The problem with OnShape and Fusion 360 is that if (when?) the services are discontinued your designs go away. You can keep the STL/IGES files but they aren't very useful if you need to quickly tweak a dimension. I've been bitten before by Fusion360 refusing to work (in offline mode) without an internet connection because it had decided it was phone-home time.
Whilst it doesn't have all the features of the usual commercial packages (Solidworks etc), SolveSpace is great, cross-platform, a single binary and my preferred open source solution for simple 2D/3D cad jobs [1].
FreeCAD [2] has also come along a lot if you prefer something more like the traditional packages, although I found some annoyances with it's dimension and constraint system when I last used it ~6 months ago due to the fact it has two different sketching modes.
SolveSpace is fantastic, and it's impressive that the entire parametric CAD package is ~2MB, it just never ocurred to me to use it for this. I'll give it a shot, thank you.
You're right about the proprietary/cloud nature of the previous alternatives, which is regrettable because Fusion is a desktop program, and could easily just be saving files locally.
I hope FreeCAD becomes a great alternative soon, because Fusion really needs competition in the space.
I completely agree and was in the same boat - I used to use Fusion because until recently the only open source option I knew about was FreeCAD, and I had various issues with stability, features and the interface. OpenSCAD & friends are neat but if like most you've learnt CAD with a more traditional package it's tricky to adapt to.
Of course the cloud systems have their advantages and uses too!
Nice. I had a similar problem when I rode my bicycle to work daily. I really didn't want to have to carry anything more than my phone.
I ended up hacking a bluetooth earpiece into a remote after seeing the idea online. It was 100% reliable over a period of over 2 years.
Having it mounted in the open like that seems like it could have a few drawbacks. I would still be concerned about water damage, even if you don't plan to ever be out in the rain. What about the fasteners you used? Are they rust-proof? Things mounted on my bicycle that have any metal parts seem to wear out over time, and plastics seem to take on some weathering, possibly from sun.
There's definitely going to be water damage on the plastic, PLA isn't very water-friendly. I'll buy some PETG to print it later on, it should be much more weather-resistant and durable. The nuts and bolts need to be rust-resistant too, yes, although I can always just change those after a while.
The good thing about having a 3D printer is that a replacement part is just 20 minutes away.
PLA definitely will absorb water, which decreases its strength, but the effects may not be that noticeable, at least for a while. I have load bearing PLA parts that have now been outside in the yard for 2 and 3 years, and they are showing no signs of degradation.
Anyway, it would be an interesting experiment for you to stick with the PLA print until you see a problem before you reprint with PETG. My bet is it will last longer than expected.
I just keep a traditional garage door opener in the inside breast pocket of my vest and jacket. Easy to close the garage when I pull out, easy to open it as I'm coming down the driveway. Harley-Davidson has an opener I could get and install directly into the bike (with a pushbutton on the "dash") but it's more than I'm willing to pay since my current solution works well enough.
Huh, I've never heard to keep weight off the keyring. I imagine this is to avoid unnecessary force on the ignition assembly? Obviously you wouldn't want to put several lbs on there, but it seems like small items would be fine.
Regardless, cool project. I'm planning to do something similar, as soon as I can get an indoor space. I might go even more hidden, and hide it under the lip of the gas tank or etc.
I was dubious at first too, when the motorcycle mechanic told me that, but then my car mechanic (completely unrelated) said the same thing, so I took all my keys out of the keyring.
Apparently, weight leads to the lock "loosening up" and can make the key "stick" a lot, unable to turn because the rods no longer align as well.
> I might go even more hidden, and hide it under the lip of the gas tank or etc.
I was thinking about doing that and connecting it to the bike battery as well, but I couldn't find a place that was easier to reach (and that wouldn't require me to lay cables all over)...
Also, it minimises chances of the tank getting scratches from objects attached to the key. At times the things on the keychain get into the well of the lock beside the key and sort of get in the way while turning it.
Nice, Stavros. I've been thinking about the same thing; having my current remote inside a jacket pocket and attempting to poke through the fabric to the button is distracting.
Thank you! I was thinking of publishing this to Thingiverse, but the enclosure is so ad-hoc that it would hardly be of use to anyone. Maybe I'll just publish the Fusion source, though.
