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

[blowfish]

Just out of curiosity, is there a reason why including 7zip in the repo would be a requirement for using it?

[Shadowmage]

Just now, blowfish said:

Just out of curiosity, is there a reason why including 7zip in the repo would be a requirement for using it?

Technically, no.  It could, in theory, reside in an external location.

For me personally -- I have a personal requirement to keep all 'necessary' files with a repository.  If 7-zip is 'necessary' to build the app, it needs to be in the repo.  If some .dll/s libs are needed for building... those have to be in the repo as well (or really should, even though in this case I've had to leave them out for licensing issues as well... GD copyright BS).

This is because I work on the mod from multiple locations/multiple computers -- I need to be able to simply clone/pull the repo and have it ready to go.  If I have to go fishing around the internet to find libraries, dependencies, or apps needed to build it... then the repo has failed at doing its job and I might as well lug things around on a flash-drive.

 

Not to worry though -- I'm building a c# .exe that will have .zip functionality built in.  Kind of a micro version of ant/etc; will take a very simple build-config-file, and, well.. build things (more packaging though, as everything will already be built/compiled by the time it is ran).

[blowfish]

Multiple locations/computers?  Heh, try 3 different platforms.  I managed to get B9 Part Switch compiling on my Windows desktop, OSX laptop, and Linux CI environment (release always happens from the CI environment though which makes things a bit easier, I only have to be able to compile and run tests on the other two platforms)

[Shadowmage]

@tater @Rodger @blowfish

Would you guys (at least one..) mind giving the following .zip a test to see if it looks packed correctly on non-Windows platforms?

https://www.dropbox.com/s/u1s4koxbdavzm2n/SSTU-testPackage.zip?dl=0

(is a repack of yesterdays release, using updated build scripts/system)

[tater]

Will do!

Doing it right now for ya.

Works fine.

[blowfish]

10 minutes ago, Shadowmage said:

@tater @Rodger @blowfish

Would you guys (at least one..) mind giving the following .zip a test to see if it looks packed correctly on non-Windows platforms?

https://www.dropbox.com/s/u1s4koxbdavzm2n/SSTU-testPackage.zip?dl=0

(is a repack of yesterdays release, using updated build scripts/system)

Looks good to me!

[Shadowmage]

Thanks for the confirmation guys; will do a full repack on this release, and re-upload it to github/etc this afternoon.

 

Edit:  Done, and done.  Zip repacked, and re-uploaded.

Edited September 5, 2017 by Shadowmage

[Jacke]

17 hours ago, Shadowmage said:

Looking into a better packing utility now.  I previously used 7-zip, but it has licensing problems that prevent me from keeping it in the release repo.  So I was looking for something a bit more built-in, which lead me to the PowerShell commands.  Apparently these use a different .zip algorithm/encoding that OSX struggles with.

IANAL, but....

Looking at 7zip,  it's licenced under the GNU LGPL version 2.1 of the License, or (at your option) any later version, with the unRAR restriction (take a look at the link), which say the portions of code used to unRAR can not be used to recreate the RAR compression code.  Looking at the GNU LGPL (along with the GNU GPL), I think that means....

You can include 7zip in the repository and even the download if your licence identifies what licence 7zip is under, without restricting what licence you want to apply to your work in the repository.  That was the whole point of the LGPL, to make things more flexible for including free software code in a separated way (ie. separate libraries or programs) with non-free software code.  It should apply similarly here.

Edited September 6, 2017 by Jacke

[Shadowmage]

Going to start working on the heat-shield fix up this week.  But going to need some info in order to balance things.

• Max mass for each diameter of MHS.  This should likely be a direct function of the size of the shield, and is a value that should be able to be calculated for all sizes given the parameters for a single size and the calc function. (HS area scales with the square of diameter;  mass/area roughly determines drag, and thus the ability to slow down appropriately).  Going to call this parameter 'mass loading' - will be specified by tons-per-square-meter.
• Re-entry profile.  -I- always use 30km as my periapsis for re-entry, but perhaps this should be adjusted (30km is the -highest- you can get a reliable capture when returning from Mun/minmus; much higher and you usually end up making a 2nd pass)?  What Pe should I use during testing for each orbit/shield size combination?  Keep in mind realistic g-loading; I don' t want re-entry profiles that subject things to more than ~4-5g preferably.
• Anything else I need to balance these with/for/against?

[Calvin_Maclure]

Just now, Shadowmage said:

Going to start working on the heat-shield fix up this week.  But going to need some info in order to balance things.

