Jacob, Welcome to the boards. I noticed on the tech specs that the foaming rate of the LW PLA is dependent on the printing temp. Do you mind sharing the temp that you were using? have you done any testing to see if higher temps (more foaming less density) creates a cleaner casting? Love the work so far! Bruce
In addition to the CBB's Q, I have a small wish list of Qs too. The statuary look great, but do you have a few close ups of either the print or casting. Just trying to get an idea of the print resolution. Along with that, what is the line width/height and what was the print time for say the wolf? Next time you cast, it would be great to see a video of the pour. You might be surprised in how much info can be gleaned from watching a pour. Since your patterns are hollow, so are your molds, and gating methodology will follow the conventional AFS or naturally pressurized methodology. For these, metal velocities and fill rates are dictated by gravity and the feed system geometry/cross section. For solid foam patterns, I can tell you that is not at all true. The metal velocities and fill times are dictated by the rate at which the foam is evaporate, which in my case is about 3x-4x slower than typical open cavity casting. Another big difference is how the evaporated gases escape. Since your patterns are hollow, they can be vented. Solid patterns cannot be vented. With solid patterns the gas escapes as they are created at the boudary layer of molten metal, liquid polymer, gaseous polymer. By comparison this is a very small area to vent through the refractory coating as opposed to the entire mold cavity and is the big reason why metal speeds and fill times are slow for solid foam patterns. One thing that might work to your favor in scaling the method, since your patterns are essentially skins, as part size increases, the amount of polymer in your pattern will increase ^2 with the pattern area while the volume will increase ^3 with pattern volume. That means your total polymer density could actually improve with size. But, the structural integrity of the pattern and ability to withstand molding pressure created by packed sand likely will not and may require some additional skin thickness or very coarse infill to preserve the shape and dimensional accuracy of the pattern. Best, Kelly
Jacob: Nice to hear from you over here on this forum. I also have a couple of questions, mostly about your Husky pattern. 1) The post under the tail, was that support generated by the slicer - or did you add that to the model? 2) If it was support, did you drill through the base and the tail to allow the tail to actually vent the tail? Don
I've been printing at a temperature of 235 C. To compensate for the foaming, you have to reduce the flow rate through the extruder. I'm using a flow rate of 52%. I do need to play with the printing temp some more, I think there's room for improvement. Here are some more pictures, the line height is 0.2mm, and line width is 0.4mm. Including the gating, the wolf was about 5 hours of print time. Lightweight PLA needs to print substantially slower than standard PLA. Specifically, I use 30 mm/s. I'll be sure to get a video next time! The post under the tail was added in CAD before slicing the model, so it printed hollow.
The print lines are very fine. That casting surface could be easily finished/polished if so desired. I'm quite impressed. You are the first in more than 20 to show a resulting casting anywhere near this quality from this method. -Well done. The print time desn't seem too bad.......I'd be afraid to hazard a guess on one of my typical parts......this one is approximately 24" x 12" x 5". It's about 4 hours of CNC Router run time for all the pieces with 80% of that being the center section. Boss 302 Inline Autolite Carb Intake Manifold | The Home Foundry That would be great, and also total casting weight with gating. A calculated weight from CAD derived volume would be fine too. Best, Kelly
Really interesting thread, for me. My initial thought here was sand/sodium silicate/C02. The OPs process seems to me to be more conventional sand casting using sacrificial patterns than lost foam? @ the OP Have you considered moving up to a 0.6mm nozzle? Starting with Cura 5.0 the line width can be controlled/varied as needed in the software. The wider line width then doesn't effect exterior resolution and we can print walls a lot faster. My standard line width is now 0.72mm. Also, keep an eye on the nozzle erosion. It is my understanding that the foaming agent in this first generation of LW PLAs is little more than baking soda. We are still in the 'Wild West' days of expanding filament printing. I'm expecting some rapid improvement in the understanding and application of this filament......especially when it comes to casting.
Am I understanding this correctly that the patterns are slurry-free and going into loose sand? Impressive!
No, they were coated with drywall mud. Having just dip coated another large pattern, I'm having a hard time imagining such a printed pattern of similar size surviving the weight of the slurry. Small parts are nothing but it's not a linear world. Yah, but I'd lump it all under the general heading of evaporative pattern casting......which are all sacrifical pattern processes. In fact, except for lost foam, they're all really all open cavity casting processes and even though shell and investment casters often dont observe the typical AFS or naturally pressurized gating and feed system conventions, they probably should for all the same reasons. As I said though, in true lost foam, not so much. Best, Kelly
