SSTU - Shadow Space Technologies Unlimited

SSTU - Shadow Space Technologies Unlimited

98.5k Downloads

Engine question and Wiki organization

taterkerman opened this issue ยท 35 comments

commented

Hey, a few things:

  1. What is the scaling for engine thrust/etc for the mod vs reality? I checked a few, and they seem to vary quite a bit, assuming the number in the VAB part description is correct. I was thinking about engines that are missing (since I don't even have stock engines anymore, lol). In game terms, there are many in the 10-30 kN range, then a huge gap. Some 50-100 kN engines would be useful for upper stages.

  2. Assuming the values in the part descriptions are correct, would editing the wiki to fill them in be good (as well as other information if you wish)?

commented

Thrust scaling is RealWorldThrust * 0.64 * 0.64. (or realThrust * 0.4096)

The gap exists... because the gap really exists. The RL10B-2 is at the 'top' of the gap at ~45kn (scaled). The next lowest engine that I've plotted stats for is the RL60 (that isn't even real), with a thrust of ~120kn (scaled).

In order to fill the gap there would need to exist real engines in the 110-290kn thrust range, which I don't think I've seen anywhere. If you can find some candidates, please let me know and I'll fill in the blanks in my data sheets.

Wiki updating -- yeah, would be nice to start getting some of the documentation filled out on the stuff that is stable; engine stats are probably about as stable as anything in SSTU, and probably a good place to start.

commented

Cool, next time I'm sitting around at night with some time (likely the next few days as my wife is on call) I'll fill those in for you and make myself useful :) .

The Indian CE-20 scales to 81.92 kN... supposedly flying later this month. (hydrolox)
indiancryoengine25

So there's 1, anyway, assuming their launch on the 28th goes well---though if it's integrated on a real spacecraft, that's effectively out of development.

commented

Another would be the Atlas II sustainer engine, the RS-56-OSA, 269kN in real life, so 110.18kN, scaled.

A hypergolic variant of the RS-88 is slated for use with CST-100 as the LES motor, but the LOX/alcohol version is 90.11 kN scaled to KSP. New fuel type though, so meh.

commented

The engine descriptions on the wiki list "Thrust," do you want all these to be the vacuum thrust, the sea level thrust, or whichever is the most common usage? Ie: SL for F-1, vac for Rl-10 based upon "Use"?

commented

Alternately, I can change Thrust to ThrustV and ThrustSL (or whatever) and put both.

commented

How are you getting the scaled engine masses? The KSP F-1 masses 11.405 tons according to the cfg file, which is more than the real one. I'm looking at TWR calculations right now.

Currently all liquid engines are entered, but I used the maxThrust value (vac) as I can simply type those in without calculation.

I'm also thinking about organization. Fuel type? Use?

commented

I'm not familiar with the wiki editing, notably the navigation sidebar. The Ship Core stuff is still set to the old naming scheme, so that Apollo=A, Orion=B, and Soyuz=C. It would be trivial to change the A to B, C to A, and B to C on the individual pages, but then I want to make sure the navbar points to the right pages.

I threw in a stub for lander core with the image of all 3 pods, and images in Ship Core General for the RCS and radial science exp.

I'm planning on adding all the engines in the imgur album of final renders over the weekend, so far I added a few as a test.

I might rejigger the text somewhat once I get all the data in (just formatting, etc).

commented

Both would be fine with me.

Have also added you as authorized to make edits on the wiki. Should jsut have to accept the invite and you should be good to go.

commented

How are you getting the scaled engine masses? The KSP F-1 masses 11.405 tons according to the cfg file, which is more than the real one. I'm looking at TWR calculations right now.

Mass is back-calculated from the scaled thrust and 0.25*RealTWR. It is common for the KSP engine to have higher mass than the real-world engine due to this.

All the current stats can be found in the linked spreadsheet:
EngineData.xlsx

Sidebar -- it automatically updates the linking when you change the page names, no need to worry about that. The only links you need to worry about updating are those that are in the body-text of the pages.

SC-A/B/C -- yeah, should probably get those updated to reflect the in-game names :)

Everything else sounds good so far. Thanks for your help on cleaning this stuff up, I've not had as much time to spend on it as I would like.

commented

OK, I'll swap the A/B/C names to reflect the current reality, and work on the TWR stuff at some point.

I can grab the different thrusts (SL vs Vac) out of the game in another window, which will be easier than calculating them. I didn't last night because the right-click is sort of flakey about closing the window if I mouse away from it in the wrong direction, lol.