• Max mass for each diameter of MHS.  This should likely be a direct function of the size of the shield, and is a value that should be able to be calculated for all sizes given the parameters for a single size and the calc function. (HS area scales with the square of diameter;  mass/area roughly determines drag, and thus the ability to slow down appropriately).  Going to call this parameter 'mass loading' - will be specified by tons-per-square-meter.
• Re-entry profile.  -I- always use 30km as my periapsis for re-entry, but perhaps this should be adjusted (30km is the -highest- you can get a reliable capture when returning from Mun/minmus; much higher and you usually end up making a 2nd pass)?  What Pe should I use during testing for each orbit/shield size combination?  Keep in mind realistic g-loading; I don' t want re-entry profiles that subject things to more than ~4-5g preferably.
• Anything else I need to balance these with/for/against?

Great stuff! Quick question about RO. It still says in the OP that RO is not compatible with SSTU. 1- is that still the case 2- eta to fix (just wondering)

CM

[Shadowmage]

6 minutes ago, Calvin_Maclure said:

Great stuff! Quick question about RO. It still says in the OP that RO is not compatible with SSTU. 1- is that still the case 2- eta to fix (just wondering)

CM

I do not use RO, so I will never be officially supporting it (until/unless my use of it changes).  Any sort of 'compatibility' or 'support' needs to come from the RO end of things.  (which means, basically, that you need to ask in the RO thread)

However I do still see quite a few people with RO installs using some SSTU parts; so I think it is less of a general 'incompatibility', and more of a 'some parts aren't balanced/patched for RO'.

[blowfish]

All the SSTU-RO compatibility is on the RO side.  Some folks have worked to make a lot of SSTU compatible, though I wouldn't say 100% yet.

[Calvin_Maclure]

2 minutes ago, blowfish said:

All the SSTU-RO compatibility is on the RO side.  Some folks have worked to make a lot of SSTU compatible, though I wouldn't say 100% yet.

Good to know! So, as I understand it, simply downloading RO will configure SSTU (and Kerbal Foundries I suspect?)?

[blowfish]

11 minutes ago, Calvin_Maclure said:

Good to know! So, as I understand it, simply downloading RO will configure SSTU (and Kerbal Foundries I suspect?)?

No idea about KF, and you will definitely find SSTU parts that have compatibility issues.  But at the very least the engines and tanks should work.

[Mike`]

4 hours ago, Shadowmage said:

Going to start working on the heat-shield fix up this week.  But going to need some info in order to balance things.

I already started to play around a bit with the current settings. First i adjusted the weights of the shields, based on the apollo heat shield which at 3.9m diameter apparently weighted about 850kg. Thus i set the mass of the 3.75m medium shield to 0.4 and its ablator to 450, yielding 850kg for that part.

Also i changed the massMults of the different types to linear, so a light shield is half as heavy as a medium, while a heavy is double the weight of a medium.

Next, i took a  look at the heatCurve, and wondered about its many points. I plotted it, and noticed it's not interpolated very smoothly, i confirmed this in an in game curve editor, so i created a new curve out of 6 points which is smooth and easier to scale because you don't have to edit that many points. I can't attach images here, will post them on the github issue.

After that i balanced the light and medium shield curves and the ablation efficiency, and i made quite some progress. Especially the medium curves i looked a bit closer at, and made 2 curves for now.

One which has a bit more headroom (to overpower the ablation for the 1.25m shield from outer Kerbin SOI, you need 1,5t and a -40km periapsis or 2t and 0km periapsis), and one with a bit less (already exploded with 940kg, eg not much more than the pod, at a periapsis of 0km).

I'll post screenshots and details of the curves at the github issue, i guess.

Edit: Curves posted in https://github.com/shadowmage45/SSTULabs/issues/553

I've also updated that post with some experimental results. Now, does anybody have an idea what might be good test/balance conditions/targets for the heavy and extra heavy shields?

Edited September 6, 2017 by Mike`

[DocRockwell]

On 9/4/2017 at 5:25 AM, Rodger said:

Who likes SSTU and Ven's Stock Revamp??

So far Poodles, Skippers, and mounts have been SSTU-ed
https://github.com/Rodg88/SSTUVens

Will this work with 1.2.2 (playing RP0)? 

 

Edited September 9, 2017 by DocRockwell

[Rodger]

2 hours ago, DocRockwell said:

Will this work with 1.2.2 (playing RP0)? 

 

No idea, but if the engine module SSTU hasn't changed much since a 1.2.2 version, it should?

[Mike`]

In case like you feel like modeling an engine again, what about the MR-80B? A small, monopropellant lander engine used to land curiosity on mars.

http://www.rocket.com/files/aerojet/documents/Capabilities/PDFs/Monopropellant Data Sheets.pdf

some more data and 1-2 pictures

[Jimbodiah]

Not sure on stock systems anymore, but I thing I used to do a 35km for re-entry. I tend to play RSS only now and use 62km there for high velocity returns or 55km for LEO.

