Question How to model complex engine transition dynamics?

[Wishbone]

Yes, you guessed it right. This thread is about NTR (nuclear thermal rocket) engines (meshing under way...) that ARE required for a manned flight to Mars.

Things that complicate realistic modelling:

1. Startup from minimum power (no propellant flow) to full thrust takes time (structural limitations on rate of heating).
2. Shutdown and residual heat dissipation take time and propellant.
3. Upon shutdown or abrupt power changes, a complicated rollercoaster of Xenon and Iodine poisoning commences, which may or may not influence max thrust achievable after a given time period.
4. Don't even ASK me about failures or full-scale modelling of neutronics+thermohydraulics.

A candidate approach:
quickly draw a few ballpark-derived nominal curves for thrust against time (startup, nominal power, shutdown), make splines of them and simply code the dependencies, without letting the user decide on the thrust level.

Two questions:
A) how to forbid the user to manipulate thrust but still retain the capability to command shutdown?
B) what information should be provided on-screen to simplify planning of ejection/capture burns?

 

[Moach]

Yes, you guessed it right. This thread is about NTR (nuclear thermal rocket) engines (meshing under way...) that ARE required for a manned flight to Mars.

Things that complicate realistic modelling:

1. Startup from minimum power (no propellant flow) to full thrust takes time (structural limitations on rate of heating).
2. Shutdown and residual heat dissipation take time and propellant.
3. Upon shutdown or abrupt power changes, a complicated rollercoaster of Xenon and Iodine poisoning commences, which may or may not influence max thrust achievable after a given time period.
4. Don't even ASK me about failures or full-scale modelling of neutronics+thermohydraulics.

A candidate approach:
quickly draw a few ballpark-derived nominal curves for thrust against time (startup, nominal power, shutdown), make splines of them and simply code the dependencies, without letting the user decide on the thrust level.

Two questions:
A) how to forbid the user to manipulate thrust but still retain the capability to command shutdown?
B) what information should be provided on-screen to simplify planning of ejection/capture burns?

a) i'd say (assuming you have your own DLL for this) set up a dummy thruster to gather input from the throttle axis, then "manually" assign the thruster level to the actual thruster...

or , if you don't need/want throttle, just setup your thruster and don't place it in any thrusterGroup... just set it on/off when the command is given


b) i'd think it would be good to show the current thrust output, propellant flow and ramaining deltaV/burn-time


now, without a custom DLL this is probably unlikely to happen... unless you can do LUA - that's a good new alternative (you still have to script, but LUA is way easier than C++)


do i serve you of any help?

 

[Wishbone]

User-defined thrustgroup sounds OK to me. Gotta try that, and the dummy main thruster that will be checked in preStep callback.

Thx!

---------- Post added 05-04-11 at 04:26 AM ---------- Previous post was 05-03-11 at 07:28 PM ----------

A quick update: I've got curves for thermal power and hydrogen flow from Koroteev's book. These are from real firings, although not exactly NERVA...

 

[Tommy]

One thing to consider is that people will need to be able to navigate this vessel. Some people will want to use IMFD, and it's AutoBurn.

I can't tell you how IMFD handles this sort of thing, you may consider asking jarmonik. Seems to me that IMFD works with some other multi-engine-type craft, so it should be possible to get your vessel to work with autopilots like IMFD or Artlavs UAP (when/if it becomes capable of interplanetary flights).

Then again, I don't know enough about NTR's to know if they would ever work with an AP that assumes "instant" dV - IMFD doesn't work very well with low thrust engines that are designed for prolonged burns (such as ionic propulsion).

 

[Wishbone]

Thanks for the ideas. Yes, Autoburn is nice, but thankfully IMFD does not rely on AB only (players do, though ). This capability should be augmented IMHO by the vessel-specific means to dial in required dV (including trailoff) and TtB.

Am not sure, however, IMFD or UAP can deal with the (fuel-saving) two-burn ejection scheme.

Another consideration: the current notional Mars mission design needs 3 engines, to handle 1EO gracefully. I'm not saying failure modelling is on the table right now, but eventually we can come to this. So, the vessel code would have to calculate time to shutdown under nominal and 1EO conditions.

Thinking about abort modes during TMI: since there would also be a conventional hypergolic-based MCC engine, it makes sense to install it as a retro-thruster, to ensure that the second ejection burn, if interrupted by a late engine failure (3EO), does not throw the craft into the vast empty interplanetary space.

 

[Wishbone]

Update: Meshed the 3-engine propulsion core stage from DRA 5.0, implemented a constant period ramp-up and shutdown. Gotta clean up and possibly implement fuel-depletion PMI and CoG shift.

EDIT: uploading to OHM...

EDIT#2: :facepalm: forgot to add a "thank you Moach :tiphat: " to the credits and docs. Sorry folks, gotta be a while till I re-upload the correction, currently on dial-up.