
Link to comment
Share on other sites
Shadowmage

Shadowmage

Posted
  • Author
Posted
  On 6/5/2017 at 4:47 PM, tater said:

RCS in general needs some options, the stock parts are unattractive, and the stand-alone SSTU ones right now are too large.

Expand  

I've been giving quite a bit of thought on how to pull off some 'better' RCS ports.  While I don't think I can fix the part-count issue (aside from offering integrated ports whenever possible), I'm fairly certain I can pull off some 'modular part' style RCS ports -- allowing for configuring the number of ports and layout (positions/angles, configurable through config like engine layouts), changing of the body style, port style, textures, and optional attachment arm selection. Oh, and scale :) (and thus thrust); possibly with an additional range of control for thrust as well as thrust scaling with model scale.

Still early in the thinking stage, could see it being developed along with the lander core stuff and/or service-module stuff, but couldn't say much more than that.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
 Share
Go to topic listing
×
×
  • Create New...