The 'Fusion' source? I assume you mean share the 3d model file... and you used Fusion 360 software to create it.
Just curious: can you share your experience modeling this? Why Fusion 360? Because it's free? Do you have experience with Solidworks or Blender or other software? If yes, how would you compare 3d modeling in Fusion vs. other modeling software?
I use Fusion mainly because it intuitive enough to figure out without doing much reading. I tried Solidworks at first, but, as a complete newbie to this, I couldn't make heads or tails of it. As far as I know, both are equivalent in functionality, but the UI of Fusion 360 was much, much better (and yes, Fusion is free for hobbyists, so that's a plus).
As far as I know, Blender isn't parametric, so it's largely unsuitable for doing this kind of work. I would recommend OnShape as a very nice starter CAD program, and it's very very easy (almost trivial) to move on to Fusion afterwards, so my advice would be: If you can get Fusion for free and it runs on your OS, try that. If not, OnShape is a very capable second choice.
Fusion is likely to displace Solidworks and Autodesk's own inventor in the market within the next few years. It's easy, powerful, and most importantly update regularly and with features people want. Onshape is pretty awful, so no threat there. Source: Spent about a hundred million hours in this sort of software ;;' _ '
What do you find awful about OnShape? It's not as nice as Fusion, but it's passable, and it runs on Linux, which is my number one annoyance with Fusion. Requiring me to reboot to design stuff is pretty much a guarantee that not many things will ever get designed :/
> This is because if, when finally screwing the parts together, the screw has to thread its way through, then the gaps between the parts won’t tighten when you tighten the screw, because there’s no travel between them.
That's why wood screws are only threaded partway up their length. Since these screws are self-tapping, they need to account for the problem you correctly identify here, where the screw can't clamp the parts together properly because they cannot move relative to one another as the screw is tightened into place. Ending the thread partway up the screw solves this by permitting the part nearer the screw head to move freely along the shaft, so the fastener can clamp them together properly as it's tightened.
The machine screws you're using, conversely, are threaded all the way up to the head because they are not designed for self-tapping, but rather to be used in holes which have had threads tapped into them. That's why they are giving you the trouble they are. Happily, there are a few different ways to make that trouble go away!
Making the entire screw path wide enough that the screw doesn't have to tap any threads, and securing it with a nut, is one option, and as you've seen, it can work. On the other hand, you lose something in clamping force, because there's less bearing surface, and you also have to choose between machining (or designing) a nut pocket into the part, or having the nut and the end of the bolt stick out past the profile of the case.
Another option, which replicates the benefit provided by the design of self-tapping screws, is to make the screw hole wider, but only in the part nearer the head, and leave the far part's hole narrow enough that the screw will tap threads into it on the first insertion. This way, you have the part near the head free to tighten, which solves the problem you saw with your first designs, and you get the maximum clamping force possible from the screw without having to tap threads for a machine screw - and, with a carefully chosen (or trimmed) length of screw, you can also have the path end in a blind hole in the far part, so that the only part of the screw visible in the finished item is its head. (If you care, of course.)
And a third option, naturally, would be to actually tap threads into your screw holes - but you need a tap set for that, and I'm not sure there's much point. You need taps to cut threads into metal, because metal is pretty hard, and you need something very hard to cut it cleanly so that your screws, which are only pretty hard themselves, aren't ruined when you try to use them. Plastic, on the other hand, is very soft, so you're not actually hurting anything by letting your screws self-tap. You just need to adjust your design to account for it.
Ah, damnit, yes, this was the second thing I wanted to mention in the post (having the last part be tighter so the screw threads its way through) but forgot to write it up in the end.
Thanks for the reminder, that's exactly how these designs should work, you're spot on.
Oh, any time. In any case, I'm sure there's someone here who won't have run across it yet, and for whom it'll come in handy.
On a related note - is it just me, or does the nigh monomaniacal focus lately on high-capital-investment tools like 3D printers and laser cutters seem like it might come at the cost of a lot that's of value? Like, for example, any interest at all in how to encourage and support kids coming up in places and families where a 3D printer, or a laser cutter, is about as likely of attainment as a nice little house on the far side of the moon? I don't know, I kind of feel that way, but maybe I'm the only one.
