Project Kliper/Soyuz-3

darrenc

New member
Joined
Nov 22, 2009
Messages
22
Reaction score
0
Points
0
Description:

I've posted a little about this project before, but since I am starting to devote more effort to it, I thought it would be worth a thread here to provide some more detailed info. about it.

The Kliper is a Russian conceptual spacecraft which, unfortunately, looks like it will never make if off the drawing board. The most recent info. about it seems to be from around 2006ish when the ESA decided not to fund some development studies.

This is where the real benefit of Orbiter comes in: It doesn't cost gazillions of dollars to develop a spacecraft :thumbup:

The Kliper is "Space Shuttle Junior". It is not designed to carry any significant payload - Just 6 cosmonauts and up to 700kg of equipment. Essentially, a space-ferry for humans.

There are many different design proposals around for Kliper, but I have decided to settle on the late 2005 winged version. I chose this model since it had some nice plan-view image which made it easier to model. Apparently, some aerodynamic studies were done in early 2006 which required a major redesign of the airframe due to excessive heating on the leading edges of the wings during re-entry, but I chose to overlook that.

The launch vehicle I have chosen is the (also vapour-craft) Soyuz-3. This is basically a Soyuz-2 on steroids. Once again, I have chosen this vehicle basically because it is easier to find information about it.

For the Launch Escape System, I have gone with the configuration which has the LES mounted on the Launch Vehicle Adaptor. This allows the LES to be used for the final orbital insertion burn, before the LVA is jettisoned, assuming an emergency-free launch. I really like this idea because:

1) It makes good use of otherwise dead weight.
2) It just looks so damned cool.

Status:

At the moment, I have only developed basic 3D models for the spacecraft and launch vehicle. The Kliper spacecraft and adaptor are probably 75% of the way there, with the launch vehicle at about 50%.

Some code has been written, but only enough for basic testing.

Direction:

The big goal for me is a good, realistic aerodynamic model. I am currently checking out various Computational Fluid Dynamics (CFD) packages for this purpose.

One area which interests me is the idea of modelling the drag and lift coefficients in a wide range of configurations using CFD, and then training a Neural Network to recognise the pattern and perform the actual prediction in the code.

Whilst I have a decent amount of Neural Network experience, I am extremely poor in CFD, so if there are some CFD gurus out there who can rescue me, I would very much appreciate it.

I don't want to let down a good spacecraft with a poor flight model.

Screenshots:

picture.php


picture.php


picture.php


picture.php
 
looks very nice!
if you need a beta tester....

are you also gonna make it AutoFCS compatible, or do we have to land it manually?
 
Cool add-on. Don't forget to use the Orbiter 2010 SDK to avoid compatibility problems :cheers:
 
looks very nice!
if you need a beta tester....

are you also gonna make it AutoFCS compatible, or do we have to land it manually?

Thanks. I'm sure I'll need some beta testers at some point, so thanks!

With regard to AutoFCS, I don't know much about it but I would definitely like to have some kind of landing autopilot. Probably the initial versions will be manual land only - I'd like to get the basics working right first before adding bells and whistles :)

---------- Post added at 09:25 PM ---------- Previous post was at 09:22 PM ----------

Looks great! The only thing I'm wondering about is if the wind shear/g-forces incurred at launch will tear the wings off of the Kliper.

It's funny you should mention that. One of the ideas I was playing with was segregating the mesh of the major parts of the service module to allow it to break up during re-entry. It's a bit of eye-candy fluff, but it sure would look neat!

As for the wings coming off during launch, I think the major concern during development was about them coming off during re-entry. Launch should be ok.

---------- Post added at 09:28 PM ---------- Previous post was at 09:25 PM ----------

I have a Nikolai Kuznetzov NK-33 engine mesh if you need it, also i can collaborate in beta testing :P

Ooh... if you'd be willing to share that mesh, I'd love to take a look at it. I was just going to fit my Soyuz with "standard" nozzles, but I may as well try and keep it real.

