He'd never designed a PCB. Now Tilt Tracker is taking pre-orders.

Melker Wändahl turned an everyday annoyance into shipped hardware: a coin-cell smart sensor designed in Flux over about four months of weekends, now selling as an 89.9 USD kit. It's the first PCB he ever designed.

“If no one had made Flux, I would probably have tried to make a similar platform.”
‍– Melker Wändahl, creator of Tilt Tracker

It started with a door he couldn't stop worrying about

Melker rents his apartment in Sweden, which means no smart lock allowed. All he wanted was to know, from his phone, whether he'd locked the door or left the oven on. No product did it his way, so he designed one.

Tilt Tracker is a small coin-cell sensor that sits on a door handle, oven handle, or garage door, watches its tilt angle, and notifies your phone if something is still unlocked or switched on. It sets up in under five minutes, runs 3 to 5 years on a battery, and needs no wiring and no landlord permission.

The catch: turning that idea into real hardware meant designing a PCB, and Melker had never done that. He's a trained engineer, with a BSc in electronics and an MSc in quantum physics, but board design is its own craft, and neither degree taught it. Like a lot of capable builders, his projects had always stopped at the breadboard.

A first PCB with expert-level problems

As a product, Tilt Tracker sounds simple. As a first PCB, it's loaded with the problems that make even experienced engineers wince:

  • RF next to metal. A BLE antenna with a CR2032 coin cell mounted directly under the board. Melker was attempting to design his own on-metal antenna from scratch.
  • A radio on a coin cell. BLE draws current in spikes that a CR2032 can't naturally supply.
  • Real layout decisions. Trace routing, component placement, and layer count. None of it is anything breadboarding teaches you.
Tilt Tracker in Flux, 2D layout and 3D view side by side. The coin cell takes up most of the board, so every other part has to fit around it.
The Tilt Tracker board had to fit BLE, a CR2032 coin cell, and an enclosure-friendly form factor.

In Flux, the tool stopped being one of the problems

His first attempt was in KiCad, which piled tool friction on top of the hard problems: small, hard-to-diagnose bugs like overlapping copper appearing out of nowhere, manual imports for any part not in the library, and no experienced EE around to ask.

The way he found Flux says a lot. New to PCB design, he was leaning on ChatGPT for everything, and realized someone could wire an AI like that directly into a design tool. Someone had. “This had already been done in Flux, so I thought it would be interesting to try it. If no one had made Flux, I would probably have tried to make a similar platform.”

The switch flipped all three problems. Importing components, design tips, component placement, the interface itself: “all of this was much easier compared to KiCad,” in his words. And when he hit questions that are nearly impossible to describe in a text box, like “how do I route this trace properly?”, he could ask real engineers in Flux's Slack support channel and get answers in context.

What changed with Flux

Before Flux, Melker had a breadboard-level idea, a difficult RF and power problem, and a PCB tool that made every unknown feel harder to diagnose.

After Flux, he had a smaller manufacturable two-layer board, a proven BLE module strategy, a coin-cell power approach that could handle radio current spikes, and a product page taking pre-orders.

That is the gap Flux is built for: people with enough technical taste to know what they want to build, but without a full hardware team around them. Flux doesn't replace engineering judgment. It gives builders a place to ask better questions, make layout decisions in context, and move from “I can prototype this” to “I can manufacture this.”

The board got simpler, smaller, and actually buildable

Three interventions took Tilt Tracker from stuck to manufacturable:

  • 4 layers became 2. Vasyl from the Flux team reworked the four-layer PCB into a two-layer design with simpler parts. The cost barely moved (about 5%), but the board got significantly smaller, which is exactly what a sensor that lives on a door handle needs. “Difficult to figure out myself,” in Melker's words.
  • The antenna he didn't have to invent. Instead of a custom on-metal antenna, the team pointed him to a u-blox module that only needs a tiny external antenna.
  • BLE on a coin cell, solved. Large bulk capacitors smooth out the current spikes, letting the u-blox BLE radio run on a CR2032. That's where the 3 to 5 year battery life comes from.
“I got help from a few employees at Flux. Especially Vasyl, who made my 4-layer PCB into a 2-layer PCB and chose simpler parts for me. This would have been difficult to figure out myself.”
‍– Melker Wändahl, creator of Tilt Tracker

Explore the design on Flux: Flux project: nrf52 tilt sensor CR2032

The result: a working product taking orders

  • A live product page: the full kit is taking pre-orders at 89.9 USD, with a crowdfunding campaign in the works to fund manufacturing at scale.
  • A final design reached in about four months of weekends. This was a nights-and-weekends project, not a job.
  • A working prototype that does what he set out to build, like an alert if the door is still unlocked after 45 minutes.
  • A manufacturable two-layer design with simpler, easier-to-source parts than the original four-layer version.
Tilt Tracker's physical stack: lock hardware, silicone adapter, enclosure, and CR2032 battery.

This used to take a team

Flux's mission is to take the hard out of hardware, and Melker's story is what that looks like in practice. One person took an everyday annoyance, designed the hardware for it on his first attempt at a PCB, and carried it all the way to a product page taking orders.

Not because the hard parts disappeared, but because he finally had the right tool — and the right experts in reach — when those hard parts showed up.

He's not stopping, either. Next up: a crowdfunding campaign to fund manufacturing, a microfluidics chip that automates cell lysis in a biology lab, and thermal-imaging drones that monitor wildlife for farmers.

Try Tilt Tracker, and tell him what you think

Tilt Tracker is in pre-orders now, and early feedback and orders directly shape whether it gets manufactured at scale. If you've ever wanted more independent hardware to make it out of the prototype stage, this is one worth backing.

Let’s start building.

Let’s build your idea together. All you need to do is describe what you want to make, or what problem you want to solve. Flux will work with you to make it a reality.

Try it Free
Close-up of a Flux-designed printed circuit board