I am planning to print some large structural components, and I intend to form them by glueing together a number of printed parts, and skinning those with composites. In other words this plastic needs to bond well to epoxy.

What is the optimal material here? My logic makes me arrive at fiber-reinforced PC as the only good choice. But please point out to me if my logic goes wrong somewhere.

PETG/PP/PA/ABS/ASA are all out of the window. I want this to bond well using run off the mill laminating epoxy. And by 'bond well' I dont mean 'cant pull it off with the first try with my fingers', but I mean I want the glue bond to be about as strong as the underlying material, and that just isnt going to happen with any of the above. I know that for every plastic there is some $100 a tube toxic goop you can get to stick with a few dozen steps of surface prep; but im not going to use that as a laminating resin.

Moreover, I want a fiber reinforced material. Not for the strength per se, but for minimum warp. When printing multiple 30x30cm pieces, I want them to slide together without gaps (or the gaps I designed into it); and if the piece as designed was intended to be straight after assembly, i want to to be straight after assembly. Its frustratingly hard to find quantitative information on warping properties. But polycarbonate without fiber reinforcement does not fit the bill, thats for sure.

Plain PLA might be another decent option. I cant find much quantitative information about bonding it to epoxy, but at least its not a known no-go. There is some worry about the heat of curing the composites deforming the PLA though.

PVB might be another good bet. Easy to print with, and great at bonding. Might be sensitive to solvents in your laminating resin though.

Of course all other criteria for a nice filament apply as well; consistent extrusion, layer adhesion, all that.

Is CF-PC the way to go? I see very little in the way of reviews about CF-PC; it does not seem to be that popular a combination, which is a bad sign, but I cannot discern much reason for it.

  • $\begingroup$ Just an idea, but if the surface was formed of trapeziodal slots (or some other undercut profile) then the epoxy would be physically connected to it instead of only chemically. I guess you would need rather runny epoxy. $\endgroup$ Sep 11 '20 at 19:02
  • $\begingroup$ Yeah; I've only recently gotten into fiber reinforced filaments and I noticed their surface microstructure trends to be quite rough. Indeed you might take that even further by modifying the surface on the gcode level. It might suffice; but it sounds like a lot of testing before you know how it plays out in practice; and I'd rather spend money on a solution that should 'just work'. But it's an interesting avenue for sure. $\endgroup$ Sep 11 '20 at 19:07
  • $\begingroup$ [* sp: trapezoidal, of course.] $\endgroup$ Sep 11 '20 at 19:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.