900 mm scaleish plank (Darkstar) - RC Groups
Thread Tools
Aug 28, 2016, 08:04 PM
Registered User
red-panda's Avatar
Help!

900 mm scaleish plank (Darkstar)


For some years now I have been tinkering with the idea of a small scale model of the fairly unsuccessful Darkstar, making several tests with discarded wings from other projects, but until now the correct cg has been somewhat elusive. I finally got round to plugging various numbers in to the Bruder cg calc and realised that the critical factor was the lifting ability of the centrebody. A sheet of 1/8 balsa was soon sacrificed to carving a 1/24 scale wing only to a decent reflex section and adding a central nose probe and weight to get the calculated cg. Span was 880mm, root chord 75mm and tip 50mm, (dimensions derived from the details supplied with a 1/72 scale model) with a constant maximum thickness of 3mm. To my surprise it glided quite happily straight off the building board, with the cg 8.5 mm behind the le. at an auw of 18gm.
Flushed with success I built a centrebody to slide over the wing. This was scale in plan at 150mm wide and 190 mm long, but kept a similar t/c ratio giving a thicness of around 9mm. With a bit of lead in the nose this also flew at the theoretical cg which was about 5mm ahead of the projected straight le. The next stage was to make a second body which this time was a scale 44mm thick, but still maintained an airfoil section. This required a bit more juggling with nose weight but was ok with the cg at 2,5mm ahead of projected le. The final centre body was coser to scale which means streamlined but not particularly an airfoil, and the cg moved furthur back to around 1,5mm aft of the wing le. auw was up to 27 gm. but my excuse is that I hadn't been building light.
I'm now fairly happy that I've cracked the cg issue, but need advice on the wing. I'm concerned mainly about the maximum thickness I can use, the optimum airfoil, and if I should increase the chord, which while losing scale appearance would help me cheat Mr Reynolds. I'd rather not increase the size too much as I live in a city and all the nearby parks are small and but could probably afford to take the span up to araound 1m.
Ther are currently no fins except for a tiny sub-fin as a launch grip, but a good handful of dihedral. I am considering elevons or separate ailerons/ elevator, with permanently open drag rudders if yaw makes it so. I'll be using a Spectrum brick with either a 6mm or 8mm motor with a nose mounted prop unless someone invents an ultra lightweight and efficient EDF in the near future.
Any comments and suggestions are welcome, but please be warned I work irregular hours so sometimes make fairly slow progress, and we're about to enter the rainy season, which impinges on test flying.
Thanks for reading.
Chris
Sign up now
to remove ads between posts
Aug 29, 2016, 07:29 PM
Registered User
Hussle's Avatar
The wiki page on it says the project was finally abandoned due to insurmountable difficulties with instability. Good luck! It's certainly an extraordinary looking aircraft.
Aug 30, 2016, 09:16 AM
Registered User
EdSoars's Avatar
Instability issues with that narrow chord and big duckbill out front is no surprise! It would be a challenge to design, trim and fly. Maybe a gyro for pitch would help, but you would need very fast, precise servos.

ed
May 22, 2017, 03:01 AM
Registered User
tspeer's Avatar
Quote:
Originally Posted by Hussle
The wiki page on it says the project was finally abandoned due to insurmountable difficulties with instability. Good luck! It's certainly an extraordinary looking aircraft.
That's not really accurate. I was a flight controls engineer on DarkStar, and the issue was controllability on takeoff. The first prototype crashed on its second takeoff because of unstable dynamics on the ground when it was neither completely flying nor completely supported by the gear. The second prototype had nearly completed its flight-worthiness test phase when the program was cancelled. The program was cancelled because the technology demonstrator, which had been sized to a specified cost to manufacture, didn't carry a payload the Air Force considered to be militarily useful.

In flight, it had a narrow range of angles of attack for which it was statically stable. The pitching moment curve broke unstable outside of that range, due to lift on the fuselage, but the fly-by-wire control system could handle it. It also had what we called the "seagull mode" - a combined body pitch/wing bending flexible mode that was actively damped by the control system. A model will probably be stiff enough not to have an issue with the wing bending mode, but the stability characteristics will probably make it difficult to fly without some kind of augmentation.

It will look strange, but the way to take off with a model is to wheelbarrow down the runway with the nose gear on the ground and the main gear off the ground, then rotate for takeoff from there while not letting the main gear touch the ground. The drag rudders on the prototype were quite effective, so the handling qualities of a model while it wheelbarrows will probably be adequate.

