48" Camelback Straightedge Pattern is in Process

Discussion in 'Pattern making' started by Melterskelter, Aug 14, 2019.

  1. Melterskelter

    Melterskelter Silver Banner Member

    Thanks to all of you guys for your encouraging remarks. They are appreciated (and needed) as I am in the slogging-along phase of filleting inside corners of the pattern, truing up variations from intended shape, sanding, filling, sanding, filling, sanding............the not so glorious aspects of pattern making. I did get 10 inside corners faired and ten inside straight segments done yesterday.

    Little will get done today as I'll be on a day-long bicycle ride on lovely Lopez Island in the San Juan Island archipelago. Fun times with about a dozen other local riders with gorgeous views, challenging hills, good food, and lots and lots of stories.


    I'll be on that ferry which is in itself a pretty ride! Carpe Diem as they say.

  2. Melterskelter

    Melterskelter Silver Banner Member

    I am trying out and liking a new process for filleting inside corners on patterns. I mix up small batches (5 to 12cc’s) of West System epoxy glue and simply dispense a little into the corner(s) while propping the pattern in such a way the the meniscus forms a smooth fillet. This method is at least twenty times faster than the frustratingly-slow Bondo or other filler methods I have used.

    In the top image I am doing a single segment of the curved bow—-the third from the left. The curve limits me to a single segment as gravity skews the fillet if I try (I have) to do two or three at once. For the curve I make the work segment tangent to the floor and lean the pattern back 30 degrees more or less.

    In this image I am filleting the 3-plane intersections 10 at once. 009DE589-C81F-45E8-8A74-10697EEE4548.jpeg

    OMM, Mark's castings and Tobho Mott like this.
  3. PatJ

    PatJ Silver Banner Member

    That is an interesting way to get a fillet.
    The filler is a pain in the rear.
    I use a filler that is softer than bondo so that it sands out more easily, but still it is really problematic.

    I have seen some use wax fillets, but I don't see how that would be sufficiently wear-resistant, unless a permanent pattern was made.

    My solution to fillets: 3D print the pattern, but that introduces its own set of problems, such as the dreaded lines in the material, fitting multiple pieces together, etc.
    And I like to work in wood when making patterns, but with the time constraints I have, 3D printing is really the only way for me to get any pattern work done.
    I can make a large number of fillets in a 3D program in a very short period of time, and they are 100% accurate and consistent.
  4. Melterskelter

    Melterskelter Silver Banner Member

    Today's shop time was spent mostly making flasks to accomodate the new patterns. The inside length is 54" and width 13" with the drag being 4.5" deep and the cope 3.5". I started out making them from 2 X material with end plates to attach studs that allow lifting and turning the flasks and also allow binding them together. I had intended to just weld 1/8 x 1" bars one foot long to the end plates and attach the bars with screws to secure the plates to the flasks. But after I got to looking at that setup, I thought I could get both side-to-side stiffening of the flask sides (desirable) and greater tensile strength (not really needed) but nice if I extended the bars full length and screwed them down full length. The increase in lateral strength is impressive. (Works like a steel-wood flitch beam.) This stiffness was a concern because of the weight and length of the flasks and mold. The mold will weigh about 450 pounds. If the sides flex some as the flsk is rotated, there is a big risk of sand dropping out. I think (and hope) this system will be rigid enough to avoid such problems. Am I posting this because I think this is an ideal solution to the problem? No, not at all. Just reporting on what I am doing FWIW. We'll see how it works.

    I made a set of steel flasks quite some time ago for the 36 straight edge. They are wonderfully durable, but they are pretty heavy weighing about 85 pounds total. this wood/steel composite flask is about 1/2 that weight. And the steel flasks were somewhat expensive and time consuming to make. If this pattern is a winner and I burn up the current flask, I may then make steel flasks for it.

    Tomorrow I will finish up the cope and start painting the pattern tonight. Paint always reveals the defects in the pattern that will require filling and sanding, filling and sanding........

    Pattern Making (27).JPG Pattern Making (28).JPG Pattern Making (29).JPG Pattern Making (26).JPG Pattern Making (25).JPG

    Jason, Tobho Mott and joe yard like this.
  5. oldironfarmer

    oldironfarmer Silver Banner Member

    Very well put, Denis.
  6. Melterskelter

    Melterskelter Silver Banner Member

    i think the pattern will be rammed up tomorrow and possible cast then as well. The casting depends on weather which is pretty unsettled right now.

