Impressive testing. I asked support for similar data a few months ago, and unfortunately was never able to get a straight answer over how to make tolerances for things like this. I was essentially told to allow for “0-1mm” of tolerance and experiment on my own. Baffling as to how Formlabs do not have published guidelines for things like this and support have essentially no idea.
I have done some testing like this before, a big difference being cleaning the interior of the threads with a q-tip to ensure residue resin does not jam screws. Also, I do think threads as fine as these benefit from printing in the 50 micro layer height for more accuracy.
This guideline is pretty good however and for the most part I have just settled on using metal inserts or tapping the holes for more reliability…
I will try on 50micron but the likelihood of printing entire larger parts like this in not high, hence the test at 100microns.
I have been using threaded inserts with FDM, but have been struggling to make them work with Thermoset plastics. I bought specials ones for this purpose, but I find that when I am finally able to press them in without shattering the part, they are not very reliable. I am open to any suggestion here
I have not yet received any feedback from support - Case #01036665:. I did upload the printer logs via the printer, just wanted to make sure I did everything correct?
Your mileage with printed threads will vary wildly with size of thread and part orientation. In general I’ve found it best to just chase the thread out with a tap afterwards.
I guess yes… The idea was to used these for items that are ‘non-serviced’ really e.g I have an enclosure print coming up that has 6 x 40mm case fans (so 24 screws) and 8 more for closing it. There will be 20 of these enclosures so I was hoping to make life easier for myself. As these are mains to DC converters, the client has no business opening and closing it With the amount of trouble I have been having inserting ‘screw inserts’ I thought it might be worth trying to print the threads instead.
I am using these the SPIRO insert supposedly design for Thermoset plastics but doubt there will be good results on this inclosure due the location of the holes. If I get the inserts to go in with a maximum pressure that I can apply by hand (anything but using a press) then I find the inserts just start spinning on place if just a little too much torque is applied when tightening the screws. If I design the holes anywhere close (even with ±0.2mm offset) to what the datasheet specify the hole should be, there is no way of getting there insert in. If I then use a small vice or something to apply pressure (as I do not have a proper press) I find that the part shatters (even with ±5mm wall thickness)
Here it the latest test model I did with which my little Arrowmax screwdriver seems most happy with.
To be fair, I guess this is type of orientation Formlabs do suggest, but this will more often than not, limit me to one part per print as apposed to 4-5.
This has turned into a pretty detailed case with a lot of useful information I was never able to get from Formlabs support. I will keep these accuracy numbers in mind going forward, just a little wild to me that Formlabs do not have this as guideline ready to go and you have to spend your own time and money testing… Maybe it’s just I’ve given up on any testing due to material cost, but I mostly think I usually just instruct clients to print like an undersized hole and I will manually tap the parts. I have had moderate success with that with Tough materials as well. I have gotten some larger self-tapping inserts to work, but the heat insert ones do not really mesh well with these materials.
Now, bigger test is can these be replicated with a flat, accuracy, and straight part with your housing model? I have no doubt Form 4 can print these little things that are meant to test and produce reasonable accuracy, but reality and real parts are obviously not these. Very curious to see if you experience warping with General Purpose V5 like grey.
This is problematic yes, more so here as it is quite a bit more costly than on the Formlabs Store.
You are correct, heat press inserts will not work as the Thermoset plastics cannot be ‘melted’… instead it will resist heat and all of a sudden become very brittle and crumble like a biscuit. The inserts I ought is specifically design for Thermoset plastics but they need to pressed in with quite a but of pressure. If you have a more complex than flat geometry, this will be problematic at best.
This is what I am hoping to try and figure out with Formlabs support’s help in the hopes of not burring through liters of resin with no or limited success, but I have no heard back from them since I contacted them.
Well, I printed 20 enclosures with Grey V5 and had no warping. Holes were not accurate though, but I could manually screw in the screws and they seem to be holding up.
These would most likely work for any of the materials without having the need to heat anything up. If you’re worried about them backing out for whatever reason, you could also apply a very thin film of glue to the outside of the insert so that it ensures it won’t go anywhere.
Just to be clear, the ones I bought are specifically designed for Thermoset plastics, so they are press-fit and no heat is required. The problem I have been facing is that if I design he hole the diameter recommended by the manufacturer, there is no way of getting the insert in without significant pressure which ends up cracking the part.
If I design with tolerance large enough so I can press-fit them with reasonable effort, the slightest over tightening result in the insert to break free and spin in one place. I have done a test where a pattern is design in to help, but with limited success. See below:
I am sure there should be a way to minimize the “printing not quite straight” on flat surfaces and square corners. I would be happy to run battery of tests on different designs to help and find a way to make these work (as far as possible) if the materials weren’t this pricey. Unfortunately it will be quite an expensive exercise to attempt on my own.
I have done this basically extensively already, I unfortunately ruled out most of the ways which this can be improved by the user. I have tried reorienting, different manual supports, everything. Unfortunately, I do think I have to conclude these are current material and settings limitations of the printer.
Again, I have given up on printing general purpose v5 materials due to warping and stopped printing engineering materials due to warping and cure failures.
Like I said before the only material I got a semblance of success out of is precision model resin, so I shall continue printing that for now. This is with latest Preform version 3.39.1, no fixes yet been issued to improve print performance on v5 and engineering materials since launch.
I had to print some parts in Black today, so I used standard V5 BLACK resin. I initially when it came off the build platform eventing look quite nice but after I removed the supports I could clearly see some layer lines that are quite prominent. Is this to be expected? Would it help to print on 50µ ?
I then tried some sanding with 1000 grit paper and applied some oil as suggested in many forums, but this removes the mat black finish.
Those layer lines would not have been normal on the laser models, but who knows with this version. Seems like Formlabs is using their customers as beta testers.
I guess I could orient it differently, but my experience so far tells me that I’d rather not print part flat surfaces parallel to the build platform if they are not on the build platform.
At this stage I feel like you “need” to media blast these parts. Sanding by hand does not work out great unless you plan on coating/panting the parts. Of course it would be just me doing something wrong…
The problem when I try to “hide” the support marks on the inside of these parts, is that those are also the larger flat areas.
I have learned my lesson in trying to print then horizontally (the surface is very wavy with lots of support marks). If I print in the orientation I did, then unfortunately the visible area has support marks and need some sanding. I use 0.45 as the touch point size. I have seriously considered media basting but the reason I bought FL was that I was under the impression I would not have to spend significant time post processing
I feel like I had better results printing with grey… maybe I will print these parts in grey later just to check. Just hate wasting resin.
