The quadcopter distilled down to its roots, adapted for FPV. The Krieger!
In the beginning there was the plus quad, and it was good. Then someone wanted to put a camera on it, so the plus became an X, and it too was good. But props were in the shot, so the X became an H, and that was even better. The H ruled FPV for many millennia, relegating the X to LOS acro. Then speeds went up and racing took over, and it was no longer necessary to put the camera up front for an unobstructed view, and DNA from an X was scraped from a fossil and they were resurrected to run amok again.
I’m not gonna lie, the Warpquad and the XBR got me thinking about making an X. And building the Tweaker made me realize that there’s very few parts in a pure FPV racer, and fuselages for 250’s were unnecessarily large. With those thoughts in mind I set myself some parameters.
Make it symmetrical in both geometry as well as mass so it’s easy to tune. Bring as much weight to the CG as possible to make it snappy. Make it 200mm (the smallest you can make a 5" quad is 180mm) in case it becomes a racing class. Make it as light as possible but strong enough for the high speeds it’s intended for. Make camera tilt easily adjustable without tools. Integrate mounting points for all components so there’ll be no nasty builds full of zip ties and double stick tape. Skinny arms are nice in theory but esc’s go there so make them 20mm wide to hold an RG20. And make it fit 5” props and up to 2208 motors ‘cause it’s going to be an overpowered beast.
My first thought was that the fuselage should be angled around 45 degrees to the motor plane for minimal drag. If the plates ran vertically they’d present less area to the wind, and they could be notched into the baseplate. The FC would mount vertically too. Vertical side plates would also make a good mounting point for a pivoting cam plate. I started to whip something up, and it looked like this.
Aside from its somewhat phallic appearance, I wasn’t happy with its size. The pod should be smaller than the Tweaker’s fuselage. I abandoned the tilt fuselage idea – it should be so short that it doesn’t need to be aligned with the angle of flight. I tried stacking the receiver/pdb/fc column on the body plate and worked on where to put the fpv system.
Oh my god this is just butt ugly, and way too tall. There’s also no place for a Mobius or GoPro.
Now this thing is supposed to be a pure race machine, but let’s face it, we all love sweet videos. So I worked backwards from an HD cam mount. I gave the top a nice long 60mm run at 35 degrees, and pushed the fpv cam mount forward. A mobius would ride on top of two standoffs, probably with a little foam in between. The whole thing got a lot shorter and much better looking. Hey we’re gettin’ somewhere!
And here it is fleshed out. 4mm arms, 2mm body plates, 1.6mm everywhere else. The vertical stanchions slot into the base and are held down t-nut style. The cam plate slides through slots and then rotates up anywhere from 20 to 50 degrees. It’s slightly wider than the standoffs so it gets pinched in place. The idea is that it’s a friction fit that’s easily adjustable at the field without tools – we’ll see how that works. The battery rides below, like an acro x quad. I showed the design to the FPV Addiction boys and of course Ryan Gury requests a 6” version. No problem, just make longer arms, such is the beauty of the symmetrical x quad.
And this is what the wind will see.
I got some prototypes made, and it's ADORABLE. It's so much smaller than I expected, barely bigger than a Tweaker, but able to take 5" props. Nice and light at 85 grams. Time to build it up and bash it around.
And here it is built up.
Up 'til recently I've developed and tested my frames in isolation. But since joining FPV Addiction I've had access to some amazing pilots to help with testing. The very first version of the Tweaker had 6 builds flying and that really accelerated the development process. I got the first prototypes of the Krieger on a Friday and Shane @onfourblades and I both had it flying Saturday. We met up at the club field on Sunday for its first real flights.
I was down to my last working fpv cam, and it had something loose inside which made the picture go completely white with turns and punchouts. I was able to get a feel for its smoothness but I couldn't do much with video constantly going out. Then Shane opened his up, and our collective jaws hit the floor.
It was easily the fastest quad we've ever seen on the track, like Quadmovr on FPV. And while some hexes might've matched its top speed, they certainly couldn't whip around and reaccelerate like it. Later we had a midair collision with the two protos that yielded this perfect shot.
And here's how it happened.
In hindsight Krieger's speed should've come as no surprise. I played around with cycling aerodynamics a bit, even got to attend Olympic gold medalist Kristin Armstrong's testing at the San Diego Low Speed Wind Tunnel. Aero drag increases as a square relative to speed, and the power to overcome that drag increases cubically to speed. The faster you go the bigger a role aerodynamics plays. I knew a simple X quad would have less drag, it was just shocking to see the magnitude of its effect in real life.
Shane took the prototype to NAFPV and raced it more, and it became very clear that it was much faster than any H quad and just as nimble. Any time there was a straight it would pull away with ease. The rest of the guys demanded Kriegers of their own.
I immediately started on V2. The basic concept was solid, it was just a matter of moving parts around and making them fit. I wanted to give the fpv cam a little more protection, so I moved it back slightly and adjusted component placement accordingly. I raised the angle of the back from 35 to 40 degrees, and put the top standoff on a slot instead of a hole to get more clearance if someone ran a full sized board cam. I also make a bigger hole in the top plate so a receiver could be sunken lower. The differences are subtle but they should make it easier to build and repair.
