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* The fancy ESP32 board could have been replaced by a commodity ESP8266.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.
Yeah, I was really impressed with the style of the project but got a little triggered by the extruded aluminum. "I don't want to pay a lot" -- proceeds to use engineered metals. And while it is convenient, it is softer and more likely to bend out of shape than just regular wood.
A better result at lower cost would come by making the entire piece from wood. You would need a few handtools and nothing else. Plenty strong and totally gap free.
Plus the huge bonus of acquiring some woodworking skills if you don’t have them.
I cant help but wonder, are these kind of posts on the HN front page because all the programmers are nervous about their jobs and looking to switch to maker/woodworking jobs? It scratches some of the same itches, for sure...
I feel like most tech-oriented folks are just simply all about solving problems.
I've always enjoyed doing stuff in the physical realm MUCH more than doing pure systems and software stuff. There's something about everything about the process between having an idea and getting a physical object at the end that you can touch and incorporate into your daily life. "I made this," is about a million times more satisfying to say than, "I bought this."
Transitioning into a maker role is VERY difficult. Every once in a while I tangle with the idea of building cabinets, tables, etc and every single time, the math just doesn't work out. Once you factor in labor costs (your time), you cannot build furniture at anywhere near the cost-effectiveness of a traditional cabinet shop. If you go up-market and try to cater to rich folks, then you're stuck with a much smaller market, which is made even smaller by the fact that those folks are looking for exceptional quality and who are usually more interested in creativity and authenticity than an amateur like me could ever master. They want a third-generation master woodworker who has been doing it for 50 years and built the cabinets that were installed in all of their friends/neighbors/other fancy homes in the area. They don't just want cabinets/furniture, they want a story to tell their friends and family.
But making my own furniture, cabinets, and shop tools as a hobby. Yeah, I dig that and get a lot of enjoyment out of it.
My grandfather was a custom cabinet maker and my dad was in construction. I've been in woodshops before I saw a computer (yes, I'm old enough for that to be possible). No matter what I've done for actual jobs, I've always enjoyed going back to building a physical thing.
engineers have long been interested in woodworking as a distraction from work. its one of the many dream jobs/hobbys for engineers when they get enough money
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.
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.
For sure.
My mental model when starting this project was "panels attached to a frame", and originally, that paneling was meant to be purely Plexiglas. When I pivoted to the bedroom lamp, I switched to wooden panels but the mental model stuck. The upside of not using glue is that it can be taken apart easily for transporting.
But yes, do not build it this way if cost efficiency is a factor.
Nice Work! Take note of the usage of aluminium extrusions, if you're not familiar with them, take a look, they are very handy and not expensive. I've built a 10 inch mini rack with them, really handy.
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
I'm very confused at the aluminum frame - it would have been far easier to just buy a $5 (?) piece of pine (or equivalent softwood) to make the back frame. The pine can also be customized with a handsaw. I guess I would have rather spent more money on the visible part (walnut planks instead of a used countertop?) than the invisible part.
Agree that the aluminium extrusion is not necessary to achieve the look. The other, more ambitious living room lamp that I had thought about first would have required it, so it was a good opportunity to try out this material anyway.
Based on the other comments, I see the unconventional design gets people riled up. Didn't know it was possible to build lamps in the wrong way, hahahaha.
It is SO hard to make things (as someone who also dabbles in wood). It looks awesome, has exactly the light patterns you want, and doesn't make an audible buzz - awesome!
And you've also learned something about aluminum extrusions!
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"
This is awesome and I love it! Congratulations, and welcome to realizing that once you can make things, you can make exactly what you want, regardless of whether it's practical or economical (as you discovered). If you have a partner, good on them for not joining the pool for 2034 or stabbing you with the aluminum partway through the project.
A couple notes:
The customary solution for hanging something on a wall such that you can hang and un-hang it is a French cleat [0]. They're best for flat-ish stuff, but they work pretty well for shallow cabinets as well. You have a cavity in the framework that you can hide it in, along with the electronics.
I don't have a huge beef with using aluminium for the frame, though I'd have used aluminum personally. I agree with others that wood would have been cheaper, but I'd be generally worried about the electronics in the cavity not getting enough airflow to stay cool. The brick wall should sink some heat, and the aluminum should help move heat better than wood would. Honestly, I see the aluminum as a pretty expedient solution to prototyping this. You don't need to save every penny if you're only making one of something.
Wood movement is a thing. I'd edge glue the two pieces of beech, and then follow the traditional solution of mounting to the frame with one fixed point and one floating point. Using a wood frame wouldn't have gotten you out of this: in practical terms, wood is constant in length, but has noticeable changes in width. Attaching pieces with the width running across the length is called cross-grain construction, and is best avoided. A common solution is table buttons[1].
