From an appeal on the Register today:
If you're a dab hand with X-Plane's Plane-Maker who just happens to fancy eternal internet glory as part of the Low Orbit Helium Assisted Navigator (LOHAN) mission, then today's your lucky* day.
We're inviting any bright spark who can turn our .SLDPRT CAD files into an X-Plane to get in touch.
The reward for this labour is the aforementioned glory, plus beer, quaffing of which TBA. Sadly, the bruised LOHAN budget doesn't allow greater recompense, and indeed our Playmonaut is complaining vociferously that we recently cancelled his hired stretch limo and bought him a second-hand bicycle.
Full story here.
(*Our particular interpretation of the meaning of the word "lucky")
Comments
Ok, Thanks for getting back to me.
Hi Phil - we've got a couple or people already working on an X-Plane model, but if you fancy giving it a go too, let me know. The more the merrier, as far as I'm concerned.
Jeez, I've just thought, I got in to X-plane in second year of uni. That's 13 years ago! Time flies.
Lester if you're serious I can help with this. I've been a user of x-plane since v5, so about a decade. I can also link you to my profile on linkedin for some more background / CV if required.
@Gary That's got to be plan A. We need to programme the thing to lock the controls on the ascent anyway, and then spring into life when "condition x" is met. I'd be inclined to say "launch event + time [seconds]" or something like that. The aircraft should glide no problem in a controls-locked state, or so the designers confidently inform me. How well it flies at altitude depends of course on how fast it's going, and it's going to have to exit the launch platform at quite a rate of knots. That velocity was one of the designers' big concerns, but they may have underestimated the force of the mighty rocket motor.
It's only going to get better as LOHAN goes down. Could you not lock off the controls until below say 50k it should be getting thick enough there and give you enough altitude for the glide. Would stop stacks of control surface hunting and possible strain on the system. Tridge might have some thoughts on that.
@ Andrew Rabbitt The high-altitude question will probably ultimately be resolved by a "suck-it-and-see" approach. The Vulture designers intend the rocket-powered phase to be a controls locked, straight-up affair. The principal concern is having enough velocity after that for the control surfaces to have any effect at all in the thin atmosphere. In the end, we can model it all we like, but even tests closer to the ground aren't going to provide any useful info about the stratosphere.
XFLR5 will do a better job of estimating high-altitude flight performance, since you can control Re effects with more precision. Whether it actually represents what's going on up there is anyone's guess. As far as the control aspects are concerned, you can learn a lot of stuff with X-plane HIL, but there are many traps for young players - you really need to understand how it is actually simulating it and drive the X-plane model accordingly. This is on top of the concerns Andrew T has regarding data latency.
Hi Andrew -
Yes, I fear the problems associated with making a workable and useful model may be too great given the time scale, especially for the high-altitude aspect of the mission. Hmmm...
Hi Lester,
A JSBSim model would actually be more useful. While you can do HIL (hardware in loop) simulation with X-Plane, the results won't be at all realistic of how the plane will fly, no matter how good you make the model. The problem is the latency of the HIL simulation, which makes the response times to control input far slower than a real aircraft. A rough rule of thumb is you will need PID gains about 3x smaller in HIL than with the real aircraft.
If you built a JSBSim model and it was accurate then you could run SITL, and that would indeed fly fairly accurately (well, at least not as inaccurately as HIL!). The problem is building an accurate JSBSim model, which is not easy. In fact, I doubt anyone will be able to build you a good enough model to cope with high altitudes well in time for your adventure.
Really the only realistic option is to fly the model with maximum logging (LOG_BITMASK=65535) and then use the logs to analyse how it flies (or plummets to the ground like a rock as the case may be).
If you do want to pursue the X-Plane option then what you'd want to do is fly it manually in X-Plane. That will be fun if nothing else, and may give you some hints on useful modifications.
Cheers, Tridge