---------- Post added at 09:30 PM ---------- Previous post was at 09:28 PM ----------

Cool add-on. Don't forget to use the Orbiter 2010 SDK to avoid compatibility problems :cheers:

I'm linking against 2006 at the moment, but I'll link against 2010 once it is officially released.

---------- Post added at 09:35 PM ---------- Previous post was at 09:30 PM ----------

Model is nice, just to add some details (windows, etc).
Some information may be found at http://buran.ru/htm/cliper.htm (russian language, bit with lots of pictures).

cliper02.gif


P.S.: VC? UMMU?

Thanks, MC. That image is actually the plan view I have been using for constructing my model! Unfortunately, the scale bar in that image is the only reference I have found so far for the actual dimensions of the thing!

I've seen that web site before, but it only just occurred to me to try Google translate on it... Hmmm... "a one-time three-spacecraft Soyuz"? I guess that's a translation of "expendable three-stage Soyuz launch vehicle" :lol:

VC & UMMU... Probably later on, but not in initial releases.
 
As for the wings coming off during launch, I think the major concern during development was about them coming off during re-entry. Launch should be ok.

What about encasing it like the American's did for the Boeing X-37 launch? So it'd be an encased payload rather than an exposed payload. The X-37 and Kliper are somewhat similar machines when it comes to structure design. Looks like an interesting project, I'll certainly be keeping an eye on how your development of it goes. Good luck :).
 
Looks very good ! I love mini-shuttles ! :thumbup:
 
A little more progress today. Windows have been added to the Kliper (though I note it has no cockpit windows!), and I've begun developing a simple multistage framework in C++.

Having never developed a spacecraft before, I was a bit surprised at the hoops you need to jump through for multi staging. This was made even more daunting by the number of stages in the Kliper.

The embryonic form of the framework allows you to do a little setup in the beginning, by deriving from two classes (MultiStage - in place of the VESSEL2 interface, and a VesselStage for each stage) and then later on call VesselStage::Jettison. I am quite pleased, since it allows me to jettison whole stage groups (ie: jettisoning the second & third stage as one unit). This is handy for launch aborts:

picture.php


Best of all, it allows me to reuse huge chunks of code regardless of if the stage is operating by itself or as part of a launch system :)
 
Vertical at the cape :)

I hope NASA doesn't mind.

picture.php


I've been working at bit more on the model (mainly the Soyuz-3 launch vehicle) and the multi-stage code. I was disappointed to see orbiter has no API functions for rotating a mesh (really a must for the Soyuz boosters), so I wrote my own code which directly manipulates the MESHGROUP structure. I guess this is not a good idea for forwards-compatibility, though...

But, over all, the multi-stage code is turning out to be quite useful. The code is being developed with re-usability in mind, so hopefully it could make a useful developer add-on in it's own right.
 
Development has continued on the MultiStage framework. I have extended the VESSELSTAGE class to include wrappers for THRUSTER_HANDLE and PROPELLANT_HANDLE. Now, when you jett. stages, all thrusters and propellants of the stage are automagically transferred to the new vessel, with quantities and thrust levels intact. I plan on extending this technique to cover all vessel resources.

I now have two main classes in the framework. VESSELSTAGE - a non-VESSEL2 class which duplicates most of the VESSEL2 functionality, and MULTISTAGE - which is derived from VESSEL2 and acts as a go-between between VESSELSTAGE and Orbiter, performing such tasks as routing callbacks to the appropriate VESSELSTAGE instance.

The big convenience here is being able to modularise - each stage can run it's own, even if it's operating as part of launch stack or by itself. Also, all stages can reside in a single DLL, reducing the clutter a bit.

One problem I have been encountered is ShiftCG(). Namely, it doesn't do what it says on the box. I have overcome this by writing my own version of ShiftCG which does the job, although now the meshes obscure the vessel label when in external view. Weird. Has anyone else had any problems using the (woefully inadequate) ShiftCG() method?

