POM Filament printing
Ho Folks.
I recently had reason to open a roll of POM Filament when a customer asked for a mart to be printed using that material. The part to be printed slides along a shiny rail, and its part of a sailboat winch.
I performed a routine calibration of extrusion multiplier, a 40mm cube in vase mode.
The cube printed onto paper that is glued to a smooth sheet on the mk4 printer. Paper used by recommendation from the supplier of the filament.
Results ? The first layer of the cube started to peel off the paper whilst the first layer was still being printed. See photo for illustration.
I tried again in successive attempts using Magigoo applied to the paper, first General purpose Magigoo, then PC Magogoo, and finally a locally made ABS type glue. All of them exhibit the same issues. In addition to the first layer coming unstuck, higher layers crack and the sides of the model contort in wierd shapes.
This is all due to the high coefficient of linear expansion of the POM material.
I did some further research and found this issue mentioned on Forums (ie Reddit) as far back as 2015, so its not exactly new.
But that was 8 years ago, I'd think ongoing research and development would have seen a better behaved POM material hit the market by now.
I found some high end POM material with embedded Glass Fibre and Carbon Fibre, but it was in pellet form, for injection moulding etc, not 3D FDM printing. And its extremely expensive at over Euro 300 per Kg.
So, my post here is a last ditch effort to find a solution allowing me to meet client expectations to make the required part from POM plastic.
Solutions sought:
a)) a way to negate the prevalent expansion and contraction leading to warping, cracking and contortion.
b) a reasonably priced POM material with altered characteristics over the classic Delrin/Acetal material, that has better dimensional stability, whilst retaining the low friction coefficient expected. (I note that POM + GF and POM + CF do not meet that requirement. Uncertain whether POM + PTFE could serve or not.)
c) an alternative material that has very low friction coefficient like POM, without the tendency to contract so much upon cooling. (Or to put it another way, it's more easily printable to get good results).
RE: POM Filament printing
I guess there is a reason why it is incredibly hard to find POM filaments for FDM printers. It is a highly crystalline thermoplast and those are very difficult to print. Poor layer adhesion points either towards a hotend temperature that is too low or the ambient temperature being too low.
I don't know what printing conditions your filament producer suggests but I would guess that POM needs a very hot print enclosure for improving layer adhesion and at least moderate the crazy warping. With ABS/PC for example I have seen recommendations of printing at 70°C enclosure temperature. For that you need of course a printer capable of that.
I could be wrong of course, with semi-crystalline materials rapid cooling could be better than slow homogenous cooling as it can alter the crystallinity but I would certainly try it hot first, otherwise layer adhesion will be catastrophic.
CF or GF filled filaments would help against warping, but they would make poor layer adhesion even worse.
If you need POM for toughness, sheer resistance and self lubrication, I guess you are better served with CNCing the part, if possible.
If PCTG is an option though you could try that instead, easy to print, super high layer adhesion, suitable for harsh outdoor environment and also fairly tough, low abrasion and low friction. If the moderate thermostability isn't an issue.
Mk3s MMU2s, Voron 0.1, Voron 2.4
RE: POM Filament printing
Thanks for your response.
Regarding temps it’s a little confusing.
Some sources say to print between 210 and 230 partly because the material starts to break down at 235 and one of the breakdown products is formaldehyde. Others say use a higher temp range up to 240. i’m using 220.
I have today tried printing on paper tape stuck down with glue, and it comes unstuck just as it did earlier.
The wsy this material prints is really shockingly bad. i found forum posts from 2015 citing the same issues. So it surprises me nobody has devised s copolymer formula to fix its strong thermal expansion characteristic while retaining its low friction coefficient.
My use-case relies on the low friction.
Thus GF AND CF variants will not cut the mustard as they degrade friction.
i am presently looking at a pc-ptfe material said to be as good as pom’s low friction.
cheers
POM is horrible to print. Have you tried printing slowly, really slowly? The drawback is that the hotend dwell time is increased and so is the risk of thermal breakdown.
If you MUST use POM then I'm with @thejiral, it machines reasonably well so, if the shape is suitable, you may be able to get your file converted for CNC processing.
In similar maritime applications I have had some success with Bio-Pro from 3D Tomorrow.
Some top-end racing skippers are happy to regard even fixed gear as consumable, if your customer is one then nylon might be an acceptable alternative but it probably will not last and you may be reprinting a couple of times a season.
Cheerio,
RE: POM Filament printing
Thanks for responding.
Yes, I have tried to print with slow speed. Currently set at 30mm/s on all speed parameters, and 10 mm/s for the first layer.
The first layer goes down ok, and levels above seem ok up to about layer 4 then I see the corners start to curl up. and by layer 10 its like a potato chip, just stuck to the paper in the middle.
I am minded to keep going and see if it will split and buckle at higher levels, but generally I find the model starts to move around and come loose, at which point I stop the print.
There are the 3 sins of this material:
1. will not stick to the bed even when the bed is covered in paper stuck to the bed using Korex glue applied when the paper joint tape is soaked in water then allowed to dry before printing. This technique espoused by a YouTuber and demonstrated. Just not working for me ;(
2. if it gets to higher level then I see splits occurring on diagonally opposite sides,
3. distortion of flat surfaces turning them into Picasso style rendition of a cube.
This was all on a extrusion multiplier test cubd model having thin vase mode walls. I decided I might be asking too much, and instead printed the xyz dimensional test model (25% infill, 2 perimeters) upsized to 40mm.
That one prints near perfect up to about layer 4 and then the corners start to curl up, potato chip fashion.
which is why I am inclined to go for one of the alternative low friction materials. PC-PTFE is in the frame as its not over-the-top expensive.
I have vast experience with Prusa’s PC-CF, which is s terrific material (not suited to this project) so PC-PTFE isn’t a big stretch, and its print parameters within the range of my MK4’s