First design changes
Time to hear this thing run!
This is what I printed out:
Upper plenum (2 pieces)
Lower plenum (4 pieces)
Intake manifold
Velocity stack (4 pieces)
It was all-in-all upwards of 80 hours of printing. All plenum pieces were printed with ASA, and the split components were bonded with MEK. The manifold was printed with CF-PP and was a complete nightmare to print.
The first issue was bed adhesion. I followed the advice of Google Sensei and applied polypropylene tape to the glass build surface. The filament adhered very well to the tape, but after a few layers, the contraction of the part began to pull the tape away from the build plate. I tried, and I tried, and I tried to further secure the tape to the bed with various clips and additional tape, but it was to no avail. After the part became over 3-4 mm thick, the tape lifted. The issue was with the length of the manifold. The side against the build plate stretched diagonally across the entire 300mm² surface. It was too long. This extreme length was creating an immense contracting force, but I was too dedicated to the print, and I had already spent so much time designing it (I am a complete beginner at CAD design at this time) that I didn’t even think to change the geometry to minimize shrinkage. After the tape didn’t work, I purchased a 1/8” sheet of 100% polypropylene sheet and tried to glue/tape/clip it to the bed but experienced the same problems.
I needed a polypropylene build sheet reinforced with fiberglass. The market options for such a sheet were limited, so I had to make my own.
DIY Fiberglass-Polypropylene Composite Sheets
How can I combine fiberglass and polypropylene? I know I needed the polypropylene to be thin, because if it’s thick, it’s going to require too much heat to melt into the fiberglass. Additionally, I won’t be able to build the two materials up as layers and produce a uniform distribution. So, my first bit of research was where to find the thinnest possible sheets. Knowing that plastic is often used in fabric due to childhood experiences of melting the straps of my backpack with a lighter into little molten plastic puddles, I searched for 100% polypropylene fabric. Bingo. It’s super common, it’s thin, and it’s cheap. It might even be underneath the chair you’re sitting on right now.
Next challenge was heat. I was thinking of using some sort of burner and pan, possibly a heated aluminum plate. It needed to be bigger than 300mm², heat up to around 250°c, and I didn’t have the materials to make such a device. Luckily, my wife loves thrift shopping. One day, as I’m tagging along for one of her thrifty outings, I see a t-shirt heat press sitting on the floor, and the light bulb goes off. No price tag, so I’m not sure if the owner is using it or if it’s for sale. But before I asked, I made sure to do a quick search to learn its approximate value and max temperature. Price new $200 - $300. Max temp 400°F. Perfect! I took it off his hands for $70, and we’re off to do some tests.
It worked fantastically! It takes a lot of time to warm up both surfaces, so it’s not a fast production process, but it produced a flat, usable build plate. The proof of concept showed enough promise to develop itself into an open source project of its own.
But the struggle wasn’t over…
The sheet had much less deformation than the solid polypropylene despite being less than half as thick (I ran out of material during testing), but it did not remain flat during printing by any stretch. (Pun not intended, but I like it) So I had it taped, glued, and clipped to the bed and was monitoring it almost the entire time. With a 1/8” thick sheet it would probably print quite well.
The end result was a heavily warped manifold. So I clamped the printed manifold to the original manifold with nuts and bolts, heated the aluminum manifold with a heat gun, and slowly heated the printed manifold until it softened. Then I began incrementally tightening all the bolts until the flange was flat. I then let it cool and sanded the flange with a sheet of sandpaper taped to a flat workbench until the printed manifold was as flat as I could make it.
With that, I sealed all the mating surfaces with RTV and bolted it all up!
Watch to the end of the video below to hear it run!