Here's a test launch of Kliper built on the framework:

picture.php
 
Whoa, nice! Good to see this is going strong after first surfaced in the "Virtual Space Agency" thread. :thumbup:
How will the cockpit work? I'm imagining it being 2 or 3 giant MFDs, for cameraMFD and other stuff, some switches around them (or overhead), and 2 side views for the windows.
Although, the Orbiter default glass cockpit might be something like what you'd see on the real thing's monitor, if it just had one of them...

---------- Post added at 06:27 ---------- Previous post was at 06:23 ----------

Hey, this would fit on an H2B... :hmm:
Russian airframe+Japanese avionics=win
 
Whoa, nice! Good to see this is going strong after first surfaced in the "Virtual Space Agency" thread. :thumbup:
How will the cockpit work? I'm imagining it being 2 or 3 giant MFDs, for cameraMFD and other stuff, some switches around them (or overhead), and 2 side views for the windows.
Although, the Orbiter default glass cockpit might be something like what you'd see on the real thing's monitor, if it just had one of them...

---------- Post added at 06:27 ---------- Previous post was at 06:23 ----------

Hey, this would fit on an H2B... :hmm:
Russian airframe+Japanese avionics=win

I'll probably put out a cockpit-less version once I've got the spacecraft/lv all working fine, and cockpit added late on (which should be the fun part!).

There will definitely be an MFD for CameraMFD. I'm not sure if I should add some kind of cockpit window so you can see where you're going... Who knows, maybe the real Kliper would have ended up with one?

I'd really like to add some nice features such as circuit breakers which actually "work". It'd be fun to simulate some emergency scenarios such as a stuck RCS and having to isolate it through the breakers or something along those lines :)
 
It's been a while, so I suppose an update is in order. Unfortunately, I only get a chance to work on this on weekends, so progress is slow :(

I've posted a video to youtube showing 1st and 2nd stage jettison, and also a test of a neato thrust-vector control algorithm I have been working on. The test consists of spinning the vehicle up and then zero-ing the rotation rate, allowing the TVC to null it out. I think it looks quite cool :)

The algorithm doesn't yet allow me to specify gimbal limits for the engines, so they go quite outside the +/-4 degree range of the RD0146E, but it was a nice little proof of concept.

Also, the model isn't complete so the 3rd stage engines just sort of "hang there". :lol:
 
The staging of the boosters is wrong... :lol: ...ok, even the animators of the Russian Space Flight agency do that wrong.

But in reality, the separation happens at the aft end first. At the bottom of the page here is a drawing of the sequence:

http://www.russianspaceweb.com/r7.html

Also are you sure the third stage skirt has four parts? All Soyuz launchers in the past had this skirt in three parts and I didn't find any change this this yet.
 
The staging of the boosters is wrong... :lol: ...ok, even the animators of the Russian Space Flight agency do that wrong.

But in reality, the separation happens at the aft end first. At the bottom of the page here is a drawing of the sequence:

http://www.russianspaceweb.com/r7.html

Also are you sure the third stage skirt has four parts? All Soyuz launchers in the past had this skirt in three parts and I didn't find any change this this yet.

Thanks for this info. I am no Soyuz expert, and finding this info can be quite difficult at times. I just decided what I thought "looked right", but I'll fix this when I get a chance.

russianspaceweb.com has been a favourite of mine for this project, so I'm surprised I didn't pick up on this myself.

Thanks again!
 
Thanks for this info. I am no Soyuz expert, and finding this info can be quite difficult at times. I just decided what I thought "looked right", but I'll fix this when I get a chance.

Don't worry, I needed a lot of strong hammering by bigger Soyuz experts until I accepted that this is the correct sequence. :cheers:. Actually, the sequence was described like this in the Foton-M accident report, which I used as reference, but did not understand it correctly.
 
Hey, I'm very excited to see this happening. I've been looking for a Kliper for ages! Don't give up on this, OK?:thumbup:
 
Back
Top