Obsessing Over the Motion of my DIY Dyson Task Light - DIYson Lamp Build Log #3
Вставка
- Опубліковано 9 тра 2022
- Welcome back to the DIYson build log where I recreate an expensive Dyson light using off-the-shelf and 3D printed components. In this build log I'm making some changes to the motion system, all in the name of achieving that smooth motion that I was looking for in the last video...and I may have found it! It starts with one big change which ends up generating a bunch of secondary enhancements to other components. It's a really fun one I think. Enjoy :)
Current project files repository: github.com/stevenbennett/DIYson
Parts mentioned in this video (OpenBuilds links are affiliate links):
20x40 Linear Rails: openbuildspartstore.com/v-slo...
Xtreme Solid V Wheel Kit: openbuildspartstore.com/xtrem...
Xtreme Mini V Wheel Kit: ttps://openbuildspartstore.com/xtreme-mini-v-wheel-kit/?ref=_2E7m4SXRIy8t
Aluminum Low Profile Nyloc Nut: hobbyking.com/en_us/m5-nylon-...
Super Low Profile Nyloc Nut: www.getfpv.com/m5-black-alumi...
My favorite filament: amzn.to/3j5RVws
Subscribe to the channel: / @stevenbennettmakes
#3dprinting #diy #maker #design - Наука та технологія
I preferred the 20x20 design more. It looks more natural. I think if you changed the base to also have a bearing, though hidden, and having the counterweight on a continuous belt with a little bit of tension, the counter weight would only need the most minimalistic of contact points with the rail to stay in line.
you could have tried POM material for good sliding effect with no scrubbing. Keep the first prototype and get a good sliding plastic. Also by time plastic gap widens makes it easier to slide.
I would have taken the first design of the weight carriage with no wheels and added small stubs that would contact the extrusion as opposed to the entire carriage just rubbing again the extrusion. You could even make the stubs inserts that you could make out of delrin or ptfe for lower friction. The stub would make the contact area a point compared to what you had which was the entire aluminum extrusion face.
Anyone who arrives at this solution is an utter genius in my book.
I was thinking the same. I really dislike the tabs and wheels sticking out and also the extra wide rail. This is a perfect job for either a nylon bush (look at igus, they do all sorts of material for this application) or a nylon T which hooks into the channel with a ball detent pushing against the bottom of the channel to remove backlash.
Alternatively you could make the wheels internal to the counterweight and make them nearly invisible (apart from mounting hardware perhaps). This would make it possible to limit a holding T to one, or maybe not even needed if you make the belt a loop fully round.
Yes, I thought of placing the two small wheels inside the carriage, hidden, so that only the wheels touch the rail. And to prevent the carriage from flying of the rail, just a tab at the bottom of the weight carriage. which in normal use doesn’t touch the rail.
Using wheels for this is overengineering to the max. You can buy COTS (off the shelf) PTFE stickers similar to those you would find on a computer mouse, they would be excellent for this purpose.
The first version of the counterweight without the wheels definitely looked the best. I bet with some design improvements and maybe sandpaper you could get it to work smoothly.
Just stumbled across your channel and I can't stop watching. I also prefer the larger wheels. I thing they give a more industrial look to the lamp which although deviates a bit from the slick design of the original, makes it visually very interesting.
I am digging this!! I wasn’t even aware of the Dyson light.
I would go with the smaller wheels and give them fairings (create shroud to enclose them)
As you were deciding on lock nuts I almost yelled at my screen, “SANDPAPER! Run them over sandpaper and make them thinner!”🤣
Definitely like to look at the smaller wheels but if the bigger wheels are smoother I don't know. It's too bad somebody doesn't make some sort of in track roller that would allow the slim design of the first version.
Nice job, can’t wait to see the next update
Thanks for checking it out, Jerry!
When testing the desk-mount my result is a wobbly connection. I guess it is due to the 500mm aluminium profile having a high torque on that small pivot point.
- I already tried editing the 3d printed base to fit a 10mm diameter aluminium tube to increase axle rigidity.
It kind of works but not to my satisfaction
- I would try to fit a bearing with 10mm inner diameter for the axle and a bigger outer diameter to have a bigger surface (limiting the torque). Some redesign for the 3D printed base will be needed tough
I will spend some time this week dialing in the desk mount (specifically the hole that accepts the 40mm spacer). I've had to really dial mine in to a super snug fit to eliminate wobble, but I've noticed that the opening in the Express STL is too big. The bearing option is also promising, and I'll be looking into that too. Thanks for letting me know :)
I've used felt to get a more silent sound. You will still hear it sliding up and down but it will be a very pleasant sound. If you can't completely high something then make it a feature.
First of all, your videos are addictive :)
One thing you can do to avoid the belt flopping when you lift the X-axis fast is, introducing a 3rd smaller pulley at the pulley assembly, between the the 2 pulleys you have, weaving the belt between all three and eliminating the slack. But this is occasional since you never slide up the x axis really fast.
Nevertheless, really great implementation
Very nice build! Maybe you can store batteries in the counterweight. This way you have less cables to hide...
