Orbiter-Forum  

Go Back   Orbiter-Forum > Orbiter Addons > Addon Development
Register Blogs Orbinauts List Social Groups FAQ Projects Mark Forums Read

Addon Development Developers post news, updates, & discussions here about your projects in development.

Reply
 
Thread Tools
Old 07-20-2017, 03:53 PM   #871
fred18
Addon Developer

Default

The multistage module has a built in attitude system, which is just made by invisible RCS thrusters. The default value for attitude thrust is a % of main thrust (can't remember how much now) in the idea that it's the thrust vectoring orienting the rocket.

I don't completely understand what you mean by using ullage as RCS.

I think that we have to divide the package in 2 parts:

- physical part
- graphical/eye candy part

for the physical part the difference of implementing an invisible RCS system instead of thrust vectoring is minimum in terms of orbiter simulation. Thrust Vectoring generates also a drift and a small amount of thrust reduction which is not there with invisible RCS, but to give you an idea about thrust reduction IIRC for the SLS a maximum gimbal angle for SSMEs of 6 was considered, and the cosine of 6 is 0.9945, which means a thrust reduction of 0.55% when the engine is fully turned, so basically always 0.
Drift could be more significant, but thrust vectoring usually happens in the first part of flight, where atmosphere still plays a role and with drag it basically nulls the drift. So in physical terms, at least as far as i'm concerned, in the orbiter simulation an invisible RCS or a thrust vectoring system produce basically the same effect. That's why MS2015 uses an invisible RCS system for all rockets.

for the graphical part that's totally another story, since it could be very nice to see exhausts and engine bells turning but I noticed that in order to have a clearly visible effect the gimbaling angle should have been multiplied because variations are usually minimum (are you usually able to notice engine gimbaling when you watch a video of a rocket launch?) and doing this was meaning complicate the things a lot in order to have a not realistic effect, then i didn't see the point.

So IMHO if you are looking to best simulate thrust vectoring physics you can try to code another rocket module, but I'm pretty sure that you'll find out that attitude control will have an almost identical behaviour than an invisible RCS system.

going to another topic: using ullage to simulate prefire. I don't think it will work: until this stage is detached, all the engines of the next stage don't exhist yet. I think the best way to do it is to make a ghost-booster (word game is unintentional..): you set up a booster with the exhaust specifications identical to the second stage engine, you assign an empty mesh to it, you put it in the same location of second stage engine and you calibrate its burn_delay in order to match your needings. Curiously if you set also the thrust and calculate fuel properly, this solution will produce a result which is both physically and graphically correct.

the Eng_dir param is for fixed rotation of engines. That was built for the Space Shuttle or similar vehicles (my fault not to have included an example of it, it would have been very interesting to show) which have engine thrust not towards their Z axis. I have to say that the launch of the Shuttle was quite nice with that effect, I'll have to add its example to the package.

Hope this helps!
Cheers
Fred

Last edited by fred18; 07-20-2017 at 03:56 PM.
fred18 is offline   Reply With Quote
Thanked by:
Old 07-21-2017, 09:53 AM   #872
4throck
Enthusiast !
 
4throck's Avatar
Default

Thanks for the detailed info.

Reading your description of the Eng_dir param you already have implemented fixed trust vectoring!
I agree that there's little difference to invisible RCS for all practical purposes. (On some launches you can see the exhaust angle change a bit for corrections, but it a small amount.)

The location of those invisible RCS may make a difference.
Do you place them around the main engine location, or at the rocket stage center ?
4throck is offline   Reply With Quote
Old 07-21-2017, 09:58 AM   #873
fred18
Addon Developer

Default

Quote:
Originally Posted by 4throck View Post
 Thanks for the detailed info.

Reading your description of the Eng_dir param you already have implemented fixed trust vectoring!
I agree that there's little difference to invisible RCS for all practical purposes. (On some launches you can see the exhaust angle change a bit for corrections, but it a small amount.)