    Here are a few pics.

    Pattern Making (33).JPG Pattern Making (30).JPG
    Letters from Freeman Supply are glues to aluminum 20ga strips 1/4" wide using cyanoacrylate glue. the edges of the strip provide alignment guides and I rule the strip at 3/8" vertical intervals. The ruled lines help get the letters level too. I use a syringe and needle to dispense just a drop or two of glue per letter. Since the gloue is as thinas water it wicks under the letter. I sop up any excess using the edge of a paper towel to wick away any extra glue so that the letters stand out sharply.
    I let the strips into the wood using a router bit in the mill and glue the strips with epoxy.

    Pattern Making (32).JPG

    Registration pins---one at each end of the pattern. 3/16" diameter and drilled and tapped so I can pull them if needed. they are cyano glued into their hole but just protrude enough so that their full diameter only enters the female side of the pattern by 1/16 inch or so. If they penetrate to deeply they can easily bind. I jsut wnat them to barely fully engage the diameter of the hole for a very short distance. Pattern Making (31).JPG

    I will not be at all surprised if the pattern needs a little adjustment to get it to pull cleanly.

    Last edited: Sep 10, 2019
    _Jason likes this.
  7. PatJ

    PatJ Silver Banner Member

    Pattern looks great.
    That is going to be a very long runner, but you seem to be having good luck filling the shorter ones, and since you can measure the temperature, you could add a bit of superheat if needed.

  8. Melterskelter

    Melterskelter Silver Banner Member


    I am also concerned about the runners and filling the mold. I am going to use the pictured runner below. I suspect it may be unnecessarily complex, but a similar runner pattern works well on the 36. The runner pattern is 1/4” thick and I narrow the gates to about 3/16 thick as they enter the mold.


    Last edited: Sep 10, 2019
  9. Melterskelter

    Melterskelter Silver Banner Member

    Just a brief note but, for me, an important one. I was able to ram up and cleanly pull the 48"pattern today. Whew---it took every bit of concentration and care I could muster. I was worried I might have missed getting enough draft into and area or that I might not be able to pick the pattern out of the sand without bumbling somehow and damaging the mold. But it worked! True, the cope half FELL OUT of the cope when I was moving the cope latterally on the barn door track that my chain hoist traverses on. Maybe that was due to applying graphite to the entire cope half of the pattern or maybe it was just the slight vibtration of the moving on the overhead trolly. In any event, I was less generous with the graphite on attempt #2 and flipped the cope right way and then traversed rather than traverse with it level in the same orientation it had when I lifted it off the drag.

    Tomorrow I will try to fill it with iron. that may be another story!

    No pics today (just trust me!) as I was totally concentrating on not screwing up the mold-making.

  10. PatJ

    PatJ Silver Banner Member

    I like that runner layout.
    Lets hope that my liking it translated into a functional runner, but it seems to be proportioned correctly.

    I noticed in the old foundry pictures that for the larger flasks, they use bars running side to side along the top of the cope, and maybe the bottom of the drag, and this prevents dropout.

    I tell myself that I have the process memorized, but then I forget to do steps.
    I am going to make a molding checklist, and have it on the molding table, and check things off as I go.

    I hope to pour some iron today, weather and schedule permitting.
    We need to figure out how to live stream these pours on the same screen.

  11. Melterskelter

    Melterskelter Silver Banner Member

    Actually I inadvertently caused the sand dropout. After I riddled a generous amount of sand onto my cope pattern, I used my hands and patted it down onto the cope pattern. I never thought that rather gentle amount of compression would be enough to form a (false) separation plane in the sand. Wrong! I never thought, therefore, thought to rake the sand surface before adding more sand. Normally I use a garden 3-pronged hand rake to break up a sand surface I’ve rammed in order to allow the soon-to -be-added sand to stick to the previously rammed sand. Had I done so, there would have been no sand dropout.

    However, it was very clear that the plane formed by the riddled sand had fallen out from half of the cope exactly as I had packed it. Lesson learned.

    I do use a written checklist on more complex patterns. Otherwise it is far too easy to miss key steps. The concentration I was referring to in this car was more to do with making a smooth straight vertical lift. In found this challenging on this long spindly pattern. There are only a few steps on this pattern. But they challenge me based on their size.