The plane will be stable and controllable in pitch when in the one-point attitude with the nosewheel on the ground an the mains in the air. In the two-point attitude, with the maingear on the ground and the nosegear in the air, it is both unstable in pitch and uncontrollable with the elevons - not a happy combination! The reason for it being uncontrollable with the mains on the ground is because the effective center of effort of the elevons is directly over the main gear. It's like mounting a doorknob next to the hinge - you can push and pull on the knob all you want, but you won't open the door, because the force on the hinge fights the force on the knob.
May 22, 2017, 05:46 AM
Registered User
Hussle's Avatar
Wow, thank you for the detailed reply! It's great to have an insight into the design and testing process.

You've given a great explanation of the handling characteristics on takeoff. The RAAF Caribou stol aircraft used to wheelbarrow down the runway as a stunt at air shows

Might you be able to tell us why the planform was selected in the first place?

Thanks for the great info,
Hussle
May 22, 2017, 05:19 PM
less is more
Knoll53's Avatar

Test fly as a glider


Assuming the wing has a workable plank airfoil, fly it first without the pod. I'd put a gyro on the elevator, turn it up to full gain and bungee launch it into the heavens. You'll figure out how to fly it before it's time to land. Oh, and only fly if the park is empty. Once everything is all trimmed out, then put the pod on and test fly.

Don't forget the video.

Jul 12, 2017, 08:39 PM
Registered User
tspeer's Avatar
Quote:
Originally Posted by Hussle
...Might you be able to tell us why the planform was selected in the first place?...
Darkstar was a stealthy reconnaissance vehicle, and its shape was dictated by stealth considerations. It could carry either a camera or a radar. It was designed to view the target from the side instead of flying straight over it. So the lobes of the radar cross section were aligned in the direction of flight, with only small lobes at right angles to the flightpath from reflections off the wingtips and side of the fuselage. The absence of vertical surfaces was also to minimize the radar reflectivity.

The radar cross section of any stealthy aircraft consists of a few bright specular reflections, resulting in large lobes, and a much smaller "fuzz ball" that is the strength of the reflections in other directions. To be stealthy, you don't want to reflect radar strongly in all directions - it's better to concentrate all the returns in a few bright lobes and then arrange the flight path so the lobes aren't pointed at the bad guys. Radar reflections tend to come off at 90 degrees to an edge, and if all the edges are parallel, then they are all reflecting in the same direction. That creates a strong reflection in that direction, but all an enemy might see is a momentary flash as it passed by and very little reflection otherwise. You can tell how many lobes a stealthy plane will have by simply looking at the number of edge directions it has. The B-2, for example, is a 4-lobe design. The leading and trailing edges are parallel, as is the wingtip of the opposite wing. The lobes were oriented in an X configuration so they weren't pointed at the target when the jet was flying toward the target. Darkstar was essentially a 2-lobe design, with the lobes oriented fore-aft because it wasn't going to be flying right at the target.

Stealth doesn't make the aircraft invisible to radar. Reducing the strength of the radar reflections means the detection range of the radar is reduced, but if the aircraft is close enough, it can still be detected. If a radar network has overlapping coverage between adjacent radars, reducing the detection range means the radar coverage shrinks so it no longer overlaps and gaps are opened up between the radars. Darkstar's 2-lobe radar cross section meant the mission plan was like threading a 2x4 through the gaps in the radar coverage so it could slip in and out undetected.
Jul 12, 2017, 08:49 PM
Registered User
tspeer's Avatar
Quote:
Originally Posted by Knoll53
Assuming the wing has a workable plank airfoil, fly it first without the pod. I'd put a gyro on the elevator, turn it up to full gain and bungee launch it into the heavens. You'll figure out how to fly it before it's time to land. Oh, and only fly if the park is empty. Once everything is all trimmed out, then put the pod on and test fly....:
I'm not sure that's a good strategy. There is a substantial amount of lift on the fuselage, and it's ahead of the aerodynamic center of the wing. At moderate to high angles of attack vortices form along the edges of the fuselage, resulting in more lift and a more unstable pitching moment. Whatever you learn by flying it without the fuselage will be quite misleading when you put the fuselage on.

I would be inclined to keep the center of gravity well forward, even though this means a substantial loss of lift on the wing from the elevon deflection needed to trim. Then move the c.g. back as you get more experience.

Darkstar had a very narrow range of airspeeds. It practically took off, cruised, and landed at the same speed. It wasn't quite that bad, but that's a good qualitative description. In this regard, it will be more like putting radio control on a freeflight design, because freeflight models also fly at a constant airspeed.


Thread Tools