[Shadowmage]

Finally unwrapped+baked the SC-V (VA Capsule) parts (first real modeling work I've been able to do in quite some time):

Will be working on textures + masks, and hopefully move this part into 'finished' status for the next update (after I fix up the configs / do some balance work on them).

 

Work is also continuing on the heat-shield rebalance / code updates.  Spent the better part of the last week trying to track down what looked like a bug in my thermal integration code.... that turned out to be a bug in the stock thermal conduction code.  Fun times (not really)...  Now that the apparent bug has been... erm.. located... I can finally start working on the balance work for the heat shields.  Most heat-shield parts should end up with a bit more tolerance / capability, as well as a slightly wider range of usable reentry profiles.  Shouldn't be a craft/vessel/save breaking change, but will know more once I get working on the balancing end of things.

[Shadowmage]

On 9/11/2017 at 3:38 AM, Mike` said:

In case like you feel like modeling an engine again, what about the MR-80B? A small, monopropellant lander engine used to land curiosity on mars.

http://www.rocket.com/files/aerojet/documents/Capabilities/PDFs/Monopropellant Data Sheets.pdf

some more data and 1-2 pictures

That .pdf is an excellent find as far as RCS / small thrusters are concerned.  I've seen quite a few of those RCS blocks on real-world spacecraft, but had never seen any actual design specs for them.

As to whether I will be making those engines (thrusters?) -- a big maybe.  I would have to see some good use-cases for their KSP-scaled-stats (which will be ~40% of the listed thrust on the spec sheets).  I could possibly see some of them being used for small 'realism' styled landers (very small) or satellite cores.

[blowfish]

On 9/11/2017 at 2:38 AM, Mike` said:

In case like you feel like modeling an engine again, what about the MR-80B? A small, monopropellant lander engine used to land curiosity on mars.

http://www.rocket.com/files/aerojet/documents/Capabilities/PDFs/Monopropellant Data Sheets.pdf

some more data and 1-2 pictures

Equivalent PDF for bipropellant thrusters, just for completeness:

http://www.rocket.com/files/aerojet/documents/Capabilities/PDFs/Bipropellant Data Sheets.pdf

[Mike`]

2 hours ago, Shadowmage said:

That .pdf is an excellent find as far as RCS / small thrusters are concerned.  I've seen quite a few of those RCS blocks on real-world spacecraft, but had never seen any actual design specs for them.

As to whether I will be making those engines (thrusters?) -- a big maybe.  I would have to see some good use-cases for their KSP-scaled-stats (which will be ~40% of the listed thrust on the spec sheets).  I could possibly see some of them being used for small 'realism' styled landers (very small) or satellite cores.

I actually scripted the ingame ant engine to be like that thing (the MR-80B, the largest thruster in the pdf), i tried to use your balance and came up with 1.54kN thrust - not much, but i calculated an ingame weight of 14kg, so it's also pretty lightweight and you can use multiple of them. What for? Indeed small, unmanned landers, like it is/was used for IRL. As i removed any stock engine from my game, i didn't have anything for that purpose - the LMAE came closest, but that thing is still comparably huge and heavy for a tiny unmanned mun-lander.

Equivalent PDF for bipropellant thrusters, just for completeness:

http://www.rocket.com/files/aerojet/documents/Capabilities/PDFs/Bipropellant Data Sheets.pdf

The R-40B at 4kN irl thrust also looks very good. Tiny, light, and good Isp for that size.

Edited September 13, 2017 by Mike`

[Shadowmage]

While working through the functions of the SC-V parts (VA spacecraft), I'm still a bit in doubt about the staging setup of the re-entry (what bits get staged off when, where the descent hardware is located).

My first impressions were that everything was in the capsule (parachutes, retro-rockets, rcs, etc).

Then I read somewhere that part of the service module / retro module stayed attached during re-entry and descent, and contained the parachutes, orientation thrusters, and soft-landing rockets.

And finally, now, I'm seeing information that the service module / retro module was in fact jettisoned prior to or early during descent, the parachutes are located in the capsule, and the soft-landing rockets are located on the parachute lines.

http://www.svengrahn.pp.se/histind/Almprog/tksalm.htm

 

Does anyone have any additional information on any of this?

 

Just to dirty things up... this image seems to show that the parachutes / soft-landing rockets were housed in the SM / retro module (which would infer it stays attached through entire descent):

[CatastrophicFailure]

Since the parachute is in the bottom of that module, I'm guessing maybe it gets released when the upper module is jettisoned after reentry?