The latest Preform is great, especially with these new supports that can arch over certain features. However, there is no way to control this useful feature. I am encountering a problem where there is one surface I absolutely cannot have any supports attached to. It’s just too delicate, and given that the supports freaking EXPLODE when snipped, it’s not feasible to have internal supports grow from this surface. There are plenty of other adjacent features for internal supports to grow from. It took me an hour to find an orientation that maximizes accuracy at critical areas but doesn’t place supports on the most delicate feature on my model.
It seems like a no-brainer to just have a “paintbrush” tool to exclude surfaces from having internal supports grow from them.
It just seems bizarre to me that you can control where the supports attach on one end but not the other, and you can’t tell PreForm where a conjoined arch would be really helpful rather than a straight vertical support.
Pennywhistles require 5 thou accuracy in critical areas, so honestly most of my time has been spent troubleshooting orientation rather than the design itself.
I’ve raised this as an issue previously too, it drives me nuts as well. The anchor points are also bigger than the touch points themselves.
I’d recommend submitting a feature request support ticket. Forums are a good place to chat with other users but not great for asking for feature requests from my experience.
For things like this, something I’ve done a couple of times, always with success, is just to model my own supports in the CAD model, and model my own raft. Then you can just print it as-is and not use PreForm’s supports.
I suppose theoretically, if you don’t want to do all of that, you could model just your internal supports, then turn off internal supports in PreForm so it can generate the rest.
There’s also third-party tools that can generate supports, I’m not too familiar with what’s out there because I’ve always been more comfortable modelling my own when needed, but one might have a way to do this.
Currently, when you generate supports, PreForm mercilessly applies them to the model. While you can edit support points, it’s a problem for internal supports because you can only specify the top attachment point. The bottom ends up wherever.
The other issue with internal supports is, for complex models, you almost always have to do manual edits to remove certain internal supports, and it can be tedious, which is a shame because PreForm’s automatic generation algorithm is otherwise pretty solid. It’s just not always cooperative.
So I have two suggestions:
The same suggestion from the forum post and that I’ve made in the past: A tool that allows you to paint areas on a surface that should not have touch points on them (including the bottoms of internal supports!). Then you can just paint the surface, run the algorithm, and this is great for workflow because it can greatly raise the success rate of the algorithm and lower the need to do manual edits. There are many situations when this would’ve eliminated a lot of work for me; the majority in fact.
As an additional request: In the support editor, when editing internal supports, it would be great to be able to position both ends. I have a lot of ideas for the GUI for this that I’d be happy to share if you’d like, but the engineers at Formlabs are pretty good with UI stuff too so I’m sure they can think of good ideas, so I’ll spare the details there.
Every time this comes up it seems to get a lot of user support, and I think it’s really worth considering.
BLol! Formlabs is hardly “big print” though - they make high-precision SLA available at a “pro-sumer” level. I think it’s just a UX oversight. Be nice to the Devs, I’m sure they’re aware of it, they just need management to approve it!
More often than not I’ve been really impressed with the solutions preform comes up with to supporting different geometries, unlike a lot of machine learning tools which leave me smacking my forehead.
Give Autodesk hell though, they’re really a bunch of bastards. No dual monitor mode after EIGHT YEARS?
I’ve actually sent some feedback personally about this as well, so hopefully this will either become a feature, or … we’ll have an answer why it’s not implemented. I fully agree it’s frustrating …
Thanks for submitting requests to keep this function top of mind! I think it would add a lot of value to what PreForm can do. I’m still troubleshooting orientation. All the orientations that don’t either put supports on the airblade or grow supports from the airblade, jeopardize the windway exit face, which is the OTHER critical surface that must be as accurate and unmarred as possible. I’m having a really hard time telling whether my new design is flawed, or whether another five thou septum is appearing on the windway ceiling from the first layer or two of that surface. I am SO CLOSE to production on these and it’s driving me nuts.
Fwiw, I just took a look at some of the software out there. B9Creator lets you edit both sides of internal supports. I found it to be intuitive but clunky, though, and the automatic generation isn’t quite as smooth as PreForm.
But, if you check it out; to control an internal support I found that you first place the top point and it’ll generate one straight down, but then you click “Modify” up in the tool bar and you can select the support and drag both ends around. It lets you customize the shape of both ends, too. You can then export the whole thing as an STL.
You might be able to get away with just doing your internal support in B9Creator then using PreForm to generate better supports and a nicer raft for the rest.
Great find, thank you!!! I’ll have to look into that for future applications, I’m sure it will come in handy! Ultimately I think the better plan, though, is to outset the airblade with a sacrificial bar and then shape it with needle files. I used to do something similar but formally modelling something to protect the end of the blade is probably better, with an indicating ridge to show where to stop filing. After examining some older prints I think having the windway exit begin on some initial layers after supports is just a bad plan; the areas where the greatest precision is necessary always benefit from being the last layers on a surface, as they have a solid structure to build upon. The airblade at least I can SEE and access relatively easily, whereas squinting down a slightly tapered 0.03" high slot to see if there is a deviation near the end is just painful.
I used to do these in two parts, allowing both surfaces to point “up” (aka toward the pool of resin), but found the indexing to be yet another source of error. There’s a reason these instruments are typically made using stock tubing sizes and clever milling tricks rather than going for the acoustic ideal - they’re just hard to make, and the more you can outsource the precision to the suppliers of your raw materials, the better.
For the devs, though, here’s a perfect example of what I’m talking about:
Figure 2 here is unacceptable - this instrument would be unplayable if the supports were clipped flush, and having to scratch into the windway to adjust the bit left over is a real workflow problem. ANY pitting caused by support removal kills the instrument permanently.
If Formlabs wants to be taken seriously in the industrial/engineering space, the ability to nudge those supports off of key features is an absolute must.
I would try printing this the other way up so that the air blade is self-supporting. I would also put a fillet or chamfer on the internal inner edge to try and get that self-supporting:
If a fillet is not possible a row of fine, 0.2mm, touchpoints on the bore just off the edge will work, though the edge itself will be slightly scalloped.
I mentioned earlier that I’d tried that orientation, but the issue was that the windway exit always had deformation at the first few layers no matter how well supported it was; it’s just a fact of printing a big face like that in that orientation. Filing back the windway face, or the windway itself, is another source of error because it can change the angle of the air jet.
I’m not saying it can’t be done - in fact I’ve done it many times; it’s just that the success rate isn’t great. I’ve also found it plays with much more ‘punch’ when that face has a crisp 90 degree edge both at the aperture and the inset your arrow is pointing to. You don’t want to bevel that unless the bore of the instrument is conical; as this design is inspired by vintage instruments, my bore is cylindrical.
I appreciate folks are trying to help with support placement, but let’s save that for another thread aimed at fellow woodwind experts, if any are on here. I can’t condense five years of experimentation and research into digestible criteria here, and it’s unproductive to be brainstorming solutions without an understanding of the acoustic ramifications of different orientation strategies.
Since this is straying off topic, but helpfully so, here’s a link to a discussion I created about orientation and supports, with more details about the project and images of what I’ve tried so far to get better consistency in the final products: Final product orientation & workflow
what about slicing in Chitubox or Lychee since those have full tree support manipulation then just export out as an stl? From my understating the only thing you loose are supports are cured differently so they don’t break away as easily. I typically use a craft knife to remove supports anyway
stickin’ it to Big Print.)