I'm thinking of doing fuel/propellant to the engines, as well, that or I might header them by that, if that sounds OK, then high to low thrust within that (since you never know how people will use engines in KSP, so booster vs orbital might not be as much of a thing).

So Kerlox engines, then hydrolox, then hypergolic, then LH2 (Nervas).

commented

I added in the 2 Agena hypergolic engines, but the only renders I could find were from a forum post you made without textures applied, so those are stubs ATM (used the images, they just aren't as pretty).

commented

Thanks, noted -- I'll get some updated renders of the LR-81's, and the SuperDracos.

Will also add some renders for the RD-171/180/181 engines.

commented

I forgot the draco engines as well, meant to get all on there, even without pictures. I haven't entirely moved the LFOs together, as they seem to have been in groups by nation, with the Russian stuff at the end... in KSP, that doesn't seem to matter much. I did make headers for LFO/Hydrolox/Hypergolic, though to see how it looked.

Note that the good image I had of the LMDE/LMAE had both together in the same image (from the forum), so I ganged those temporarily.

I'm also adding alt image tags, though I got lazy and didn't finish those yet. For now they say "SSTU F-1 Engine" (for the right engine name for each image). If you have specific text you prefer for the alt tag, let me know here, and I'll make sure they are all labeled. That goes for any image already the wiki, just let me know.

You posted renders on the forum for the 171 and 180, so those are already on there, but the link can be changed if you have a single spot for all of them.

commented

The RL19s ate clearly vacuum optimized, yet some have a high % of their vac thrust at SL in the SSTU specs. Perhaps they should all (certainly the ones with clearly vac optimized bells) have a SL thrust of 0 (I'm looking at all of them because of the wiki, lol).

commented

But that would give the wrong performance in the range that they should be allowed to operate in. It's not that the engine would generate zero thrust at sea level, it's that you would not want to operate the engine at sea level because bad things would happen.

If you want to make all the engines consistent, use the relationship which gives the correct thrust in the operating range, i.e. (thrustvac) - (thrustsl) = (areanozzle exit) * (ambient pressureSL)

commented

I understand that in the game, this is to provide a thrust curve based upon altitude/pressure, right? Some of the RL10s have a pretty high expansion ratio, and yet go no where near zero thrust wise. I'm not disbelieving you at all, but Merlins are outliers with all of them at 0 SL thrust, vs every other vac engine having a non-zero value.

commented

In that case maybe adding more points to the curve is necessary - two doesn't really capture the behavior well. But determining where flow separation should occur is also not trival, and pretty difficult to determine just based on known parameters of the engine.

commented

Really in the case of KSP, it becomes a game balance issue more than a simulation issue. Merlins in SSTU end up being the go-to engine because they appear early, and cluster well n the available tank sizes (since I have no stock engines any more , lol).

So the 0 value doesn't matter that much, I was just curious about it.

commented

SL engine ISP & Thrust on vacuum/upper-stage engines -- The stats that exist on the engines are a result of what information was available, and guesstimate attempts to make them unusable at sea-level.

The problem is that for vacuum engines there is seldom a sea level thrust or ISP specified. As such I guessed at some values that would render the engine mostly useless at sea-level. This is the case with the Merlins, RL10's, and J-2X. In fact the only 'vacuum' engine that I have ever found SL stats for is the classic J-2 (and RS-25, if it counts as a vacuum engine...).

As @blowfishpro points out is not really possible to derive the sea-level stats directly from the vacuum stats with the smattering of information that is available; with a few more numbers I could probably plug the stats into the RPA tool to calculate some theoretical sea-level stats -- but I would need information that is not readily available for some of these engines (chamber pressure, expansion ratios, fuel mixture ratios), and even then it would only give theoretical values.

If you would like to propose a standardized method of statting the vacuum engines' sea-level stats, I would be okay with that. Probably something like 15% of vacuum thrust at 1atm seems appropriate for most of the purely vacuum engines. Some of the hypergolic 'thrusters' could probably be set to 50% of vacuum thrust for their SL stats as most don't appear to have pure vacuum nozzles.

commented

The Merlin vac engines need their SL thrust improved. The worst SSTU upper stage engine in terms of difference between SL and vac thrust is the RL10-B2, and the SL thus is ~19% of the vac thrust. Many others are substantially higher, even well above 75%.