You keep referring to these parts as if they're exclusive to open builds. 2020 extrusion, 2040 extrusion, concentric nuts, bearings, and v-slot bearings are things you can get from most hardware and hobby shops, and even certain rainforest themed online websites... Open Builds also has them, but they're more expensive there.
Fair enough, folks can buy most of this stuff from whichever vendor they prefer. I do think there are practical benefits to limited to BOM to as few vendors as possible; one being that it avoids unexpected incompatibilities that arise from using parts that are similar but not identical (this may be a peace of mind feature for me, but I find it valuable). Another consideration is around minimizing/consolidating shipping costs (this may seem trivial but in my experience adds up). Either way, you're right to point out that the options aren't limited to Open Builds, and there might be better alternatives for many people depending on their circumstances.
I'm binging this project of yours right now, so I haven't caught up to the current state, but it seems to me that the reference design may be using a closed, or looped belt, and their counterweight is just attached as a tracked "barnacle".
I'm going to keep silent and not spoil anything for you :)
Looking forward to the electronics portion! also, can't you solve the bottom asymmetry using a 3d printed part? something perhaps, to allow you to center the pin you use for rotation
You read my mind, Carlos! Look for updates on this in the next video :)
Why not use a button head screw and put the regular nylock nut in the plastic? Would be lower profile and more minimalist.
How are the electronics going? Looks like they used a magnetic contact for the power, and the rest is in the lighting unit? The more I’m messing with this, the more I appreciate their engineering. Pretty awesome. I applaud and appreciate you for doing this!
Hey Joey. The electronics and lighting are going well, thanks for asking. I'm a little busy with work recently so I'm behind on the video, but I'm eager to get it out in the next week or two. I totally agree with you, the engineering behind the original lamp is really impressive. The more I learn about it the more the high price tag seems justified. I hope I'm able to come up with a good alternative design, we'll see :)
What if u use ribbon/dupont style flat cables 🤔
For the wheels, go for the bigger ones, they look good, the small ones are a bit weird.
U can flip the setup, use allen button head (on the nut part) and for the other side u can go for regular nuts, or use threaded inserts, u can try this ones and make the other side ''blind'' (just for a clean look), if that works...omg haha🤯
Have you thought about using retracting cable springs as counter weight like they did with Voron Switchwire gantry?
Steven and Joey: check out Modustrial Maker today. sure enough: LED Lamp. a bit different than this project, but same goals... sort of :)
Very cool! There's something about making lights that's so satisfying...
Small Wheels Everywhere
Have you thought about running the counterweight wheels but internally (so rotated on the back plane), as your using lead balls I would imagine they would fill the void just fine and then you would get back that minimal discrete look. Or do you see some complications doing it this way?
This is great, I was looking at ding something similar, and maybe adding a linear rail on the back side would guide the counterweight? Also, are you posting your step/stl files along the way? I’d love to do this along side with ya
Hey Joey! I can definitely share access to the STL files. I'm curious about your preference with these; would you prefer to have STL files available as soon as each build log comes out (even though they're likely to change) or would you prefer to get them closer to the end of the project when things are more finalized?
@@StevenBennettMakes if you are building this in real-time, I'd recommend you wait until the end to share files and parts lists -- to give you the chance of change your mind along the way or improve on something.
After each log would be great! I have a few things in mind that would fit my needs, and then maybe If I can’t get it right, I’ll just use the end one!
The F3D files or step files would be best, because I can make changes as needed as well. Totally up to you!
Cool! I think I can release STLs and STEP files between videos with the disclaimer that the designs will almost certainly change. I will look into it over the next few days. Thanks for the suggestion :)
@@StevenBennettMakes thank you! I can also send you my email if that’s easier!
you can just rotate the screw for more clearance
did you try something like the rectracting mechanism from a tape massure?
I would try to build the wheels into the weight, hidden inside and kept the smaller profile of the orignal design. The smaller wheels should be able to fit inside?
Why not slide a tslot into that vslot and make the tslot half as wide and use 2 mini wheels on each side of the tslot. Way lower profile same benefits
As I am binging I am noticing you keep referring the aluminum extrusion as linear rails, they are not!! Linear rails are whole separate beasts with their own ball bearing fed carriage and hardened steel rails. The ones you got are simple aluminum extrusion with family of standard sizes like 2020 or 2040.
openbuildspartstore.com/v-slot-20x20-linear-rail/
Totally see what you're getting at, but I'm afraid you'll have to take it up with OpenBuilds :)
This guy is right. Open builds use them to make linear rails but they are just aluminum extrusions. Linear rails have a rail, and a bearing block and nothing to do with aluminum extrusions. A stock ender 3 uses extrusion and v slot wheels but they would never sell it as having linear rails.
Haven't watched all the videos yet, but 1) I prefer the 1x1 aluminum extrusion no matter what. So I think you should have spent more time trying to get that smooth than switching to the 2x1. Latest vids look like you did? 2) if you have to go with the 2x1, make ALL wheels small for uniformity.