The two protos took an immense amount of abuse that day with no ill effects (mine crashed repeatedly due to lost video!), but I could see that the pod could be a weak point. I reduced the thickness of the X plates (2mm to 1.6), and thickened the vertical plates (1.6 to 2). Here's V2.
Our local group has a Slack where we brainstorm and waste time. The Tweaker was named in there when @shane_fpv made a joke about how we're addicted to FPV like meth heads and tweakers. Naturally the Krieger was presented to the group for naming.
Privately it was codenamed Dong after the first drawing. Most subsequent name ideas were X-based, with some really good ideas like Xen, Xentric (eccentric), Addix, Addixion. I particularly liked Xentric 'cause, let's face it, all RC guys are a little...off. I also thought it could be named Whiplash in honor of Shane's ludicrous speed video.
One day we were making some 'phrasing' jokes when @luckylegacy13 suggested an Archer based name, and then came up with Krieger. Mad scientist with questionable ethics? Sold! I also wanted to revisit my comic strip drawing days and create a sticker for Krieger. Another one of my clubmates Chuck had suggested the blue meth for Tweaker, and now I feel a lot of pressure to include something equally fun in the next kit. I hope the sticker fits the bill.
So there's V2 all built up with Cobra 2206 and Kiss 30a. My V1 was full of gremlins so I never got much of a chance to fly it so this was my first chance to really open it up. I was surprised to find that it needed higher P on pitch and roll, as well as higher rates for fast flips and rolls. I assumed it would need low gains and rates because of its small size and low rotational inertia.
Maybe I'm imagining this but it seemed to fly really 'symmetrically', with no preference for any given direction, truly nothing but a thrust vector in space. Or maybe I'm just too used to flying the Fast Forward, which is biased to forward flight. Once I got used to it it flew like a dream. With lots of cam tilt it just zips along effortlessly, frighteningly fast.
Three more V2's got delivered Wednesday to the wrecking crew in Long Island, two for 5" and one for 6". This is what I love about these fanatics: all three were built by Wednesday night!
Math Interlude (feel free to skip ahead)
Just for fun I thought I'd dust off my aero calculations from my cycling days. First, let's assume that a Cobra 2206 spinning a 5045 bullnose pro generates 860 grams of thrust. That's 3440 grams total. The Krieger weighs about 500 grams without an HD cam.
If you vector 500 grams vertically you end up with 3403 grams of forward thrust (500^2 + 3403^2 = 3440^2). That's a whopping 81.6 degrees of forward tilt without losing altitude!
Now let's assume there's an H quad with the same power train and weight. It will also fly level at 81.6 degrees. I modeled a quick 230mm H quad with a 40mm wide body and 35mm tall standoffs, and placed it next to the Krieger, both at 81.6 degrees.
And here they are viewed straight on as silhouettes.
I'm not sure if the prop disks count as circles for drag, so I did it with and without props.
Using Photoshop to count pixels, Krieger has either 8.5% or 41% less frontal area.
Now, the aero component of the power equation is:
watts = Cd (coefficient of drag) x a (frontal area) x air density x velocity cubed / 2
Since we're assuming the same powertrain and weight for both quads, we can put watts (Krieger) = watts (H quad). If we also assume they have the same Cd, then we can cancel out air density, Cd, and the 1/2 at the end, leaving us with:
a (Krieger) x v (Krieger) cubed = a (H quad) x v (H quad) cubed
If we call both a and v for Krieger 1 then we have
1 = 1.085 v (H quad) cubed
1 = 1.41 v (H quad) cubed
Which leads to the velocity of the H quad at 97% or 89% of Krieger. Visually it sure seems like 89% is more accurate, which would suggest you don't count prop disks when calculating drag. If you know please get in touch.
Hey there! The Maths is over!
As I type this Jay and Shane are plotting silly tricks with their Kriegers. Video soon as I update this page!
Goalazoooooooooooo! Jay intentionally scores a goal with his 6" Krieger.
We also discovered that the Krieger is perfect for shooting rear facing video: just turn the camera around and it's in perfect position. Way easier than an H quad!
I make prototypes on the light side and then test to see what breaks and strengthen as needed. Shane flew into a pvc gate full throttle and so now the arms will not have cutouts.
Henry lost video while flying mine and dropped it from way up. The battery cracked the bottom plate on impact, so now the lower plate will revert to 2mm. I also added angle indicators for cam tilt as suggested by our club president Rick. Jay's x plates cracked, apparently flexed by the arms, so I added a 4mm spacer to make the central unit almost solid. Now it's time for the factory to start cutting.
Here's some shots of the first run. Krieger's another model with many interlocking pieces, and the factory nailed it again. Tolerances are so tight that pieces fit together almost seamlessly.
I have a feeling that the X quad will become the standard for racing. I'm starting to see other X's hit the market already, but I'm loving my Krieger for its quick cam tilt changes, 5" or 6" compatibility, HD cam carrying capability, and FC crash protection. And of course there's this too.