Warts and Seams

[jimc]

that looks much better. yeah 5.5 is alot of coast. i just want to throw this out there. when the coast is pretty high like that you will need to make it alot lower when you come to a model that has small detailed features because the features might not even print. i always try to keep coast and restart as low as possible to get the job done. your setting are good now for this model and probably anything else of similar size or bigger. just keep a mental note of that so you know what the deal is if you print something tiny and it doesnt come out right..

[ednisley]

[quote="pyronaught"all dimensions exactly match the CAD model[/quote]

Makes all that aggravation worthwhile, doesn't it? [grin]

Now, you can expect results like that on a regular basis...

Well done!

[pyronaught]

that looks much better. yeah 5.5 is alot of coast. i just want to throw this out there. when the coast is pretty high like that you will need to make it alot lower when you come to a model that has small detailed features because the features might not even print. i always try to keep coast and restart as low as possible to get the job done. your setting are good now for this model and probably anything else of similar size or bigger. just keep a mental note of that so you know what the deal is if you print something tiny and it doesnt come out right..

Yeah, I just now found that out when I went to print one of those cone shaped adapters and started seeing larger and larger gaps as the velocity decreases further up the cone. What worked great for the base was not so great at the top. I guess there just is no 100% solution to the wart vs hole problem. Since it's easier to remove a wart than to fill a hole, it's probably better to err on the side of larger warts and go with lower coast settings. I wonder if an acetone vapor bath would melt them off or at least blend them to an unnoticeable level?

[pyronaught]

Makes all that aggravation worthwhile, doesn't it? [grin]

Now, you can expect results like that on a regular basis...

Well done!

I do wish I did this earlier though, since now I'm having to go back and edit my thin wall tube models to eliminate internal gaps. Like you said, it's easier to just model with wall thicknesses that are chosen to work well with the extruder width rather than screw with the settings to try and get an arbitrary wall thickness to work.

BTW, you should write a book with all that detailed information you've been posting in your blog. I'm amazed at how little information there is published about 3D printing beyond just the basic intro stuff. I mean, 3D printing is not THAT new, it's been around at least 10 years now at the consumer desktop level right? How can something so ripe with technical details exist for a decade without anything beyond "3D Printing for Dummies" being published about it? When I asked the clerk at the bookstore about this he said technical books in general are sort of a dying breed, probably due to how quickly the information gets outdated and higher losses on really thick tech books when they can't be sold. I guess everyone gets their technical information online now too. So perhaps the irony is that more people would likely read your blog than if you wrote a book with the same information.

[jimc]

really in most cases this is not too much an issue but a model with just a cylinder there is always a seam at some point. in most instances you just drop that seam on a corner somewhere and then you whole model is blob and seam free.

what i do is model a 2mm dia x 10mm cylinder. export it as stl then just keep it somewhere. when you want to locate a seam, drop the cyl on the s3d table. drag it around the bed to the location you want the seam then double click on it and it will give you its location. plug that info into you profile in s3d and your good to go.

[jimc]

it's easier to just model with wall thicknesses that are chosen to work well with the extruder width rather than screw with the settings to try and get an arbitrary wall thickness to work.

absolutely and this applies to any manufacturing. the manufacturing techniques always have to be taken into consideration when modeling a part be it a milled part, injected, 3d printed or built from wood. its all part of the gig. thin walls though are really about the hardest thing to deal with in 3d printing and your jumping right in at the beginning plus using abs instead of pla to learn with so im sure its a little frustrating. 3d printing does have quite the learning curve. we have all been there at one point

[pyronaught]

really in most cases this is not too much an issue but a model with just a cylinder there is always a seam at some point. in most instances you just drop that seam on a corner somewhere and then you whole model is blob and seam free.

what i do is model a 2mm dia x 10mm cylinder. export it as stl then just keep it somewhere. when you want to locate a seam, drop the cyl on the s3d table. drag it around the bed to the location you want the seam then double click on it and it will give you its location. plug that info into you profile in s3d and your good to go.

Someone else here mentioned that little cylinder trick, I'll have to try that. I'm surprised S3D doesn't have a 2D mode with cursor X,Y readout for getting coordinates like that.

I've been making my own tear-away support tabs for overhangs that are high-up on the model since all the support material in S3D seems to run all the way to the table and waste a lot of time being generated. The little white spots left behind after tearing off the tabs can be removed pretty easily using the heat gun trick, so I bet trimming a seam off with an exacto-blade and then blasting the scar with a heat gun would be another way of hiding this blemish.

[jimc]

if you havent found this option yet concerning the support. in the 2.2 update they now give you a dense interface layer so you can do a very low support, say 10% then make the top "x" amt of layer can be done at 60%. saves alot of time and materials and gives a better underside surface. i have asked in the past for a top orthographic viev like we have in rhino for accurately moving parts around the table. nothing available yet. we will see what happens in the future. if more people ask for it it might help.

[chad]

Hi guys,

This has been an interesting topic. I also have struggled with blobs at times. Aside from all of the great analysis already discussed I found that doubling my retraction speed from 2000 to 4000 helped me quite a bit. Just one other value that you might want to try tweaking and see if it helps or hinders...

chad

[pyronaught]

That outside to inside setting seems to have a pretty radical effect on increasing smoothness. I'm going to have to start using a separate process layer for threaded segments just so I can only use inside to outside on threads and then revert back to outside to inside for everything else. Blobs get worse on inside to outside too.