Hi, I hope this finds you well.
I recently posted an introduction as a new builder/pilot and had since built a little 3" with 1504’s using the ImpactRC Alien frame.
It’s been a ton of fun to play with, coming in under 250g even with a 850mah 4s (flying a 730mah in practice).
With the way stock runs in this hobby and with uncertainty around global trade I thought for the following build I would just double up on components, and that way if there is an issue sourcing something I can at least keep one flying.
As I had cleared my favourite hobby store of their last 3/3.5 inch frame I took it as an excuse (clearly didn’t need much) to design some custom plates, I also thought it could be a good practical way of understanding the processes involved.
Introducing El Toro, the 3(-3.5 maybe)", chonky boi freestyle frame.
Modelling was done in ZBrush 2022 (primarily sculpting software), as I had a licensed copy and a lot of comfort with that application. After export, elements were imported to FreeCAD and converted to step files through the Part → from Shape → Simplify → to Solid process.
I had watched a fair few hours of Chris Rosser videos at this point, and had absorbed a bit in terms of topological optimisation and target resonance characteristics, so was under no illusions of how far from optimal some of my choices may be.
I will actively incorporate best practices in future designs, but for this I wanted the only constraints to be on the physical machining (no sharp internal angles etc) with a focus on seeing it through to the end, rather than getting lost so much in trying to build a “winning” frame (as the road to any sort of competitiveness tends to be pretty narrow).
I chose a bull for the overall form as a nod to traders and investors, and to go with something a bit different (you rarely see wide frames, I’m sure for good reason).
It does feel like giving up space without much in return asides from some weight to throw around, but also provides a nice area for carrying the battery in a horizontal configuration and centers everything like a toothpick.
I designed this jigsaw junction puzzle piece connection for the arms, to increase rigidity while remaining serviceable.
I question if anyone will rush to steal the idea as it does add complexity in being unable to simply slide an arm out, but it serves a core purpose while giving a striking visual through the underside window.
All the cut outs topside are functional, either providing through holes to the screws that release the plate clamping the arms down or providing a clip for straps and other accessories.
The underside is almost all for show, exposing some of the internal design choices.
The frame attempts to portrait not just some character, but some characters. In stylized form the top plate plays home to a bear, a wolf and a fish, forming another nod to the market and acting as a signature from another life.
The bottom a shaman waving gloves that depict cheering (or screaming depending on the angle) hooded passengers.
The TPU mounts do a good job of maintaining the overall squared off aesthetic, while the front attempts to draw subconsciously towards a depiction of tusks on a samurais mask.
I like the way the design works but in terms of pure practicality they could offer more in the way of protection and functionality.
They also are not the easiest to print, though tree supports and good part cooling seem to be the trick.
I played with some printed Velcro snapped in place to the topside snigglet cut outs. It was interesting, if using double sided tape on a flat side of printed Velcro fixed to the battery, it created a pretty compelling battery pad.
It deserves more exploration, however I had some issues with the size of the Velcro beads crumbling under use and when lazily resized a reduced effectiveness, so abandoned in favour of the single strap for now.
For electronics I doubled up on the sometimes poorly reviewed, yet reliably cheap and cheerful Speedybee F405 mini stack.
I also used their nano ELRS receiver and chose a RushFPV Tiny Tank for VTX.
Cam is the baby ratel 2. The motors are Xing 1504 3100kv, using 4s 730mah Coddor Li-HV batteries.
I guess riding high with over confidence from the first two builds I rushed in finishing up the soldering of the last component (VTX to flight controller), in doing so I left a little extra wire at the end of the leads and rather than unsolder and reseat I attempted to snip an end off, where by the force of the clippers against solder lifted the VTX pad (one of the few pads I don’t think you can remap).
I believe I can repair it, the pad has kept its form and it was a clean tear. I have a scraper pen to reveal a trace if need, solder mask and a UV torch, even a small hot bed.. but no conductive epoxy, which I think would really be best way to attempt a rescue, or maybe I can use it in a digital build but for now I swapped it out with another.
I had watched many videos of failed 3d printed frame attempts and read lots of messages of why if you are going to 3d print a frame it should be either some optimised organic shape or an enclosure, but I knew I didn’t want to build those.
I knew it should be machinable in CF plates at the end, and understood that you could get away with more on smaller quads.
So I thickened the plates for 3d printing and hoped the design of the independent arms with the clamp plates would introduce some level of damping while remaining stiff enough to not rattle out or overheat the motors.
I used PET-CF (as it was the strongest filament I had, initially for motorbike accessories). I dried the filament for 10 hours prior and kept it in the drier while printing for consistency.
These were printed using a .5 hardened steel nozzle with a .18 line width, 5 walls and 100% infill
The initial prints where good but slightly off in size (exporting to STL and then reimporting to the slicer had messed with the scale), it took several goes to get as close to as few gaps between walls but after a few attempts and a dry component run I felt good to solder everything down.
I think PET-CF is a nice fit, as it is very firm with high temperature resistance (looking at the VTX) but I am concerned it might shatter too easily.
The plates come in at 46.7 grams though demand a lot of hardware in the form of nuts and bolts.
Up until this point I had thought myself rather clever cutting slits in the TPU and mounting the front standoffs with a 20mm clearance in such a way that just allowed the 3520 Gemfan hurricane props to pass by.
I had based the motor separation on the Juno 3.5, and had never once thought to check for overlap between the span of the arms.
They just slightly overlapped and I was mildly crushed by such a rookie mistake, but it had initially started out as a 3" build concept and I wasn’t about to change the design of the arms at this point, so I switched to my trusty 3" wind dancers.
First hover tests went really well relative to my suitably low expectations, I wasn’t sure if it would fly-a-away or shatter to pieces on landing.
While fighting the wind I managed a few small punch outs and while landing on a chair the drone bounced off hitting the concrete on the end of one arm, yet came away appearing undamaged.
It’s a small success as it felt in that short time just like hovering the Alien, though I expect issues will become apparent.
I have included initial black box logs as a point of reference though there isn’t much to take from a hover test:
Blackbox for El Toro.zip (3.2 MB)
Will get some park data tomorrow along with looking to order some carbon fibre cuts for comparison.
I am asking for recommendations on that front, if you know a good service for single lot size CNC machining in this region, please share.
All the best, thanks for reading.
