geekhack Community > Other Geeky Stuff

3D printing

<< < (2/2)

Metum7:
Thanks for your replies.

I realized maybe I didn't specify enough what I wanted to print as there were a lot of things in the answers regarding that. I want print the two sides of a split keyboard case for a handwire(specifically the dactyl manuform), and the design(although I modified it a tiny bit, but not in any major ways) is reliable and tested. I do have a little experience with 3d printing from engineering courses in my high school, and I was planning on doing more research on the subject, but yeah. Anyway, most of the on demand services I've seen have priced each half of the board as around 35 bucks, but I don't know whether it would be cheaper to try and get it printed somewhere else(the main 2 options would be the college my dad works at and my school, although my teacher from the latter said they usually don't do prints or other manufacturing things for students' personal projects so it would probably be the former). I would have to get my Dad to ask his colleagues about that one, and it would be quite a hassle from what I've seen, so it would be easier if I could just get it printed from some sort of on demand service. I don't really know. Also, although I may get a printer in the future, this is my first time actually using a 3d printer for my own personal purposes, so I didn't think it would be a good idea to jump the gun right now.

Sorry if this post is a little incoherent, my mind was kind of all over the place when writing this

Thanks again

Leslieann:

--- Quote from: MIGHTY CHICKEN on Thu, 17 November 2022, 21:11:05 ---Also apologies if I passed off as rude in any way earlier, the question was directed towards playbox who seemed to be taking words off of ya. I wouldn't assume there's no value in a service or no reason to get a 3d print without a printer.

--- End quote ---
No, you're good.  :)
You weren't wrong, just that it depends and at the time we had no idea what he wanted.  There's a lot of variables in keyboards and 3d printing.

I wouldn't say there's no value in it but it is a catch 22.
To get the most out of it you need experience but if you have experience you probably don't need the service.




--- Quote from: Metum7 on Thu, 17 November 2022, 21:36:49 ---(although I modified it a tiny bit, but not in any major ways)

--- End quote ---
Famous last words...
It's easy to mess something up in digital that used to work fine, and "tiny bit" is relative. What may seem minor can lead to major problems, I can't tell you how many minor changes I've made led to massive redesigns or conflicts.


Make sure the on demand service is pricing with the infill and perimeters you need, when I did work for them they always without fail, under estimated the price (by a ton). Also layer height matters a lot, smaller layers feel better but crank up print times.

The difference? (based on a 5x6 Dactyl design on Thingiverse)
1 perimeter, 10% infill, 0.2mm layer height = 300g plastic and about 12 hours print time.
vs
3 perimeter, 40% infill, variable layer height or 0.12mm height = 350g of plastic and about 30 hours of print time.

12 hours print time (plus prep and cleanup), sure, I'd do it for $70 but for 30 hours, probably not and if you're on a low end home based printer that's probably going to be closer to 45 hours and that's quite a long print time for those machines.


3d printing is only cheap if you own the machine and even that's not a hard rule.
Worse still, most of these designs are by people with 3d printers meant for people with 3d printers and some are done by people with no 3d printing experience at all. And truth be told, even when they "work" they're often not great or as simple as they seem. Remember, it's a bunch of amateurs and they often work with whatever they have, a bunch of random parts and cheap digital calipers. No hate, just saying don't trust everything on these sites. I've seen designs where they used 5 different screws because they only had 2 of each size, designs by people who just imported some random cad file to be "helpful" and others were just so badly proportioned they would never work in reality which is why they designed it, posted it and never showed off a finished model, they never got it to work. I trust almost nothing I download that is size specific regardless of how many others printed it and run as many tests as I can prior setting the whole thing to print because I've been burned too many times.

Metum7:
Ok, so long story short, I should make sure that the designs have finished products associated w/ them, make sure my changes are not going to break the design, but after I've done all that, and done some more research on 3d printing in general, do you still think getting it printed from some other service is viable? The only things I really care about are cost and at least decent quality, but I guess that may change when I do more research.

Leslieann:
Making sure it's a finished product and making sure you didn't break the design both involve actually printing it.

You can look at it all day long in CAD and miss the obvious, until you have it in your hand you may never notice a hole is missing or something is off by 0.75mm, skewed, or the design has a CAD error making it unprintable without massive errors. Then you find and fix the errors and repeat the cycle. There's no way around this and it adds up fast, this is why most people who use a service end up buying a printer. You say it's only $70 but after 1 revision you're already 1/2 to 3/4 of the way towards an Ender that lets you prototype for pennies.

I know it sounds like I'm being overly critical, I probably am, but you're also concerned about cost and I'm basically telling you it's probably not going to be as cheap or nice as you think. 3d printing is messy and it's even more messy when you rely on multiple 3rd parties to have everything right. In this case you have the original designer, your changes, the service, and the printer, that's four chances for something to go wrong on something that requires decent precision on machines known for a lack of precision. Again these designs are by people with 3d printers for people with 3d printers, their tolerance for what passes for working or good enough may not be the same as your level of working or good enough. What passes for good enough tends to trend down as you make more and more revisions to get something working whereas you may be expecting something relatively close to something you could buy and slap together in minutes.

In the end you're going to have to pay your money and cross your fingers.
The more research you do the more likely it will come out how you want, but you're still very much beholden to these 3rd parties to get it right.

Metum7:
First of all thanks. That cleared it up a quite a bit. I guess I have a choice to make, but your guys' answers made it much easier to figure out what the choices actually are. Now I'm off to sit and think about it for a while.


Thanks again for your help

Navigation

[0] Message Index

[*] Previous page