Well, I got my printer for $250 new, and the print quality is fantastic, on par with some I've seen that cost $1000+. I'm not sure if you consider that high-capital, but I've definitely seen single tools (e.g. saws, and, hell, phones) cost more than that.
I definitely don't think this is moon-house territory, the technology is very accessible, price-wise, especially if you consider that these tools are easily cheap enough for libraries and other communal spaces to buy and make available to the public.
There appears to be a lot of choice and fiddling required to become proficient in 3d-printing. Without having a personal killer-use for one I've so far managed to get several things printed by downloading the designs from thingiverse and paying other people to make them for me, on a per-print basis.
Having 3d-printed things is pretty cool, makes my electronics projects look a lot more professional, but I still suspect if I bought a printer of my own I'd be frustrated and annoyed.
Fair point. But how much effort went into getting that print quality out of a machine at that price point?
I'm still uncertain that the heavy emphasis is of net value, but I'll need to spend some time considering why I feel that way before I can make a coherent argument.
I couldn't tell you, a lot of it is just me becoming more experienced at setting its parameters. I made a few minor upgrades, but I don't think it took too much effort to get that quality out of it.
Now you've got me curious. What's the make and model? And what are running costs like, in terms of feedstock et cetera?
I can hardly fit a Bridgeport mill in a one-bedroom apartment, after all, but a 3D printer might well be doable, and even if I still can't work in metal with one, I'd still be able to do a lot more than I easily can now...
It's a Wanhao i3 duplicator. Filament costs around $25/kg, and your mileage will vary, but I think I've only used one spool (1 kg) of filament in the time I've had it or so. $50 of filament will get you a lot of prints, as most prints are a few tens of grams.
Wow, costs have really come down since I last looked into the space a couple of years ago.
Thanks for the info! One last question - I'm seeing .1mm cited as minimum wall thickness in a few places. Legit or bogus in your experience? I'd love to be able to 3d-print complex light diffusers, but that just sounds too good to be true...
Yep, what the sibling comment said. In practice, I've gotten great results with this machine with 50μm step in the Z axis, although yes, my X and Y axes are limited to 400μm.
Also yes, optically clear prints are a no-go, as far as I know. You can get transparent PETG, but it will end up looking rather milky after printing. Here are some videos:
That's well within tolerances I could live with, I think. And optical clarity is not a requirement! I'm just thinking about complex-shaped translucent diffusers for a design of accent lamp I'm thinking about making - if it works out, I might see if I can sell them to some of the hipster shops around here, and I can see where diffusers might add some appeal.
Thanks again for all the info! I really appreciate you taking the time.
0.1 mm layer thickness is standard for an entry level machine. But in XY it is hard to create lines more narrow than the nozzle diameter. Typically 0.4mm, down to 0.2mm is available. There will also be visible lines still. And optically clear prints are ver challenging with FDM.
No idea if it would be good enough for your usecase.. Visit a makerspace or use 3dhubs to give it a try?
One other solution is that whoever it was in hollywood that first declared fanny packs to be 'uncool' needs to be 'drawn and quartered'. My little tiny remote is easy to access while i'm on my bike thanks to this amazing technology called a fanny pack.
I don't know how easy it would be, even with a fanny pack. There's going to be other stuff in there too, and even hitting the button when I'm holding the remote in my hand is hard with a glove, as the buttons are almost recessed. This design has the button protruding, making it much, much easier to press.
I would be afraid to have my garage able to be gotten into by anyone who happens to know where I live and has a screwdriver. But from a technical standpoint, I'm impressed with what you did, and as a mechanical engineer would love to have a 3D printer of my own! Nice work!
Buildings are different here, I live in an apartment complex and the garage is like a parking garage. Anyone can walk in behind someone who forgot to close the door, but there's not much they can do, since it's just parked cars and apartment doors.
For a few months, we even had the garage door stay completely open all the time, due to some power problems.
Whilst it doesn't have all the features of the usual commercial packages (Solidworks etc), SolveSpace is great, cross-platform, a single binary and my preferred open source solution for simple 2D/3D cad jobs [1]. FreeCAD [2] has also come along a lot if you prefer something more like the traditional packages, although I found some annoyances with it's dimension and constraint system when I last used it ~6 months ago due to the fact it has two different sketching modes.
1: http://solvespace.com
2: https://www.freecadweb.org/