• This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn more.

The CyberQuad

Cyberdactyl

Misfit Multirotor Monkey
#1
Ok, here’s a bit more detail. I’ll post this to force me to actually start cutting this weekend. Looks like thunderstorms most all weekend, so I’ll probably have the time inside. :rolleyes:

I’ve calculated the AUW, and it looks to be somewhere in the 540-560g range, depending on the battery and how well I manage wiring. The CF frame, without hardware is 116g. With the robust Rctimer BC2208-8 (22mm x 28mm) 2600KV motors I'm using, as well as using a KK2 and regular size VTx, I feel good at keeping it in the mid 500's. If this works well, I could easily step up to more expensive and exotic hardware and shave off 40-80g I suspect.

I’ve meticulously gone over it to keep the size pretty much as small as possible. Since I use a wet tile cutter, and Proxxon rotary tool with a cutting wheel and a grout removal bit, I'm trying to keep the cuts simple, but functional. I’m restricted to four basic design parameters.

- 6” props.

- As some may have guessed, I want absolutely the thinnest booms in planform as possible. Mostly to show it can be done successfully. I have it down to ~3.1mm for the entire radius of the thrust column, except for the motor area. If you pause the video at about the 6 second mark, you can see a representation of the method on the starboard front boom I’m using for wiring down the booms. It’s 5mm x 0.75mm copper strip, similar to these with shrink shrunk tight and adhered to the boom. I’ve tested it and it works well. I’ll use bullets under the motors for break-away safety.

- I want as much mass as reasonable floating on silicone bearings with the camera.

- I want the entire volume to be as small as possible, working around inexpensive and easy to use components.
I’m using the KK2.1.5, an Orange 6ch Rx, a 600mw Vtx, and 20A Afro Slim ESCs, and a 1300mAh 3S or 4S battery.

You can get an idea of the size of the quad by the ESCs in the back. Everything is to scale, the components are within a couple millimeters, the frame is within a half millimeter. If you still can't visualize, the center structure is about the size of a baseball. I’m using 2mm CF plate entirely, but …may... use 2.6mm G10 for the motor support pieces, since those pieces are pretty small and 2.6mm offers a bit more width for the motor support cross piece. The split-piece inset method worked well for the CyberHex, so I’ll use it again.

One thing I did violate vs my CyberHex is not having the Inversa chip at the exact center of mass and lift. It's forward about 18mm in planform and ~25mm low in elevation. More than I'd like for something this small. However I do have the overall CG within a few mm of the quad's lift plane in both axis. I'll see how it goes, but this isn't for acro, so I'm not overly concerned.

This also uses the hard-clamping technique and only six compression columns (screws) and slot-n-hole for the booms. I've always been shocked at the liberal use of hardware on BlackOut type quads, we'll see if mine are enough. The rear upright supports also serve as an ESC rack, and I have them vertical right outside the thrust column for cooling. What is not seen in the video is, of course, the wiring and the 5V UBEC that will go in the center between the clamping plates. What will be a challenge is to keep the wiring clean and managed, but not so tight to transfer vibration to the isolated structure when the wiring makes the physical connection across the rigid boom and clamping plates to the floating cage.

Also, I’ll be using single strand nylon ties pulling the vertical supports tight to stiffen the structure underneath the FCB plate and underneath the top battery float plate (the holes underneath). Since I'm using 2mm I want some failure point of my choosing. :eek:

As some may have noticed, currently this does not have DVR. However a Mobius could be slipped in to replace the Sony PZ0420 later..

A lot of cutting ahead. . . :eek:

[video=vimeo;100373044]https://vimeo.com/100373044[/video]

https://vimeo.com/100373044
 
Last edited:

Balu

Lurker
Staff member
Admin
Moderator
#9
I'm always irritated when I see something with the battery on top. It feels unstable to me, like it is going to tip over all the time.
 

Cyberdactyl

Misfit Multirotor Monkey
#11
I'm always irritated when I see something with the battery on top. It feels unstable to me, like it is going to tip over all the time.
Actually, with the battery above the props, the z-axis CG is still below lift plane. In most quads, the batteries loosely approximate the total mass of the motors, and their z-axis CG is always under the lift plane, so the additive mass of everything else, still keeps the z-axis CG below the lift plane. And ideally, and especially in acro, you want the z-axis CG as close to the lift plane as possible.

