Are the 3D Printed parts a bottleneck?

#1

Hey @chrismakesstuff I was just wondering if the 3D printed parts are the principle bottleneck to scaling out shipments? I saw Greg posted the other day on Facebook that you were up to 40 printers now. I totally get why it’s a great way to get the first few dozen machines out and you can quickly make any necessary adjustments to the designs etc as the feedback rolls in from the lucky first shipment recipients.

But once you’re 100% confident the parts are all dialed in, wouldn’t it make sense to scale up the 3D printing work by using someone like Xometry who are front ending hundreds of shops and also have large scale 3D printing capability? Detailed photos of an SLS print here: Xometry review - Professional 3D Prints for Hobbyists?

I didn’t even know about these guys a few months ago but I’ve done a few “test” projects with them to see how it goes, and they’ve exceeded my expectations. Spin 2 of the battery adapter arrives with the designer tomorrow (we got a few measurements off in our design, the production was dead on what we asked for). I suspect they could turbo charge your 3D printing capabilities and also they have hundreds of shops they have vetted that they sub to, so you can parallelize massively and only have to manage one supplier. Importantly, you could likely move to SLS as the material which will give a very high end and rugged finish as well.

I’m 99% sure my new GRR-Ripper safety push block is SLS printed and it has a lovely finish, if you happen to have one of those around. Heck sign up via the link in my other post and take advantage of that $50 credit on your first order and order up some of your harder parts to get a sense of the quality and compare. Running all those printers must be a pain in the butt! I bet you can get a couple samples of your own parts at virtually no cost…

-Jeff

#2

I think cost would be the key factor. Print services are not cheap in any way. Ya maybe it would speed up things but they would be looking at 5x the cost than printing their own.

#3

@chrismakesstuff I almost forgot. It won’t take much time to check out Xometry and get a quote. Somehow they can quote instantly right from the STL files. It’s basically <90 seconds to upload and have them process it and then you can play “pix and mix” with materials and quantities and timelines and get instant pricing. Very low barrier to getting the business info you need to compare with other options.

I don’t know how they do it, but it’s like Amazon level ease of ordering for manufacturing. I’ve never seen anything quite like it in the B2B space. As you can tell, I’m a fan after spending a month and my own dollars trying it out :slight_smile: Anyway, enough posts from me, time to go make some sawdust and noise and finish the drawer jig I started…

-Jeff

#4

I would have thought so too, but that hasn’t been my experience. Perhaps the particular part I was making was better suited, I don’t know, but it’s so easy to get quotes and adjust quantities and finishes it’s worth checking. Maybe it doesn’t suit these particular parts, but I have a hunch it does suit some or all of them.

Are the STL files around anywhere? I know the shapes are visible in OnShape but can I download the STLs? Seems to me it wouldn’t let me when I tried. I wouldn’t mind running a few quotes just to get an idea of cost myself.

-Jeff

PS - I did get some local quotes on that battery adapter in the project link and the pricing was right around $50 for a one-off PLA print on a low end printer, which doesn’t compare at all to the finish and ridgidity of the SLS prints. Mind you this was “some guy on Craigslist” but I was intrigued that his quote was basically at the same price point but for a lower grade material at quantity one. Also care and feeding of a farm of 3D printers isn’t free on many fronts…

#5

Our printing is running quite reliably right now and we’ve set up a fairly efficient process which I’m happy with. We’ve certainly never had to produce at this scale before so there was a bit of a learning curve, but there’s also something to be said about changing too much at once between our beta-tested Kickstarter machine and the one we’ll be shipping out to all our backers.

Our focus right now is on getting through the backlog of orders, so I think making more changes right now wouldn’t be a great idea. Certainly we’re already working to implement new manufacturing and assembly techniques on the next batch of machines which can make the process of order fulfillment and production more scalable.

To directly answer your original question, 3D printing isn’t currently the primary bottleneck. There’s a combination of trained assembly and quality assurance techniques that need to be brought up to speed in order to meet fulfillment requirements. This can be attributed to the boards coming in late and requiring modification, and to the recent changes in the dust shoe design which we felt were necessary to implement.

2 Likes
#6

Hi Chris,

As a fellow Canadian manufacturing/tech entrepreneur, I totally understand where you are coming from. There are a lot of moving parts in getting a physical product designed and evolved to production before even considering supply chain and manufacturing processes and QA etc.

