I Spent 8 Months Building a Wall Lamp From Scratch

A developer with zero woodworking or LED experience documents the 8-month journey of designing and building a custom wall lamp for the bedroom. From buying a second-hand beech countertop for €15 to routing pockets for custom steel mounting plates, the build involved extensive planning, CAD modeling, and a few bets from skeptical colleagues. The final lamp features individually controllable SK6812 LEDs, WLED software, and a clever hidden mounting system, proving that ambitious DIY projects are possible even without prior expertise.

I estimated that starting and finishing this project would take me 2 years, with a 99% chance of it never being finished.
  1. mft_

    Nice write-up and a nice outcome, but speaking as a reasonably successful home/amateur woodworker, this is very over-thought and over-engineered. I say this not to criticise the creator as I appreciate that they are new to this, but to help others considering such pursuits.

    Some constructive feedback:

    * There is no need for the extruded aluminium frame - some cheap wood would have sufficed as a frame

    * The CNC'd hanging brackets are totally unnecessary - a couple of holes, one either side, with a keyhole hanger over each would have sufficed

    * The balancing and 3d-printed endcaps is unnecessary assuming you've got reasonably well-placed hangers

    * The fancy ESP32 board could have been replaced by a commodity ESP8266. (You could get rid of the ESP altogther and just buy pre-wired LED strips, but then you lose the opportunity to program nice lighting effects.)

    As a result, the bill of materials is multiples of what is could have been. I estimate that you could replicate this in a single weekend for ~80-85 EUR (15 EUR countertop, 10 EUR wooden frame, 5-10 EUR plexiglass, 40 EUR LEDs + PSU, 5 EUR ESP8266, 1 EUR for two keyhole hangers) assuming you have a few screws and some wood glue available, plus whatever the cost of the markerspace is. Everything could be sourced from your local DIY store, plus an order from Aliexpress for the electrical components.

  2. MisterTea

    No disrespect but this is pure vibe engineering. Aluminum T-slot? CNC machined parts? CPU? Unnecessary design leading to being way overbuilt with lots of material and energy wasted. My advice would be to take a wood working class where they let you build a simple project of your choosing. Your wall lamp would be a perfect project, plastic accents included. You'll learn a real skill.

  3. rainbowDolphin

    Great job MB, on both the lighting fixture and the blog post. Design #1 was certainly the right call.

    As for the comments about it being over-engineered, I agree. Yes, you could have used French cleats over the extrusions, and yes, you could have used a chisel and avoided having to print a routing jig.

    But I’m sure you learned a lot, as evidenced by the difference between the first and latter pockets.

    Time in CAD, 3D printing, and structural materials selection is crucial for production runs, but you might be surprised by how much sooner you can finish a one-off project by prioritizing a pencil, paper, and hand tools (where sensible of course).

    I say this not to criticize, but to encourage, as I’m sure you would like to finish the next project much sooner.

    You ended up making a fixture that looks and functions as well or better than the €600+ one, and in that sense, this is a resounding success.

    You also started with IKEA detritus for cost savings and ended paying more than €600+, which is pretty funny! Might not be a bad idea to start the next one with some nicely figured solid wood board (and maybe use those routing skills for some inlays…)

    But again, great job.

    Sincerely,

    Former carpenter, machinist, & lighting fabricator/designer

  4. JR1427

    I think you could decrease the costs a lot. I don't see the value in using aluminium extrusions to stand it off from the wall. You could use softwood for that, and glue, and negate the need for expensive custom connectors.

  5. abrookewood

    At last ... realistic expectations! "I estimated that starting and finishing this project would take me 2 years, with a 99% chance of it never being finished. So I just hung paintings instead"

More from this day

2026-09-10