CNC Router Upgrade Parts

Discussion in 'Lost foam casting' started by Al2O3, Feb 28, 2023.

  1. Tops

    Tops Silver Banner Member

    Very nice Kelly. I and my CNC are both impressed and jealous all at the same time (remember, don't humanize machines, they hate that...:) )
    What will be your clearance underneath the dust collector? Optimized for 2" foam?
     
  2. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I will have somewhere between 10 & 12" under spindle and Z travel, and that dust shoe is adjustable in height from 0 to 10". For machining hard materials, like hard plastics and aluminum, I'll either raise the cutting surface or install a platform that does same and allows the Z axis to be (near) fully retracted and at its stiffest, thus, the reason to be able to raise the stationary Z dust shoe that high. Make sense? A lot of people dont get purpose of the stationary Z dust shoe. If you primarily machine plate/board stock of uniform height, there's no reason for the dust shoe to travel with the Z axis. In fact, you are better off because the skirt/brush clearance can be set at an optimal height, and then the cutter just plunges in/out of the the stock and shoe as the whole shebang sweeps motion in the X & Y.

    If you have a 3D surface that varies significantly in height, you need a shoe/brush that travels with the Z for more effective chip collection.

    Best,
    Kelly
     
    Tops likes this.
  3. Tops

    Tops Silver Banner Member

    Thanks for explaining it to me Kelly, that is going to be nice!
     
  4. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Between other projects, still trying to keep this one moving along. This is the core of the traveling gantry beam. It gets cast in two parts, screwed and glued endwise, and .25" tooling plate screwed and glued to each back face, and 20mm prism rails attached as well.

    The goal to is strike the best compromise between weight and rigidity. Aluminum is 1/3 the weight of steel but also 1/3 the elastic modulous/stiffness. With the the ribbing, I figure it will be about half the weight of a steel beam of the same dimension and at least if not likely stiffer torsionally.

    Each casting half should be about 16lbs and likely +8lb of gating and sprue, so a 25lb casting or so. I added a lot of gating but the cross section of the webbing is only .188". All the contact area minimizes the travel distance of molten metal and better probability of success at lower pour temps. We shall see.

    CNCd and then glued the two halves (of each half of the beam) together thickness-wise.

    1 CNCd XBeam.JPG

    Gated and ready to coat.

    2 Gated XBeam.JPG 3 Gated XBeam.JPG 4 Gated XBeam.JPG 5 Gated XBeam.JPG

    The Gantry Beam installed height is adjustable for large cutting envelope and better rigidity at lower height.

    6 CNC Layout.jpg

    Best,
    Kelly
     
    Tobho Mott, Rocketman and Tops like this.
  5. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I got the patterns coated. I got there but not without incident, and as usual I found myself second guessing some decisions.

    7 Coated Patterns.JPG

    First observation was how much heavier these parts were after being coated. They always are but these parts especially so. There were several contributing factors, first because they have a lot of surface area to coat, but also because I skipped addressing a couple things I’ve previously learned about such parts, and not only did I find myself with more than the usual level of pucker about breaking the pattern while dipping, I actually did break the first Y-branch off the second pattern I dipped, causing it to fall into the vat of slurry and causing some minor pattern damage. At that point, I had quite a mess on my hands because I no longer had a means of hanging and handling the pattern to allow the slurry to drip off, and I of course have a broken feed system and dent to fix on a pattern which was now completely coated in pounds of wet slurry.

    In hindsight, I should have attached a piece of wood across the two Ys in the feed system and opted to attach the first branch along with the sprue after the coated pattern had dried. I also should have extended the gating/runners at the bottom of the pattern and attached another piece of wood so I could support and handle the pattern at both ends, and use both to allow the pattern to rest at an angle over the vat and better shed coating off inclined surfaces instead of puddling on the horizontal surfaces. This would have made the support points much stronger and allowed the coated pattern to immediately shed a lot of weight before I set it on its drying legs.

    I also have the viscosity of my slurry at a level better suited to coat smaller parts with minimal effort in a single dip. It works well for that but for these larger patterns, that’s a significant aggravation because it causes much more slurry (weight!) to initially adhere to the pattern. I’d be better off thinning the slurry and dipping twice if necessary for these larger parts, but I’m unwilling to alter 30 gallons to do so.

    To recover, I scrambled, wiped the slurry off the broken Y, did the same on the mating spots still attached to the pattern, used light compressed air to further clean the broken joint to bare polystyrene, and hot glued the pieces back together while the pattern was still sitting buoyantly halfway submerged in the vat of slurry. From there I did manage to hang it and used a corrugated plastic board to transition it to sitting horizontally on its drying legs. A lot of slurry drained off the part onto the board which I later scraped back into the vat. I decided to go to work fixing the dent while the slurry was still wet.