    Good luck on your pour! Luck seems to always be something of a factor.

  12. PatJ

    PatJ Silver Banner Member

    Thanks, same to you.
    I am getting the process more down to a science, but there are always those wildcards.

    I try to refine the technique with each pour, and pick up tricks and tips from you and others who do iron, and incorporate those as I go.
    I am trying the mold wash for the first time, and am interested in seeing if that improves surface finish.

    I hate that I have to wait until tomorrow to break the castings out of the sand.
    That is one advantage of pouring aluminum; you can see how it turned out in a very short time.

    I did not get any instructions with the mold coat (typical for foundry raw materials), and so I am not sure exactly how thick to apply it.
    I sprayed it on the mold cavity in a medium coat, and on the runner with varying thicknesses, from thin to quite thick and smooth.

    Should be interesting to see the results on the surface finish.

    I also left a non-critical part of one mold uncoated for a comparison.

    Last edited: Sep 11, 2019
  13. Melterskelter

    Melterskelter Silver Banner Member

    Actually a small part like you are going to cast will cool pretty quickly, so you could open it an hour or two later. It will be down to 400 or 500 degrees by then. You can check that by drilling an 1/8th hole down to the casting and using a kitchen thermometer to check. I’ve checked my SE’s 4 or 5 hours later and they are down to 5oo degrees or so.

  14. Melterskelter

    Melterskelter Silver Banner Member

    I did do a trial pour of the pattern on Wednesday. There is good news and some bad news as well.

    The good news is that the runner system seems to work well and the mold filled fully(until it leaked). That is very encouraging as getting a complex mold to fill completely can be challenging. On the 36" Featherweight I went through quite a few iterations of runners and gates prior to getting it to fill reliably. The surface finish on the casting looked great and the detail rendition on the lettering was excellent.

    Now the not-so-good news: I am kind of embarrassed that I forgot to put the mid-clamps securing the center portion of cope to the drag on. I blame this rookie oversight on having visitors at my foundry. (I love having visitors, mind you, but on a pattern there is a lot to think about and remember. Explaining everything as you go can be a bit distracting just like it is when working at a machine etc.) Since this is a long and relatively shallow mold and since I use wedges under the drag at each end to get the mold level, the drag predictably must have sagged a bit under the weight of the sand. This resulted in a very thin gap or near gap between the cope and drag. Then, when I added molten iron, the head pressure must have wedged the two parts apart allowing a 1/32 gap. That is plenty for molten iron to insinuate itself and as the iron enters is adds to the horizontal surface area of molten metal which further increases float pressure. At that point the result is sadly predictable.

    Lesson RElearned! I have not had a breakout like that in 18 months. Super frustrating as a lot of work and a weather opportunity goes down the drain. Oh well, I'll give it another go next week. It's just part the game.

    a couple of pics---might as well share the misery:



    You can easily see all the fins especially in the center four bays of the casting. You can also see that that each end of the casting looks pretty normal---that is because the fatal sagging happened in the middle of the mold, but the ends were held together by clamps that hold the ends together. One interesting thing I noticed is that I pulled this failed casting from the sand while it was still red hot and it cooled pretty quickly in the open air. Every once in a while I heard a pretty sharp TINK. And then a chunk of fin fell from the casting. The explanation was that the thin fins cooled really fast and contracted quite a bit whereas the main part of the casting cooled a bit slower. So much tension developed between the fina dn casting that the metal fractured.

    I will use my calibrated Wilson Hardness Tester to see how much hardness developed in the casting because of this rapid cooling. I usually let the casting cool at least 6 to 8 hours in the sand so that the casting is down to 400 or 500 degrees before opening the mold to avoid hardening.

    One other note: The casting weighed 34 pounds as miscast and, when corectly cast, should hit the target of 37 pounds very closely. that means it will machine to 27 to 34 pounds depending on the owner's needs.

    Covering my face with my hands as I post this....

  15. Tobho Mott

    Tobho Mott Silver Banner Member

    You're allowed to make one small mistake every 18 months (at least I hope so, for my own sake). :D

    Bet you'll nail it next time!

  16. PatJ

    PatJ Silver Banner Member

    That is a bummer about the flask sagging.
    I had a mold split and empty all the bronze out on the ground (I have a video of that).
    There is so much work leading up to a pour that it makes me very nervous during the pour, knowing that I can easily botch several hours of work.