That looks really interesting. What ESCs do you have spec'ed? I don't think I'v ever seen anything quite that narrow.
I have four. I scaled them directly from 20A(30A) Afro Slims. I plan on replacing the 20g wire with 16g and keep fast moving air on them with windows cut in the clear shrink to optimize their 30A capability.
 

FinalGlideAus

terrorizing squirrels
#12
Looking good cyber. Question: why wouldn't you put the camera above the props? For FFF they'll be right in view all the time especially if you tilt it back for better vision. Or am I over thinking things and you will be mainly hovering with it?
 

Cyberdactyl

Misfit Multirotor Monkey
#13
You're right, there will be prop in my view with the card-cam. But I wanted to protect it from damage in a crash and set it down between the booms and double purpose the front uprights for mounting and also for protection. But my design has the props spaced close in the X and the Y, much like a regular tightly spaced X config., so keeping the props out of view, above or below would have been a real challenge.

I either would have to raise the cam very high above the prop plane, or very low. . .or extend the cam out away from the frame. And as you said, I'm going to want the cam angled up for fast forward flight. And to keep the props out of view for a 2.5mm (~100 deg) lens would mean having the card rather high on the frame to not be seen in the bottom of the image. If I mount a Mobius later, it will stick out another ~30mm, and the lens angle is narrower, so seeing the props won't be a problem.

Most Blackout-type H config has a rather wide spacing between the starboard and port props. With 6" props I wanted to have the same flying clearance (or better) than a 5" H-quad.

I'm not sure if anyone noticed, but I did allow for the card-cam to be swept up, or down 20 degrees. There's no servo actuated movement, but at least it allows me to set it up considerably for fast forward flight.

 

crlock

Senior Member
#14
that is why i love this hobby :) outstanding work cyber, as always.
would you be kind enough to share the dimensions of the sony camera? i'm going to use that specific camera for fpv :) but it is still on the boat somewhere on the sea.

keep up the good work
cheers!
 

Cyberdactyl

Misfit Multirotor Monkey
#15
It's right at 38mm x 38mm. The depth varies with the lens, but the most minimal lens makes the card depth around 28mm, that includes a few mm for the sockets and SMDs on the back.

However, the PZ0420 has an outer "strip" that can be removed to reduce the overall width and height by 8mm or so. Here's Bruce Simpson explaining it.
 

Cyberdactyl

Misfit Multirotor Monkey
#17
Had a slight setback. Got home tonight and found the silicone bearings in the mailbox. The are about 50% larger than I thought. I'll need to to tweak the model a bit and widen their mounting area by around 4mm.

The little bit of good news is they raise the gap between the top floating plate and the upper 'dirty' clamping plate by about 3/16" giving it more clearance. I'll have a battery strap and nylon ties in that gap, allowing them a tad more breathing room.

 

stay-fun

Helicopter addict
#18
I'm always irritated when I see something with the battery on top. It feels unstable to me, like it is going to tip over all the time.
It blew my mind recently, but actually, a quad is more stable with it's CG above the props. I learned about it on helifreak, it's about a pendulum effect: if the quad, with CG above the props, wants to lean to the right, the thrust of the props will move the rest of the quad back under the CG!
 

Cyberdactyl

Misfit Multirotor Monkey
#19
Yes, I've ran into this a few years ago while researching to do a hyper-realistic 'hover-craft' in vacuum for a fanciful turret from a game I played in the late 90's. I discovered it was actually easier to maintain balance with a mass considerably higher ..IF.. the mass is going to be above the lift point.. But having the z-axis CG too far "above" the lift point has its negative aspects as well. :rolleyes:
 

stay-fun

Helicopter addict
#20
But having the z-axis CG too far "above" the lift point has its negative aspects as well. :rolleyes:
Absolutely, gotta find that sweet spot! Hm, how would one do that? After all, the gyros are taking out all instability... Try to fly a quad without gyros, and see when it is most stable? :rolleyes:

Actually, the pendulum effect using a broom on a quad (or your hand) has been demonstrated: https://www.youtube.com/watch?v=w2itwFJCgFQ