For affixing router templates, double-sided tape works well. You want the stuff with the thickness and texture of duct/duck tape, sometimes sold as carpet tape. Don't use the thin stuff from the office supply. DON'T REUSE THE TAPE. It never sticks as well the second time, and a router can do a lot of damage in a hurry if the template moves. You can also just use a single sheet of paper glued between the template and the workpiece. This is called a paper joint[2], and it's commonly used for split spindle turnings and bowl turning.
Anyhow, the design is great, and it brings a lot more to the space than either of the commercial options. Keep on building, and please keep posting!
Now that you mention it, French cleats and other rail-based mounting solutions were on the table at some point. I discarded the idea because I thought it couldn't be made to work with the shifted center of gravity (I didn't have a solution in mind at that time). Well, I took an informative detour there.
Cross-grain construction is a new topic to me, thank you! I will take a look.
You can do the same with wood, just do not glue it. On the other hand there is something to be said for making something that looks a certain way (even if it is hidden in the final presentation) for the enjoyment of the craft person. On the other other hand, it can cost more and be more time consuming.
- a few filter-on-top-bag-on-bottom dust collectors
- a router table
- squares, clamps, rulers.
Nothing in here strikes me as particularly exotic to a woodworker US. What are you seeing that's inaccessible in the US? For my money, Europe has cheap versions of track saws when there are no cheap versions of those in the US, so that's one difference, but that's not in the post.
The only things I can think of are: cheap track saws in Europe, and combination jointer/planers. There are two makerspaces near me with wood shops, and one has a French combo machine while the other has two separate machines. It saves floor space, but instead of one machine breaking, two do, so there are tradeoffs there.
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.
Yeah, I was really impressed with the style of the project but got a little triggered by the extruded aluminum. "I don't want to pay a lot" -- proceeds to use engineered metals. And while it is convenient, it is softer and more likely to bend out of shape than just regular wood.
I absolutely agree.
A better result at lower cost would come by making the entire piece from wood. You would need a few handtools and nothing else. Plenty strong and totally gap free.
Plus the huge bonus of acquiring some woodworking skills if you don’t have them.
I cant help but wonder, are these kind of posts on the HN front page because all the programmers are nervous about their jobs and looking to switch to maker/woodworking jobs? It scratches some of the same itches, for sure...
I feel like most tech-oriented folks are just simply all about solving problems.
I've always enjoyed doing stuff in the physical realm MUCH more than doing pure systems and software stuff. There's something about everything about the process between having an idea and getting a physical object at the end that you can touch and incorporate into your daily life. "I made this," is about a million times more satisfying to say than, "I bought this."
Transitioning into a maker role is VERY difficult. Every once in a while I tangle with the idea of building cabinets, tables, etc and every single time, the math just doesn't work out. Once you factor in labor costs (your time), you cannot build furniture at anywhere near the cost-effectiveness of a traditional cabinet shop. If you go up-market and try to cater to rich folks, then you're stuck with a much smaller market, which is made even smaller by the fact that those folks are looking for exceptional quality and who are usually more interested in creativity and authenticity than an amateur like me could ever master. They want a third-generation master woodworker who has been doing it for 50 years and built the cabinets that were installed in all of their friends/neighbors/other fancy homes in the area. They don't just want cabinets/furniture, they want a story to tell their friends and family.
But making my own furniture, cabinets, and shop tools as a hobby. Yeah, I dig that and get a lot of enjoyment out of it.
My grandfather was a custom cabinet maker and my dad was in construction. I've been in woodshops before I saw a computer (yes, I'm old enough for that to be possible). No matter what I've done for actual jobs, I've always enjoyed going back to building a physical thing.
engineers have long been interested in woodworking as a distraction from work. its one of the many dream jobs/hobbys for engineers when they get enough money
Who knows? :) I have to say that for me, making something physical is a greatly fulfilling diversion.
Yep, feels nice to work in both mental problems as well as physical ones.
I think AI has played a role in more analog hobby adoption.
Based on this post the makers are safe.
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.
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.
For sure. My mental model when starting this project was "panels attached to a frame", and originally, that paneling was meant to be purely Plexiglas. When I pivoted to the bedroom lamp, I switched to wooden panels but the mental model stuck. The upside of not using glue is that it can be taken apart easily for transporting.
But yes, do not build it this way if cost efficiency is a factor.
Nice Work! Take note of the usage of aluminium extrusions, if you're not familiar with them, take a look, they are very handy and not expensive. I've built a 10 inch mini rack with them, really handy.
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
I'd have put the effort into replacing the "Raufaser" tapestry with a plastered wall, but that's personal preferences. ;-)
German authenticity. The wall also isn't flat, so the lamp sits hollow in some spots and light is bleeding in all kinds of places.
