12/6/2025 - Planning

Checking online for inspiration on how other cyberdecks look and knowing what parts they use has given me a pretty good idea of how I want my cyberdeck to look and what functions I want it to have.

For example, I want this to be a fully functional data processing device, even off-grid, so that I can receive and process data live. This can enable pretty cool stuff, like live 3D visual simulation of a rocket trajectory. In the future, when I want to do some lightweight hosting, this can also serve as an ultra-portable server!

With all of this in mind, I think it would make sense to start by picking a case to restrict the size of the device.

A more cost-effective solution would be to use CM5 with an IO board as compared to a Pi5 on its own. This also allows expansion like M.2 SSD and makes display connection easier with full-sized HDMI ports.

From my observation, general electronics like panel mount connectors, cables and miscellaneous stuff like hard cases are cheaper on Shopee as compared to Amazon or Digikey. This is highly likely caused by Chinese sellers being able to market directly on Shopee, so I'll list out the rest of the parts needed on Shopee.

and some miscellaneous stuff that is not too expensive and that I have in stock. The ground station module will be in a different project as it is linked with the flight computer.

What's next will be the keyboard and pointer interface. I'm unsure if I should make a custom mechanical keyboard for this project since the cost is already quite high. For now, the running total stands at a hefty 165 USD, so a cheaper alternative would be adapting something like the one attached below.

A final look at what we get with this setup.

The problem with this project is that it requires most of the tinkering to be done after you get the parts. This is because the materials that I get will constrain what I am able to make. The dimensions have to be measured physically for things like panels to be cut and for 3D printed parts to be designed.

12/7/2025 6 PM - More research and sourcing

After some thorough research regarding the power source and pointer input, I realised how hard it is to keep cost low for these two components.

For power source, we need a UPS to seamlessly switch from the mains power source to the battery whilst also charging the battery when plugged in. From what I see this is the reason why most UPS board are very expensive because they have to integrate a BMS, Voltage Regulator (buck-boost), battery charger. The MakerFocus UPS board seems really nice for covering all my needs but it comes at quite a high cost of 23 USD without the battery. I have decided to opt for a cheaper Chinese counterpart that will only cost ~2 USD, but we can work on tinkering it to give it more USB outputs and a switch to control the battery, because from how it looks now it doesn't have a switch and was built for true uninterrupted supply. I have considered building my own PCB here, but an estimate of cost(parts+pcb+shipping) comes out to over the MakerFocus UPS rendering it unfeasible.

With this addressed, I researched deeply into the input options. A mouse does not seem like a good idea because you will need a flat surface to operate it. A USB trackpad or trackball will cost more than 25 USD, this is bad for our budget so I had to work around it. Looking at repos to gain inspiration online, I saw that there are many options for custom built trackpads and trackballs, amongst the different options, ploopy came to my attention the reason being they make both trackpad and trackballs and all their work are open source so it's easy to reference and learn. Here comes to however yet again. Pixart sensor are out of stock everywhere and have strong NDAs meaning they don't have datasheets publicly available. Not only that, ploopy products are at a very high price point around 90 USD, even if we assume a 50% profit margin it would mean that the cost to make is around 40 USD way over our budget (the only reason they get lower is because of purchasing in bulk). This leaves me in a very awkward position, I either have to bite the bullet, work with a CLI-only OS(I would really prefer not to), use a joystick (this just popped out of nowhere) or put a flat piece that can be attached to the side of the case for a mouse.

That's all for now. I will continue thinking through the different possibilities.

12/7/2025 10 PM - Hinge for case

Another thought that came up was the hinge, as the waterproof case comes with a friction-less (not literally, just the type) hinge unlike those found in laptops which are friction hinges they cannot hold themselves in place. This will mean that we will need to come up with a solution for that.

First thing I think of is a gas strut, these seems like they would work great but they come at high force rating and are hard to hold in position (they typically go to their max position). So for this to work it will have to be a weak and small version to fit.

Next we have a folding support hinge, these are cheap and are easily available from hardware stores. However, they can only lock in one position. So once it's set it's there.

Other than these two I saw someone stick an item to the outer side of their case to hold itself open using the environment and the surface it's seated.

I have also thought about having a stand attached to the back of the case, this will have the problem that you will need a surface behind it to function.

Another thought that just came up is to have a 3d printed part attached to the screen side panel, so that when it opens up it can click into place onto something on the keyboard side.

From what I see, the best solution here is to either add a friction hinge or something to support it externally. If going with the friction hinge approach, it would preferably be installed inside the case on the panels or externally only with adhesive. This is because I want to keep the device waterproof. Detent hinges are also an option but if I were to add a hinge I might as well add one that can stop at any angle.