This week’s update: I spent a fair portion of my time this week continuing the modeling of the shaper. I released the first video in the shaper project series. I purchased steel for the shaper and mill. I built a hot wire foam cutter and used it to make the column front pattern. I used the foam pattern to cast the column front, although I did have some issues with the casting. Finally, I started making the pattern for the ram.
This post contains affiliate link. I make money if you use these links to buy things.
Modeling update. I am modeling the shaper in Fusion 360. The plans for the shaper are found in The Metal Shaper book. I am adapting these to metric quantities using a conversion table in the user defined parameters. This is a feature of Fusion 360 and makes adapting the design very easy and hopefully portable for those who may still want to build it using imperial dimensions.
One big area in which I feel I am spinning my wheels: structuring the component tree within Fusion 360 to most efficiently build the model. For instance, in preparation for casting the column front, I modeled the books dimensions for that part. However draft is required if molding the pattern with green sand. After casting, I have to machine off the draft in some places, drill holes, tap holes, not to mention dealing with sprue and runner placement, removal, and clean up. What is the most effective way to structure these components in Fusion 360? I don’t have a definitive answer, although I am honing in on it. I have tried creating a body in the root of the component that is the “ideal” shape of the part–no draft and no machining. I then copy it to create a sub component which will be the pattern. I will split that component into cope and drag half components if required. I additionally have started creating lost foam patterns. Then the pattern(s) get copied to the casting. The casting get copied to machined casting with draft removed where necessary and holes drilled and tapped. This seems like a lot of work, especially now that I have discovered lost foam which eliminates the need for draft. Take a look at the latest model to see examples of this structure in the column front and ram components.
I have also been playing around with generating drawings from the Fusion 360 design. Fusion’s capabilities for drawings are not nearly as polished as the modeling environment, but they are useable and have improve greatly since last year. I made progress with using part numbers in Fusion. This should provide me some ability to generate bills of material (BOM) and potentially even doing some quality assurance automatically.
I placed the order for $540 worth of steel stock in metric dimensions. Wow! That really seems like a lot, but it should be enough to do the shaper, mill, and have quite a bit left over. This includes about a $70 premium to have it shipped. In my cost estimates, I will probably omit shipping when I amortize the cost of the steel. The steel should arrive next week. I still haven’t ordered hardware, so I don’t think I’ll be assembling the column next week even though I’ll have all the parts in hand.
One of our Makercise patrons requested that I attempt some lost foam casting. I decided the column front would be a good place to try it. I built a simple hot wire foam cutter to make this possible. Currently I have just clamped it to the side of a scrap of melamine, since this was just for proof of concept. The cutter is made from some scraps from poplar boards (I milled the boards using my homemade chainsaw mill), a transformer from a broken lamp, dimmer switch, stainless steel hanger wire (this is 22.7kg or 50lbs wire with strands that are 0.33mm or 0.013″–not information found on the package). I referenced Grant Thompson’s styro slicer video to get going, but I opted not to save $5 by not buying a pack of springs at the big box store. During my testing, I observed why the springs are a common design element: the wire stretches as it warms. My rig is tensioned by a long threaded rot opposite a fulcrum from the wire attachment eye hooks, therefore I have to tension the wire after heat is applied. This will probably cause the wire to burn out faster, but it seems to be working for now. Eventually, I may make a more permanent setup.
After making the hot wire foam cutter, I cut a handful of foam letters and shapes to get comfortable with the cutter. I decided to cast some test coupons before spending time making the column front pattern. I did not had good luck with my one previous attempt at lost foam–most likely because the sand was too dry. For this new attempt I decided to be more scientific with my test coupons. I placed one in the sand bare, one in the sand with a partial coating of plaster of paris, and one in the sand with a full coating of plaster of paris. The plaster of paris did a phenomenal job of preserving surface detail and keeping sand from sticking to the coupon.
Based on the coupon success, I cut parts for the column front and used hot glue to assemble it. I melted a half pot of metal using the last of the ingots left from the lathe project. I am running out of metal and will have to start breaking down the transmission housings before casting again. (Oil burner scrapping side project in my future?) I also felt like my propane tank was on ‘E’ or almost empty. Short story: I cast the part without enough metal. I believe the part will be useable, but the sprue and gate are nearly absent. I’d rather be lucky than good.
Finally, I started using the foam cutter to make pieces for the ram pattern. I will probably use a 3/8″ (9.5mm) shaft to melt a semi-cylinder into the front of the ram. That hollow space will provide a space to place a 9.9mm arbor for the rotating head. I should be able to ream the rest of the way after casting and assembly of the ram and ram cap for a tight fit on a 10mm shaft.
Next week I plan to order hardware for the project, finish the ram pattern, cast the ram, and start assembly of the front column. Again, fairly ambitious plan.
PS. Here is the video from earlier this week in case you missed it.