One further argument to replaster it ;-)
Glad you got it done!
I'm very confused at the aluminum frame - it would have been far easier to just buy a $5 (?) piece of pine (or equivalent softwood) to make the back frame. The pine can also be customized with a handsaw. I guess I would have rather spent more money on the visible part (walnut planks instead of a used countertop?) than the invisible part.
Thanks, it took a while.
Agree that the aluminium extrusion is not necessary to achieve the look. The other, more ambitious living room lamp that I had thought about first would have required it, so it was a good opportunity to try out this material anyway.
Based on the other comments, I see the unconventional design gets people riled up. Didn't know it was possible to build lamps in the wrong way, hahahaha.
It is SO hard to make things (as someone who also dabbles in wood). It looks awesome, has exactly the light patterns you want, and doesn't make an audible buzz - awesome!
And you've also learned something about aluminum extrusions!
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"
Temperature safety test was a nice touch; looks incredible.
I thought the article was going to be about making some wire from scratch and milling a light bulb socket!
This is awesome and I love it! Congratulations, and welcome to realizing that once you can make things, you can make exactly what you want, regardless of whether it's practical or economical (as you discovered). If you have a partner, good on them for not joining the pool for 2034 or stabbing you with the aluminum partway through the project.
A couple notes:
The customary solution for hanging something on a wall such that you can hang and un-hang it is a French cleat [0]. They're best for flat-ish stuff, but they work pretty well for shallow cabinets as well. You have a cavity in the framework that you can hide it in, along with the electronics.
I don't have a huge beef with using aluminium for the frame, though I'd have used aluminum personally. I agree with others that wood would have been cheaper, but I'd be generally worried about the electronics in the cavity not getting enough airflow to stay cool. The brick wall should sink some heat, and the aluminum should help move heat better than wood would. Honestly, I see the aluminum as a pretty expedient solution to prototyping this. You don't need to save every penny if you're only making one of something.
Wood movement is a thing. I'd edge glue the two pieces of beech, and then follow the traditional solution of mounting to the frame with one fixed point and one floating point. Using a wood frame wouldn't have gotten you out of this: in practical terms, wood is constant in length, but has noticeable changes in width. Attaching pieces with the width running across the length is called cross-grain construction, and is best avoided. A common solution is table buttons[1].
For affixing router templates, double-sided tape works well. You want the stuff with the thickness and texture of duct/duck tape, sometimes sold as carpet tape. Don't use the thin stuff from the office supply. DON'T REUSE THE TAPE. It never sticks as well the second time, and a router can do a lot of damage in a hurry if the template moves. You can also just use a single sheet of paper glued between the template and the workpiece. This is called a paper joint[2], and it's commonly used for split spindle turnings and bowl turning.
Anyhow, the design is great, and it brings a lot more to the space than either of the commercial options. Keep on building, and please keep posting!
Thanks also for listing parts/suppliers/etc.
[0] https://en.wikipedia.org/wiki/French_cleat
[1] https://www.popularwoodworking.com/editors-blog/attaching-a-...
[2] https://www.wwgoa.com/video/bowl-turning-paper-joints-000224...
Thanks a lot for the comment.
Now that you mention it, French cleats and other rail-based mounting solutions were on the table at some point. I discarded the idea because I thought it couldn't be made to work with the shifted center of gravity (I didn't have a solution in mind at that time). Well, I took an informative detour there.
Cross-grain construction is a new topic to me, thank you! I will take a look.
extrusions also let you re-tap and move standoffs later, glue doesn't. worth the premium imo
You can do the same with wood, just do not glue it. On the other hand there is something to be said for making something that looks a certain way (even if it is hidden in the final presentation) for the enjoyment of the craft person. On the other other hand, it can cost more and be more time consuming.
I love European woodworking equipment. It's so much better than what we can get in the US.
The only tools I can see in the post are
- A horizontal band sander
- A table saw with a rip fence
- a few filter-on-top-bag-on-bottom dust collectors
- a router table
- squares, clamps, rulers.
Nothing in here strikes me as particularly exotic to a woodworker US. What are you seeing that's inaccessible in the US? For my money, Europe has cheap versions of track saws when there are no cheap versions of those in the US, so that's one difference, but that's not in the post.
Do you have examples? I have seen a few US woodworking/carpentry videos and the tools seem pretty much the same.
The only things I can think of are: cheap track saws in Europe, and combination jointer/planers. There are two makerspaces near me with wood shops, and one has a French combo machine while the other has two separate machines. It saves floor space, but instead of one machine breaking, two do, so there are tradeoffs there.
That's a nice lamp!
[flagged]