Given talk of attempting to land F9 stage 2, clearly the thrust must be >0 .

commented

Most vacuum engines don't even work at SL, and given the huge nozzle, I'm inclined to believe the Merlin Vacuum engines are no different.

I am mostly sure that attempts to recover the second stage would be non-powered, probably involving an inflatable heatshield and parachutes.

commented

In that case, shouldn't all the game's vac engines have no thrust at SL?

Regarding Flacon 9 resuse, I think the intent is indeed dracos or superdracos, that old video shows the exhaust coming from 3-4 smaller engines around the vac engine bell (which retracts in the video).

commented

Sort of. KSP doesn't really have a way to deal with the way this actually works though. The reason you can't run a vacuum engine at sea level is flow separation - once the exit pressure is far enough below ambient the exhaust will separate from the nozzle wall and become unstable. But if you take the thrust in the range where the engine operates and extend it down to sea level, you will not get zero.

commented

I do think that the nozzle areas would be a good first pass. Someone would have to go into the models and measure them though. I'll see if I have time soon...

commented

The expansion ratio data is in the spreadsheet, and the SpaceX engines are clearly outliers there, with 2X the expansion ratio for the vac engines of typical upper stage engines.

In actual play, the only issue might be the TWR of the stage when it is not yet out of the atmosphere, I only know what the TWR curve ends up looking like at high alt compared to other engines with a similar thrust (which would likely be a cluster).

There is a bit of discussion in the spacex thread, and I was not aware how thin the bell was, so that there could be damage at high atmospheric pressures that would prevent using it entirely (which makes the 0 value make total sense, as the rest is not modeled in KSP at all). The other engines might not have thins limitation due to their construction, and substantially smaller engine bells.

commented

It was pointed out on NSF that the nozzle extension on the Mvac engines is added after testing it without the added bell. Someone suggested that perhaps that part could be jettisoned for landing operations. It's all pretty unknown, maybe we'll see some information after the FH test, if they attempt landing it.

That extension is apparently only ~0.3 mm thick for most of it, which would explain the fragility.

commented

Yeah, detachable was spitballing ideas, anyway, not too serious, as I'm pretty sure it's never been done in RL.

Engine out capability would be interesting, however, and I could see mods that add failure modes using that (it would be cool to watch the other engines trying to compensate).

commented

@taterkerman Nice work on the wiki updates -- engines page is looking good. Should hopefully be able to get you some renders of the rest of the engines sometime this weekend.

Detachable / jettisonable nozzles -- not something that can be done with the current plugins. Would need quite a bit of custom coding to make it doable. Might consider it in the future if/when I update the MEC module -- contemplating adding 'engine-out' capability / enable/disable of each engine position.

commented

Is there a repository for all SSTU images for the wiki?

Technically, yes, but It isn't really used as it should be currently.

A separate repository exists for the wiki that is liked to every main repository. You can clone, push, etc to them just like any other.

Probably something I should look into for long-term hosting of files used on the wiki / etc. Keep them local to where they are being used. (currently half are hosted on the wiki, and most are linked from imgur).

commented

Is there a repository for all SSTU images for the wiki?

Once uploaded to this forum, they have a URL, but does that remain indefinitely, or must the issue remain open? I can start a general wiki image thread in that case, and dump images there. Stuff like this:

cfg use 1

vab cfg

config containers

commented

I threw the images above in there of illustrate a provisional instruction set of Station Core on the wiki.

commented

req_parts

commented

@taterkerman if you are familiar with actual Git use, there is a repository that is specifically for the wiki. Unfortunately, it has no web-interface -- the only way to upload files to it is to use git-push.

https://github.com/shadowmage45/SSTULabs.wiki.git

If you clone that locally, you should be able to add commits and push changes. It should contain all of the text and image files, and you can edit everything locally using a text-editor (e.g. the wiki pages).

If that is something you would be interested in, please let me know. I can (most likely) help guide you through the process of getting Git setup, getting the repository cloned, and showing how to pull/merge and push changes.

commented

Cool, thanks.

I'll read up little on it, first, so you don't have to hand hold me if I can figure it out.

I installed git desktop, and cloned the wiki. I'll mess about at some point, and post questions if I have them. The real issue was if there was an image repository, or if they can be within that wiki repository itself (and presumably they will then just use local links since they'll be in the wiki's directory).

commented

Btw -- I added updated renders of the few engines that were missing them (LR-81, Lunar engines). Wiki has already been updated.

Will try and find time to do up some renders of the SRB nozzles a bit later this week (both lower and upper stage variants).