It may not seem necessary or valuable at the moment to look at options, but supply chain management is dynamic. It’s great that you’ve got the 3D printing farm doing what you need right now but that’s a lot a points of potential failure as well as care and feeding. Consistency over large volumes and variations in material supply quality as well as wear on the units or operator error could bite as you scale, which can cascade back in to unexpected costs and delays.

As my first boss once said to me “Always remember, your B-plan is more important than your A plan”.

As a fellow entrepreneur, I’m going to give you a second little nudge to invest the 30 minutes or less it would take to grab your STL files, set up a Xometry account and have their automatic quoting engine give you pricing and lead times. Even if you don’t order a sample with the US$50 credit if you sign up here https://mbsy.co/BWNW9 it would be worthwhile knowing in the back of your head how you could scale out (or not) with Xometry or someone like them.

It is also worth understanding what, if any, positive impact it could have on your margins. Afterall, hopefully some of that margin goes in to designing the next generation Longmill when the time comes. And given the leap from the Mill One to the Longmill, I know we’re all excited to see what you dream up over the next year or two as you learn from the Longmill deployments.

As an added benefit, SLS has strength similar to injection molded parts. I was a bit alarmed when I read this in the first assembly post on the forum: " This one was very likely my fault as I knocked the left Y-Axis rail onto the floor and cracked the left rear foot. I patched it up with some epoxy as I was just a crack along layer lines." Of course, any part could crack, but you would be much less likely to see that on an SLS printed part. Given the stresses the machine could be under during heavy load, increased strength in the printed parts is a good thing…

You’ve got the files, account setup is <90 seconds and then it’s just uploading and selecting your printing options and your quotes are in real time as you adjust options so you get a real time sense of the impact of quantity and finish adjustments on unit cost. Purely from a business “what-if” scenario basis, it’s information worth having for the <30 minute investment.

-Jeff

#7

I downloaded the mid rail support file from their open source page.

I tossed it up and played a little with different materials and methods. Most
were coming out around $60/ea USD.

Maybe there are better options (materials ect)
that could be selected???

I’m all for upgrading down the line maybe for certain parts but ya as I expected
not cheap from a print farm. An equivalent part in PLA $2/ea CAD

LongMill - Open Source

#8

That’s quite a difference, for sure. How did you arrive at $2/ea as a cost?

-Jeff

#9

The best thing about this project is the Open Source part of it and if I wasn’t one of the first people with the machine I wouldn’t have thought twice about cracking the first part. I think many other solutions may have caused more issues as the part was the only issue with a fall from 30". The aluminum rail did not bend, as was my first worry, with it only being supported on one side. But if the epoxy didn’t work my plan would be just to print another one, no shipping or anything just print overnight and have it in the morning.

Now you bring up the 3rd party printing services, I have recommended them to a number of people as I have seen them do some very neat things. But I had never thought of using them as you describe for manufacturing and it makes allot of sense. I think that could also be one of the neat things from this type of project too where different people experiment with different solutions for a better overall product.

Lastly you mention SLS and that is one printing technology I have been very exited about. My initial exposure was on a video from 2017 by tested.com (https://www.tested.com/tech/3d-printing/634216-everything-you-need-know-about-formlabs-fuse-1-sls-3d-printer/) & (https://www.youtube.com/watch?v=IrFYt3uUdrI). In the video they printed a nylon bicycle peddle that had been used daily for a year and it looked the same as the ones coming off the printer. With this printer you can fill the entire build volume in X,Y & Z with parts with no support or adhesion to the build plate with FDM or SLA. I know there are other SLS systems as well to fabricate in metal and other materials.

The link for the part I printed is - https://cad.onshape.com/documents/c671532d980f277a13d1ace4/w/95c27aef1c0bb10a840ac8bd/e/3e6995fc4bcf06b277ac8177 - right click on the highlighted tab at the bottom and choose Export…
I chose, and the rest defaults.
Format - STL
Units - Milimeter

#10

Thanks for the link to the LongMill Onshape page. I didn’t have that one yet, but now if i need to I can reprint the one foot I cracked.

#11

I work with 3D printers all day… Something like that would be very inexpensive to print.