Cast iron lost foam

Discussion in 'Lost foam casting' started by ESC, Jan 30, 2019.

  1. ESC

    ESC Silver Banner Member

    I brought this over so it doesn't get lost in the sprue thread.
    I'm building a reciprocating hammer to form some of the details on the 1/4 scale Willys. I couldn't find a drop of 2.5 x 1.5 DOM tubing so I whipped up a foam of the ram guide housing and poured it. I have done cast iron in the past and was a bit too casual with placement for depth and compaction and with the density and fluidity of the metal it washed out the core and I ended up with this misrun.

    I made two more foamies and decided to try one with a linseed oil bonded core I had left from the planishing hammer in my regular foam sand oriented vertically and another with a core horizontally in some dead Petro Bond I reconditioned with the muller. I stacked a cheek on a cope and drag flask and actually used the Jolt Squeeze for the first time to just set the sand, no rammer. I added a couple of holes in my flotation ring pointing down so it is easier to plunge to the bottom. Then gentle puffs lift the sand so the pattern can be pushed into position.





    It has been freezing here, but one hour from a cold start I pulled at 2400*, slagged and poured with no additional additives. I was using some old iron, and remelted the previous dud casting. I had poured a chill wedge when I did that and had beautiful grey iron all the way to the tip of the wedge. This indicates that there is enough Total Carbon to make soft iron at that 1/8" thickness and since the flange and wall of the ram guide are 1/2" thick I don't expect any issues.

    It's like Christmas, waiting for the glow to fade, but I don't usually pull the iron until it's cool.

    Attached Files:

  2. PatJ

    PatJ Silver Banner Member

    Looks nice.
    I have several things I need to cast, but I just dont' want to deal with the cool weather right now.
    Hats off to you for getting out there and doing it.
    Should be interesting to see how it comes out.

  3. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Think I like your gating version into the opposite end of the square flange better than first go. I might have been inclined to gate into the top face of the flange with the part canted at 45 degrees. I haven't done iron LF but I suspect good packing is very important. Is that drywall mud? How did it hold up to the iron temps? Any melting? Did it prevent sand burn in? I've read that iron can pick up a little carbon from the decomposed pattern material. Anxious to see the result but I think you are wise and better served to wait. Also wise to keep that Kush cup outs are much more likely with the denser metal.

  4. ESC

    ESC Silver Banner Member

    Spoke too soon Pat. They are ugly, but it is my own fault.
    Core blow from the vertically poured attempt. It burped, but I kept pouring and might not have had enough head on the horizontal mold because I ran out of iron.
    No sand burn in and the drywall mud looks the same as with aluminum. I had folds and inclusions in both pieces and that might not be enough superheat. The first pour was at 2650 which gave a really nice fill and surface on flange and the start of the sleeve. I canted the first one and it had a good flange and barrel. I think you had mentioned that cast iron was bottom fed while aluminum was top fed. My feeling is that the temperature difference creates a larger void area in front of the metal and as dense as iron is it can erode the coating and wash the sand in front of it. I go back to my pour of the exhaust spacers in
    AA. They were deeper in the sand and fed horizontally, they came out sweet.



    The iron was machineable, and this piece could probably be used, just too ugly, so back in the pot.
    Conventional split pattern already under construction.

    This is the chill wedge from the first pour. Just a couple of strokes with a file as a test.

  5. PatJ

    PatJ Silver Banner Member

    Well at least you are out there pouring iron.
    That is an accomplishment in the winter.

  6. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Bummer. Guess you gotta go for what you know. I see lots of LF iron casting videos in commercial foundry settings. Which of the gating produced the near success?

    The fact that drywall mud seemed to suffice is good information and a bit surprising to me. I would think higher refractory coating would be necessary. I'm always hesitant to post about things I haven't actually done but I'll make an exception. I have LF cast lead and that is the ultimate in high density LF and I learned some things from it. It produces very high metalastatic/head pressures. At 12" of depth it would compact the mold sand whereas the mold of above that point remained stable. When I used fresh dry sand and better vibe, the depth at which the mold remained stable increased. I eventually repositioned the pattern so it had less total depth. If the height of the Kush cup above the sand exceeded the depth below the sand, I experience run outs because the sprue pressure exceeded the ability of the sand to retain it.......or weight the top of the mold.....or use a vacuum assisted membrane.