    I don’t have any pictures of the repairs since I was in a scramble, but there was a relatively small affected area, and I probably could have just left it, but it would not likely have cleaned up in finish machining, and it was on a flat easily accessed exterior surface, maybe one square inch and 1/16” at the deepest spot, so I used a small putty knife and lightly wiped off the coating in the area, further cleaned the dent with compressed air, and used the same heated putty knife to apply wax to the dent which blended well. It only took a couple minutes, after which, I brushed coating onto the affected area and touched up other coated areas I had disturbed while handling.

    I set them outside and the exterior surfaces promptly dried but as usual, the interior surfaces were still quite wet. They would have dried naturally in a few days, but I decided to help them along by making a simple cardboard duct and moving some air through the interior. I left it overnight and the next morning everything was very well dried.

    8 Forced Air drying.JPG

    So now, they are ready to be cast. By the time I attach the sprue and add my largest pouring basin, it will be 40” from the top of the basin to the bottom of pattern. That will require two sessions with my large molding rig with two flask height extenders, 400-500lbs of sand, my large furnace, at least 30lbs of aluminum in my A60, and my gantry crane to assist in the pour.

    Here’s a video discussion of the above.



    Best,
    Kelly
     
  6. Smoking Shoe

    Smoking Shoe Silver

    To me that is more impressive than a text book perfect pour.
    Takes a calm kind of 'scramble' that only comes with experience to recover with your dignity intact.
    :cool:
     
  7. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I wasn’t able to get at it until almost Noon today but got everything staged and made the pour

    9 Staged for Pour.JPG

    Cut up two ingots. Each is about 30lbs. That A60 is so big it just swallows that ingot and is only half full which also makes it easier to manage during the pour.

    9.1 Ingot.JPG

    Here’s the pour:



    By golly, it came out looking just like the pattern:). Here it is degated. I estimated the part at 16lbs and the Gating at 8lbs but with the slug from the pouring cup and left over ingot, I suspect the gating will be less. I'll weigh it tomorrow.

    10 Degated Casting.JPG 11 Degated Casting.JPG 12 Degated Casting.JPG

    It was getting too late in the day to pour the second one and I was tapped after shoveling 1000lbs of sand in the afternoon Sun.

    Here’s the dented area I patched. I must say, I had a hard time spotting it but you can see where the glue seam on the pattern briefly disappears and it runs out to the edge. Glad I did it.

    13 Patched Dent.JPG

    I’ll cast the second one, remove the rest of the gating on the mill, check for flatness and tweak as/if necessary, then heat treat them before final machining and assembly.

    Best,
    Kelly
     
  8. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    I'd be willing to trade a little pride to save the pattern ;). My Father always told me a good craftsman can cover his mistakes.

    Best,
    Kelly
     
    Jiminy likes this.
  9. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Poured the second casting and enjoyed a similar result. I finished degating them and they were both very straight and flat so now on to heat treat, machining, and assembly. I was pretty close on my weight estimates.

    14 Gantry Pour Weights.JPG 15 GB Castings.JPG 16 GB Castings.JPG 17 GB Castings.JPG 18 GB Castings.JPG




    Best,
    Kelly
     
    Wachuko, BattyZ, DaveZ and 1 other person like this.
  10. BattyZ

    BattyZ Silver Banner Member

    Assuming using the quick set craft glue with weights like you have in the past for these halves?

    Overall just stupendous work Kelly! Really happy to see no cold shuts for those webs. Especially in the middle. Just further pushing along what is to be expected, if one takes time and effort doing LF castings.

    The boxiness of the gantry should really aide the cnc in fucntion.

    Good on you! Mine has been dead in the water for months. Maybe sooner than later will get some time for it.
     
  11. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Yes, Arlene's Tacky Glue, as sparingly as possible but about any polyvinyl acetate glue will do it. A piece of particle board and 40lbs of weight on top makes for a nice thin, tight, joint. Sort of amazing how strong foam is in compression. I think I could have stood on it.

    With the part only weighing 50% of the pour, it's quite the low yield and would never fly in a production setting. The lines can sort of blurr between what's actually a runner and what's a gate in lost foam casting, but I guess I'll call those four long EPS rectangular cross sections runners and the thin cove sections on them that contacts the pattern the gate, along with the additional branches into the interior of the part.

    I try to keep the sprue area at least as large as those combined primary runner areas, but that contact area on the pattern is very large. If you were thinking of it in terms of AFS system of gating, instead of 1/4/4, the sprue/runner/gate area would be more like 1.2/1/8.

    In thin walled lost foam casting, the evaporation rate of the foam slows the metal velocity even lower than than open cavity sand casting, it loses heat to the mold quickly and can't flow as far before freezing, thus all the crazy gating to minimizing the distance of that path through the thin sections. Even so, if you watch the video, the total duration of the pour is only 20 seconds, so less what stays in the cup, about 24 pounds of aluminum went down the spue, but certainly not in a uniform manner......much more the last 10 seconds than the first.

