Flock of KestreLs?

[Chad]

John,
I took 2.25″ off the upper tube, and 0.6″ off the lower tube. Made some minor internal changes too. Fins are reshaped and fully airfoiled, but the span is not that much less than stock. Maybe 0.2″ shorter. My final CP/CG is about 1.7 calibers- pretty conservative. I didn’t want wobbles.

[Bdale GarbeeParticipant]

Just noticed this thread. I’m working on a KestreL too, though very much not stock. In fact, my “Sharp Stick in the Sky” flown at MHM used some of the parts, as I sort of kit-bashed a Wildman Jr and a KestreL plus a few other random parts like the conical nose cone that’s been on my shelf a while… my KestreL project is “K Up Yonder” and will undoubtedly appear at NCR North no later than O’fest… my big personal goal for 2010 is to “need” the high altitude window at least once, and this is the project that will get me there.

I’ve commented on other threads about the known/obvious limitations in Rocksim. Some are in the docs that came with the program (yeah, I know, I didn’t read them either until I started getting weird results!), and some I’ve learned by talking to Tim. One of the “rudest” for high dynamic flights is that the nose cone drag model is really simple and based on surface area, so a 1:1 conical nose cone will simulate as the optimal choice for altitude.

There are at least two ways to cope with this. One is to use a tool like RASaero to generate Cd values for your airframe for different velocities and plug those in to Rocksim to see how the results are affected. Since the Cd changes radically as you cross mach, using a single Cd value is at best an approximation of what’s going on, but you can at least get more believable results for different nose cone geometries this way.

The other approach is to just change simulators. I’m personally using OpenRocket for all of my new designs now. It’s an open source rocket design and simulation tool written by Sampo Niskanen at Helsinki University of Technology. One of the pieces of documentation provided is his masters thesis, in which he describes the approach he took to handling drag at higher velocities. I’m not a real expert on these things, but his approach made sense when I read the text and then the code.

I don’t have enough actual flying experience with things simulated using OpenRocket yet to know how great the models actually are, but the results seem much more “believable” to me. Since it’s open source, if they turn out to need tuning, we can do that. And I really like the ability to plot lots of attributes through the flight and then pan and zoom on the plots to see what’s going on. If you’re serious about simulating things like the KestreL, I’d suggest giving it a try. As of version 1.1.0, it even has the ability to import a Rocksim file… though to date I’ve done all my simulations with it by starting from scratch and so don’t know how well the import really works.

In any case, have fun! That’s what this is all about… 😉

Bdale

[AdrianParticipant]

I have found that the import works well, but the estimated drag is really, really high compared to both RASAero and actual flights. IIRC, RASAero predicted my J flight going over 22k, which it did, but Open Rocket was predicting something like 17k. So don’t use it to figure out if you’re going to bust the waiver if you’re close. On another rocket, I had to make the body tube almost zero length in Open Rocket to match the Cd that RASAero was coming up with. Hopefully this will get straightened out.

I love the rocket design interface though, and the way the predicted altitude gets adjusted real-time as you move the sliders for the different part dimensions.

[Bdale GarbeeParticipant]

Adrian,

Did you play with the surface finish setting? I noticed that OpenRocket picks something rough as default. You can change that in the general screen of any of the exterior components, ticking a box to change it for all exterior parts. That seems to be the largest “knob” on drag that needs to be set to something non-default to get rational results in OpenRocket that I’ve run in to so far.

Bdale

[AdrianParticipant]

Yes, I’m pretty sure that was with the smoothest possible settings.

[Bdale GarbeeParticipant]

Very interesting. Makes me wonder just how high my KestreL hack might get, then… hrm.

[AdrianParticipant]

I’d check it in RASAero and see how that compares to what you got with OpenRocket.

[John A. WilkeParticipant]

I’d like to keep track (at least in my head) of who breaks 20K for their first time – and did so with a KestreL. I think I have heard from 6 or 7 fliers who had their first 20K+ flight and they did it on a KestreL…

In our club, Chad comes to mind? There were other KestreLs that got up there, but I think they were from guys who had broken 20K previously?

It took me something like 3 or 4 tries to recover from over 20K. While it is getting easier, it is still an achievment!

[Chris LaPanse]

I’d like to keep track (at least in my head) of who breaks 20K for their first time – and did so with a KestreL. I think I have heard from 6 or 7 fliers who had their first 20K+ flight and they did it on a KestreL…

In our club, Chad comes to mind? There were other KestreLs that got up there, but I think they were from guys who had broken 20K previously?

It took me something like 3 or 4 tries to recover from over 20K. While it is getting easier, it is still an achievment!

Well, it won’t be on a KestreL, but I’m taking my first legit shot at 20k at the July launch, using my L3 rocket. I’ve got an M1939W, which sims to ~19.5k, and so far every supersonic flight has exceeded the sim.

[Author]

Posts