Snarls' 500mm Spider Quad Build

[Snarls]

Hello everybody. A few months ago I built a large quadcopter (680mm) capable of carrying a large camera gimbal as a project for my last year in highschool. The project was a great success and I ended up with a very stable, gentle flying quadcopter. At the same time I learned what was good and bad with the design in terms of how I wanted to fly. Even after downgrading to a smaller GoPro gimbal some of the problems I faced were: trouble mounting hardware on a small body, too large of a size even when folded, and exposed electronics. This first quad served me well, but now it's time to move on.

After reading a post by ananas1301 about his CineFold quadcopter I was deeply inspired to build a very similar design with a focus on aerial photography and portability. With my budget and resources, instead of carbon fiber and aluminum I will be using 1/8" plywood, 1/2" square wood dowels, and plexiglass.

After measuring my electronics I determined that a 500 size quad was the best size. I will be recycling the same electronics from my first quad for this build with the exception of the FC and new FPV gear. I am swapping out my Multiwii Pro with GPS to simplify the build and focus on the flying.

Electronics:
-4 X DT750 motor
-4 X 1147 props
-Q-Brain 4 X 25A ESC
-New MW Flip 1.5 FC
-Turnigy 9X receiver
-Homemade gimbal
-2200 mAH battery
-New Skyzone Plug and Play 200 FPV Set
-New Quanum FPV goggles

Based on the size of my electronics and ananas1301's design, I planned the frame in Sketchup. For a little fun I brought my plans into a rendering program to get some realistic images:

Folded into a small profile


Here is the gimbal and GoPro I will be attaching later on:


I printed out the plans for use as templates for cutting the plywood. Even with a new scroll saw I got for Christmas cutting out every hole in the frame took a good amount of time. I am a little more inaccurate and create more chipping on the scroll saw than I do with a hand coping saw, but I attribute this to my newbie scroll sawing experience.

 

[Snarls]

After going away on vacation I came back to the build ready to dive in. I cut the arms to length from some 1/2" square dowels from Home Depot (although I prefer the ones from Lowes which are poplar and straighter). Without the aid of a drill press I somehow managed to drill fairly straight holes right to specifications.

The frame experienced a lot of chipping, especially near the holes and cutouts so I decided to reinforce the holes and bottom of the frame with some CA and handy capillary-action.

To cut out the slots in the landing gear/skids I used a jig and a router bit on my Dremel. The plexiglass is relatively thin and unfortunately cutting the slots resulted in some small cracks. I'm afraid a hard landing may cause breaks, but hopefully the slide-ability of the landing gear will negate that.



I then proceeded to spend two days trying to paint the frame black and the arms white. I quickly discovered that it is incredibly hard and annoying to spray paint in windy, below freezing weather. I managed two light coats before giving up. In the future painting with craft paints from a bottle is probably the best option.

Assembling the frame was easy and amazingly only one of the holes was off. Despite all the cutouts, the frame with the arms and nylon spacers installed feels very rigid. All together the newly assembled frame comes in at only 170 grams compared to the old frame which was 270 grams. That's 100 grams or 37% less mass!

 

[Snarls]

Today I finished what I call the "electronic sandwich." The electronics fit so snugly in the frame, especially the Q Brain because its height is exactly the spacing between frame pieces. The flip 1.5 FC had to be shifted upward to account for the pins, but luckily this placed the board precisely where the motor to motor lines intersect. Due to the height of the receiver and the cables coming out of it I had to mount the receiver upside down, not that it makes a difference.

My only concerns with the wiring is that the USB to the Flip is unreachable, and that the power wires to the Q Brain cross right over the FC. Is telemetry possible with the Flip? If so I can solder some pins and hook it up for easier access to MW config. As for the power cables, hopefully that won't lead to any EM interference issues.

Everything on this quad just somehow fit together perfectly. As I already said, the Q brain is exactly the right height, and the FC is in the perfect location. The wiring is everywhere, but it is organized. The wires to the ESCs in the back reach with no slack while the ones in the front have a lot of excess length than is packed into the frame. It also looks like I'll be able to mount my gimbal up front without hindering the frames fold-ability. After all this I found that without a battery, the quad has a CG right in the center of the FC. For now the battery will be mounted on top around the center, but with the addition of a gimbal in the front I will be able to move the battery to the back as originally intended.

Final Specs:
-Motor to Motor Distance: ~500mm
-Collapsed Length x Width: 42cm X 18.5cm (w/ Props) 34cm X 18.5cm (no Props)
-Easily fits in a backpack
-AUW No Batt: ~800grams (very roughly because my scale only goes to 500 grams)
-Hover Time: 15 mins
-Non-Aggressive Flight Time: ~10+ mins


In short, the build went incredibly smooth. Let's hope getting this quad tuned and ready for AP and film will go just as smoothly!

 

[cranialrectosis]

Love the Sketchup views. Will the copter balance with the lipo all the way back?

