fp.

We built a wireless, low-profile macropad with a display called the Creavi Macropad. It lasts at least 1 month on a single charge. We also put together a browser-based tool that lets you update macros in real time and even push OTA updates over BLE. Since we're software engineers by day, we had to figure out the hardware, mechanics, and industrial design as we went (and somehow make it all work together). This post covers the build process, and the final result.

Hope you enjoy!

Comments

sulicat•2mo ago

Looks really nice! It looks real professional!

Is there an ESP on there? Are you using lvgl for the display?

cmpx•2mo ago

Thanks!

We planned to use ESP, but turned out it needs a lot of energy, so we are using nrf52840 MCU. Yes, we are using lvgl! Such a nice library!

stavros•2mo ago

This is fantastic. You've detailed a lot of the stuff that ruined my projects, though I never really actually sat down to learn about parasitic inductance and how to combat it. Now I look at the datasheet and make sure I don't miss any of the recommended caps and resistors, though I'm still not at your level of complicated design.

The finished product looks great, well done! By the way, why not add two more buttons at those two blank corners? Is it because the labels wouldn't fit on the display? You could always replicate the layout onto the display itself and put the labels on that.

cmpx•2mo ago

Thanks!

About the buttons in the corners, I wish we could use that space too! :) Most of the electronic components (MCU, power supply circuit) are located in the corners, while the remaining PCB area is occupied by the mechanical switches, display/battery, and the encoder.

stavros•2mo ago

Ahh, I see. Wouldn't it work to put those on the underside?

cmpx•2mo ago

Great idea! Actually the mk3 of the PCB had these parts on the bottom. The reason for putting them on the top layer was that this way we can decrease the thickness of the macropad (we don't need much space between the pcb and the backplate). For now it is basically a tradeoff. One way of improvement could be, that we don't use a BLE module, but solder the MCU directly to the PCB and use smaller passive components. This way the max. height of the components could be decreased and we could put them back on the bottom. This would also mean I have to design the RF parts, which is very far outside of my knowledge.

stavros•2mo ago

Hm yeah, there isn't really much space for the module, huh. One idea would be to cut the PCB and use castellated edges to solder the module to, which will make it thinner by one PCB height, but it'll be tricky.

Plus, it looks like your module has multiple rows of pads, but I'm not sure about that.