    And the think is that the second time around is not near as fun as the first mold making.
    I think I cast the base on one of my engines 7 times, with two castings per try.
    I was trying a lot of different techniques and sand types, and was learning all sorts of painful lessons. No pain, no gain as they say.

    I was using clamps for a while, and that was problematic, so I have resorted to using metal weights on top the molds.

    That is a great effort, and quite an impressive piece, even if flawed.

    You will get it next time.
    Seems like proof of the runner concept anyway.

  17. Melterskelter

    Melterskelter Silver Banner Member

    For me, clamping the mold shut is different than using clamps which I really dislike. I have seen Youtubes of folks using common woodworking bar clamps on molds with 2' bars sticking up all around the mold making an interesting obstacle course for maneuvering a crucible of metal.

    Except for the first one or two pours I ever did, I have used clamps consisting of tabs of wood spanning the joints between the cope and drag or, even better, oversized pieces of plywood laid on top of the cope with all-thread secured to the drag and piercing the plywood. A nut on the all-thread pulls the drag into the cope and pulls the plywood down onto the top of the cope preventing separation of the cope and drag and also preventing the sand from lifting. I try to keep the all-thread pieces short so they only protrude above the plywood by a half inch or so.

    I am not sure that weights alone on the cope would have solved the problem on this pour as I think the drag sagged away from the cope. A lot of weight would probably sag the cope down onto the drag preventing a leak. But then you are bending the cope and drag both. With the all-thread/ply clamping, the sand is held down definitively and no increased sagging is induced.

    Pattern Making (36).JPG
    I don't think I will forget to install this on the next pour.

    Mark's castings and Tobho Mott like this.
  18. Melterskelter

    Melterskelter Silver Banner Member

    I have tried packing this pattern several times and had trouble with the cope pattern piece FALLING out of the cope as I lifted the cope mold/flask with the chain hoist. I guess I included enough draft at 2 to 3 degrees and the graphite I dust on is allowing (too) easy release of the pattern from the sand. I think the slight bumpiness of the lift from the chain hoist strongly contributes to the problem. If this mold only weighed 50 pounds and I could simply lift it smoothly by hand, I doubt the pattern would fall out. But the bumpiness is equivalent to light rapping on the pattern while gravity provides downward traction.

    Anyway, I tossed around a bunch of strategies intended to retain the pattern. The one I settled on was to use retractable pins that extend into the sand on each end of the pattern: I simply drilled slightly undersized holes in the ends of the pattern at about a 45deg angle and then tapped the hole in the wood pattern. Into that 3/8" hole I screw in a 1.25" long 3/8" set screw that has been turned down to a shallow taper for 3/4" I extend the pin and then pack the cope being sure to pack the sand tightly in the area of the pin to ensure good sand strength. After separating the cope and drag, I use a t-handle hex wrench to withdraw the pin into the pattern and then draw the pattern using a guide (more on that when I have pics). This worked very well for me yesterday. I plan to simply bore/clean out the pinholes in the sand to be sure there are no loose bits of fractured sand on the edges. I considered packing the holes full of sand, but the holes are not easily accessible.

    If the weather holds, I plan to pour the mold today. You can be sure I will be clamping the cope to the drag.

    IMG_5994 (2).JPG IMG_5995.JPG
    IMG_5996 (2).JPG

    Tobho Mott likes this.
  19. Melterskelter

    Melterskelter Silver Banner Member

    Here is the guide I used to help me make a straighter lift on this rather long pattern. I simply place it so that the long back is parallel to the sole of the cope or drag pattern and the legs are just barely (1/64th") outside the ends of the pattern. I can sort of lean my hands on the 4" tall back of the guide and use it to help stabilize my arms and to provide a stable surface against which I slide out the pattern. The ends help prevent side-to-side wandering of the pattern as I lift it. It felt like a real help getting this pattern cleanly out of the sand. I guess if I were more coordinated or steadier, such a device would not be needed.
    e upload_2019-9-20_11-6-39.png liftguide.jpg

  20. PatJ

    PatJ Silver Banner Member

    That is an interesting problem (the pattern falling out of the mold), and not one I have ever heard of.
    You must have a very smooth pattern.

    Good luck with your pour.
    We will be rooting for you.


Share This Page