 

[Snarls]

Will the copter balance with the lipo all the way back?

With the weight of a gimbal and GoPro up front I am hoping it will, but I'm not certain. Without the gimbal and camera I will surely have to move the battery closer to the middle. Luckily all the cutouts in the frame will make good points to run velcro straps through.

 

[C0d3M0nk3y]

Will the copter balance with the lipo all the way back?

That was my first thought too.

 

[Snarls]

Just updated the second post after taking a week off.

 

[cranialrectosis]

Yeah, painting takes time and is a PITA but it looks soooooo good.

 

[Cyberdactyl]

Looks good.

I'm curious about how much paint was actually sticking to the plywood spraying below 32F?

Rattle-can paint particulates are a good 15-25F colder than ambient merely coming out of the can, spaying below freezing, I'm somewhat surprised if it didn't act like dry powder.

 

[Snarls]

painting takes time and is a PITA

I agree. I do tend to look forward to it because it's the time where the build comes to life.

I'm curious about how much paint was actually sticking to the plywood spraying below 32F?

Luckily I made the wise choice of bringing the paint inside to warm up before rushing outside, painting, and bringing everything into my basement to dry under fans. Even so, every spray would turn into a plume of mist and proceed to be whisked away in the wind. I would say only about 10% actually stuck to the pieces.

 

[Ace2317]

Have you tried tightening all the bolts minus the extra four that you're taking in and out to fold the booms? On my Bat Bone, I just overtightened the bolts by about a half turn and I can fold my booms in on my Bat Bone without having to add screws. I've never had any problems with the booms moving. However, with a wood frame it may work differently. I would be interested to see how it works.

 

[Snarls]

Have you tried tightening all the bolts minus the extra four that you're taking in and out to fold the booms?

Yes that idea comes to mind because I am beginning to be concerned that a crash on a boom will result in the frame breaking and not the boom. On my old quad the arms could fold in upon crashing, but mounting the booms with two screws in this new build will prevent that. I will try the Bat Bone style friction fitting, but as you said I am concerned the booms will shift in flight and the props will come in contact with each other. Currently I am trying to find a material that will secure the booms through the second bolt hole, but in the event of a crash will break and allow the booms to fold in.

 

[Ace2317]

You could probably do small zip ties. They would break in a crash and not hurt the booms (theoretically, anyway) and you could cut them when you want to fold it. A bag of the smallest ones costs little to nothing for a ton of them.

 

[Snarls]

Just updated post #3 with the completion of the build! Now on to some tuning to hopefully get a flight video out soon. Depending on how well that goes I will also mount the gimbal.

 

[Snarls]

Woke up this morning excited for a maiden flight and it was snowing! Luckily by the time I got my gear it had stopped, but part of me wanted to maiden it in the snow. I took it out having not touched any of the settings in the Flip 1.5 and the thing took off with ease. The PIDs seemed to do the job, but I will adjust them a tad for my comfort. Other than that, the throttle needs some expo, the yaw is not sensitive enough, and I need to be more confident in my flying skill!

 

[JAG]

Care to give some details on your gimbal setup?

 

[Snarls]

Care to give some details on your gimbal setup?

Sure, the gimbal frame itself is made of two ply plexiglass that was bent through heating into 90 degree angles. The gimbal motors I use are these ones from HobbyKing. The board is the 2-axis gimbal controller also from HK.

I am running BruGi_050_r217 I believe, or it could be the version before that. Tuning is a pain, but it works. Unfortunately my control board died randomly (possibly shorted from a bolt) and now only talks to the computer. Luckily I have an extra board to use on this project, I just hope I have my PIDs saved somewhere...

Edit: Forgot to mention that there are two places to adjust the gimbal for balance. You can slide the gimbal side to side where the roll motor is, or you can raise the camera platform up or down where the pitch motor is.

 

[makattack]

Nice! I like your designs and DIY ingenuity!

 

[Snarls]

Nice! I like your designs and DIY ingenuity!

Thanks! As Josh Bixler says "it's because I'm cheap"!

 

[Snarls]

So Maybe Motor Layout is Not All That Important

Took the quad through two batteries today. After reorienting the Flip 1.5 for better access to the USB port I loaded the PIDs from my old quad to the new one and took it out to fly. It flies much better now except there is a very slight wobble/oscillation that I'm thinking is just from slightly high I gain. With fairly slow non aggressive flight around the yard I was able to get 8 minutes of flight with my alarm set at 3.7v. Later I did a strict hover test with the alarm set at 3.5v and got 15 minutes of hover time!

For the first flight I decided to not use the second bolt holes and instead rely on friction to keep the arms in place. I was flying around the yard enjoying how well the quad flew when I noticed that the arms looked strange. I came close to check out my new motor layout:
*ignore those cat drawings

I'm surprised how well it flew even in that H-Spider configuration. Even two of the props were overlapping with 5mm of clearance.