    The method allows me to pour quite a bit cooler and has produced high first pass yield/success. But I must say, you'd better have a generous size pouring cup and be ready becuase after those longer runners fill, it takes metal like someone flushed the toilet!

    It has a large Polar Moment of inertia compared to just a box section. It's a long way to go for a beam, but hey, how often do I build a machine?

    Best,
    Kelly
     
  12. BattyZ

    BattyZ Silver Banner Member

    It would be so epically cool to watch this mold fill in the real world. Although if we had that tech, everyone might be casting everything. Somebody needs to make some clear sand...

    Maybe so but it looks like it came out of a production run. Also, you didn't have to spend a million dollars on CNC pressure-cast steel dies! Watching the video, that thin webbing is more impressive. A camera pointed at the pouring cup would be super cool to watch.

    I hear ya! Surprises me every time.
     
  13. BattyZ

    BattyZ Silver Banner Member

    Machined parts coming soon??
     
  14. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Since you asked........:)

    After spending a day at 450F.......

    19 Heat Treat.JPG

    I did rough machine all six sides so everything is square and parralel within a couple thousanths. I only had to take .030" of the surfaces that will be faced and .045" off the rest.

    20 Machined.JPG

    They better look the part now but may get a very light finish cut after final assembly and after which I'll add all the mounting features. I still need to buy the 1/4" tooling plate to skin both sides.

    21 Machined.JPG

    I also need to get the rectangular tube for the rails and bed. My buddy told me he'd grind them for me.

    Best,
    Kelly
     
    Rocketman likes this.
  15. BattyZ

    BattyZ Silver Banner Member

    Yeah, buddy!! They look great!
     
  16. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Still plinking away. Having completed the rough machining, I drilled and tapped the holes in the end of each casting to join the two castings and eventually mount the beam. I used Loctite EA 9460 structural epoxy and threaded fasteners to join the two castings and ¼” MIC 6 tooling plate skins. The physical properties of the epoxy are dependent upon the joint thickness. A joint thickness of .005” is optimal and produces the maximum shear strength of 3500 psi so to control the joint thickness, I made the 7 shims that were installed at each fastener to join the beams.

    22 Shims.JPG 23 Shim Installed Epoxy.JPG 24 Castings Assembled.JPG

    With over 120 fasteners for the skins, I wised up and bought the shims from McMaster Carr in packs of 50.

    I glued the two castings together and let it cure,

    24 Castings Assembled.JPG

    Then I match drilled all the holes through the skins and castings in the same operation to save some time, sort of…….that was 120+ drilled, tapped, and countersinked 5mm x .8mm holes!

    25 Match Drill and Trim Skins.JPG

    Then glued the skins in place using all of those shims and fasteners.

    26 Glue.JPG 27 Skin.JPG 28 Skin.JPG 29 Skin.JPG 30 Skin.JPG

    At this point I need to call in a favor from a friend to machine the registration lands where the linear rails mount because my old knee mill has neither the travel nor the accuracy to do the job.

    That was a heck of a long way to go for a gantry beam. Hope it performs well!

    Best,
    Kelly
     
    Wachuko, BattyZ, Rocketman and 3 others like this.
  17. That looks great!, I'm trying to think how to sand cast it: probably have sodium silicate cores with location dowels that leave holes in the finished casting everywhere. Definitely a lot more work than lost foam in making the pattern.
     
  18. garyhlucas

    garyhlucas Silver

    I just had a box tube machined for the frame of an automated lathe. It took a CNC machine with 8 feet of travel. It's kind of neat, takes a polypropylene pipe right off an extruder and transfers it into the maching and grips both ends with expanding arbors, then grooves and chamfers both ends with two cross slides. Push a button and the tailstock and one cross slide move to a new postion for a different length.
     
  19. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Thanks Mark. If it wasn't for the glue flange on the wide side, it probably could be conventionally sand cast with draft without cores, but that would add considerable weight and added machine stock. Making a tool/pattern for such would have been a lot more work than programming and CNC'ing the foam patterns (at least for two). Having no draft made for easier fixturing and less stock removal but overall, machining, assembling the castings, and skinning the beam was as much work as the casting themselves. Producing those two castings via LF method was much less work and cost than other methods, at least for me. Glad to have it, but glad it's done. Can't say I enjoyed tapping all those holes. Not sure I'll ever be able to look at a 5mm flathead screw the same way in the future without thinking of this job.

    Best,
    Kelly
     
  20. BattyZ

    BattyZ Silver Banner Member

    Kudos Kelly. That gantry should be solid!
     

Share This Page