Whilst customising an old Yamaha DT-100 I
needed wanted to replace the huge standard tail-light. I also upgraded to LEDs to reduce the current draw on the bikes generating system. Below is the finished results.
The whole tail unit featured an aluminium registration-plate holder, registration-plate illumination, LED indicators, LED brake/tail-light and was designed to comply with relevant ADRs.
The designed started with an LED trailer brake light purchased from the local auto-store. The curvature of the mudguard was estimated, a 3D model of the light was created and then the part was designed in Solidworks. The Solidworks model was then saved as an .stl file before it was entered into ReplicatorG to create the g-code for the printer.
This is what the part looks like straight after printing. The support structure is still attached.
The part was printed with a layer height of 0.2mm and the layers are clearly visible.
Finishing post-printing required sanding to smooth out the external surfaces, etch-priming (so the paint will stick to the plastic), a layer of spray-putty to smooth the surface more, more sanding, another coat of primer, tapping of the threads and a final coat of paint.
The material is ABS, an engineering plastic the same as your computer mouse and keyboard is made of. It has plenty of strength however there is a definite “grain” due to printing layer by layer and strength is higher with the layers versus across the layers in tension. Post treatments like acetone vapor baths help smooth the surface and bond the layers together.
Yesterday was much more successful with the quad-copter project.
There was still a few intermittent problems with motors not being able to spin up. I gave the battery a charge to make sure it wasn’t the LVC (low voltage cutout) causing the problem but that didn’t solve it.
Next I used the programming card to re-programme the settings in the ESCs (electronic speed controllers). The first time didn’t work and the limited instructions weren’t much help but a second attempt was successful. Tip – double check the jumpers are in the right position on the programming card and leave it long enough to properly program the settings.
The final result was a few successful hovers at around 1.5m altitude (this was just testing it in the common area in front of our house and there were hazards like cars and small children nearby to limit the flights). Initially it had a nasty oscillation in the roll axis but adjustments to the PI settings in the KK2.1.5 control board reduced this. Further tuning in a suitable area will have it flying well.
Above is the initial assortment of parts. I already had the transmitter and just added another receiver.
The original landing gear used “pop-sticks” as the sprung element in the gear. These weren’t strong enough in the initial tests and the landing-gear was redesigned.
The Anycopter build is great as a simple first time build to get started. I recommend watching the FliteTest video on the construction. I’ve learnt a lot and can already see areas I want to improve on to customise it for my planned usage.
The website and blog are now up and running.
Today I also had a successful lift off with the prototype quadcopter.
It’s based on the Anycopter design from Flitetest.com. With my custom designed 3D-printed motor mounts and landing gear. The gear uses fibre-glass rods salvaged from a kite that didn’t fly too well.
The landing gear works well but the ESCs (Electronic Speed Controllers) seem a bit delicate. Self-leveling was switched off on the KK2.1.5 control board and there were a few rough landings, now the front two props aren’t spinning.
Welcome to the Speeddevice blog.
Speeddevice is about engineering, design and research with a constant desire to improve performance and learn.
I hope to have a collection of my projects as they proceed as well as a collection of thoughts, ideas & information.