The location of those invisible RCS may make a difference.
Do you place them around the main engine location, or at the rocket stage center ?
1 meter from the rocket center of mass, that's why they don't produce the drift, basically you can scale the distance by setting the pitch/roll/yawthrust. It's the very same of what Vinka did with Spacecraft
fred18 is offline   Reply With Quote
Old 07-21-2017, 10:22 AM   #874
DaveS
Addon Developer
 
DaveS's Avatar


Default

Quote:
Originally Posted by fred18 View Post
 1 meter from the rocket center of mass, that's why they don't produce the drift, basically you can scale the distance by setting the pitch/roll/yawthrust. It's the very same of what Vinka did with Spacecraft
Well, that's a bit of cheat actually. All thrusters produce a bit of waste thrust unless they're specifically located at ideal points. I know of no spacecraft or rocket stage that ever had it that way.

Here's two charts that show the RCS for the shuttle and as you can tell by the Thrust Components sections (X is FWD/aft, Y is left/right and Z is up/down), most jets most definitely waste thrust.

https://www.dropbox.com/s/rklsqpjltv...tions.jpg?dl=0
https://www.dropbox.com/s/60rqwh1lgj...ions2.jpg?dl=0
DaveS is offline   Reply With Quote
Old 07-21-2017, 10:28 AM   #875
fred18
Addon Developer

Default

Quote:
Originally Posted by DaveS View Post
 Well, that's a bit of cheat actually.
I'd say yes, but it's impossible to avoid it for a general module such as Multistage. It needs to grant proper attitude control to any possible rocket the user may design, so it basically needs to turn the rocket properly when used. The invisible RCS doesn't even use fuel. It applies the very same method Vinka used for multistage.

Of course for detailed realism the best option is always to code a specific vessel in every detail, but as you know Multistage is a general module, and for this purpose I think it does its job quite ok.

thanks for the charts!
fred18 is offline   Reply With Quote
Thanked by:
Old 07-22-2017, 08:46 PM   #876
4throck
Enthusiast !
 
4throck's Avatar
Default

I agree that for a generic module must be simple.

My questions were all related to getting realistic performance from the autopilot.
I'm sure that a way of setting the placement (the center) of the invisible RCS would add to realism.
But I do agree that it must be simple and work as a general solution.


All this is positive criticism, Multistage is a great achievement and I'm a fan!

Last edited by 4throck; 07-23-2017 at 01:55 PM.
4throck is offline   Reply With Quote
Old 07-23-2017, 03:39 AM   #877
boogabooga
Bug Crusher
 
boogabooga's Avatar
Default

Quote:
Originally Posted by fred18 View Post
 Hi there,

@boogabooga and the others: would you please check in orbiter2010 if the camera now works properly with this modules? thanks!

Fred
I only tested the default scenario with SLS. It seemed that the camera works better in Orbiter 2010 now. At least for the module you provided there in the thread. I haven't tried what is "official" on O-H now.

But, I still strongly disagreed with how the "G force" is being calculated. I strongly recommend that you implement the algorithm that both asbjos and Face provided. Or, use the "thrust/drag only" algorithm that you had going for the Orbiter 2016 version.
boogabooga is offline   Reply With Quote
Old 07-23-2017, 12:00 PM   #878
fred18
Addon Developer

Default

Quote:
Originally Posted by boogabooga View Post
 But, I still strongly disagreed with how the "G force" is being calculated. I strongly recommend that you implement the algorithm that both asbjos and Face provided. Or, use the "thrust/drag only" algorithm that you had going for the Orbiter 2016 version.
It should be simply thrust / mass. I hope that in the OH is calculated in this way for both o2010 and o2016, if it's not it means i forgot to update one of the versions, i'll check
fred18 is offline   Reply With Quote
Reply

  Orbiter-Forum > Orbiter Addons > Addon Development


Thread Tools

Posting Rules
BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
Forum Jump


All times are GMT. The time now is 06:40 AM.

Quick Links Need Help?


About Us | Rules & Guidelines | TOS Policy | Privacy Policy

Orbiter-Forum is hosted at Orbithangar.com
Powered by vBulletin® Version 3.8.6
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright 2007 - 2017, Orbiter-Forum.com. All rights reserved.