    Sooooo, with that experience,

    1. I suspect you'd need to burry that deeper, and make sure you have fresh dry sand and very good vibration and packing.....wouldn't rely on fluidizing ring to set the pattern and only jolt. The reason being the higher density metals just exert more pressure on the mold making packing more important.
    2. Bottom gating does seem more common in iron but in the commercial settings I see examples from top to bottom and in between. The reason it may make more sense in this setting is the increased temperature of the iron along with higher sprue pressures make the rate at which the pattern evaporates uncontrolled whereas introducing the metal in the bottom of the pattern and filling upward makes the fill more tranquil and controllable like conventional sand casting, instead of just falling in on the foam pattern.

    My 2 cents. Got one more try in ya? It's our first iron LF post on the forum.

  7. ESC

    ESC Silver Banner Member

    Not now, it's holding up the build. Check the link I sent.
  8. Ironsides

    Ironsides Silver

    That was a great wedge test, grey all the way. It is a pity that the casting had all those wrinkles in it.
  9. PatJ

    PatJ Silver Banner Member

    Melting and pouring iron successfully can be tricky.
    And "success" is more than getting the iron to a good pouring temperature.
    Success is as much about the molding process as it is about the iron.

    There is a reason I use bound sand, and that it because the binder I use is designed specifically for use with iron, and it works very well.
    Ironsides said I need to learn how to work with greensand, and I tried that late last year, and it was a spectacular failure, even though the iron was plenty hot (perhaps too hot).

    I would like to try lost foam, but like water-based greensand, there is an art to using it, and a learning curve too; sometimes a very long and tedious learning curve.

    For now I am sticking with bound sand because I need to get some things cast, and don't have a lot of time or desire for lengthy experimentation.

  10. ESC

    ESC Silver Banner Member

    That was a great wedge test, grey all the way. It is a pity that the casting had all those wrinkles in it.

    Yes, it was really good old Iron. I didn't mention it, but I shortened an old wood lathe and had a saddle from a small lathe Southbend that I modified for my crankshaft grinder. I started with a little bit of charcoal and oyster shell pulverized and just a handful of the Desulco carbon increaser.
    Those folds didn't show up in the first pour, so I think it was just too cool.
    Last edited: Jan 31, 2019
  11. ESC

    ESC Silver Banner Member

    My 2 cents. Got one more try in ya? It's our first iron LF post on the forum.
    Kelly, I got two good castings today so I can continue with the build. I'm thinking the lower anvil arm would be another candidate for foam, so I might give that a try. Looking back, the first pour almost made it and it give me a chance to bottom feed.
  12. ESC

    ESC Silver Banner Member

    I'm going to need a ball turning attachment for the anvils of the ShopDog, so I built a foam pattern for one that would mount in the compound of my Logan. I re-thought my approach to this and decided to build one similar to the boring head style so clearance under the ball isn't an issue. I sprued up the pattern I had already made, and then added a spiral test piece to see just how far my metal would run. Nowhere near as elegant as Ironside's or Kelly's, but I thought it would give me some results to help with future attempts. Here are the two sprued up. I used the heat gun to shape the spiral it shrunk slightly and lost some of its cross section. since I didn't care about the surface finish I didn't coat them. After I poured the ram guide I ran too low on metal after filling the ball turner to fill the spiral, so I put the crucible back in the furnace and had another charge ready in 20 minutes. Something had happened to the spiral and the start was broken from the sprue. It poured as I would have expected and I only found out about the disconnect when I pulled it. I cooled it and measured 16 inches from the sprue to the cold shut end. The metal had somehow found a path past the break and then formed back into the rectangular shape. There was soft clean metal all the way to the end and a nice grey fracture.



    The ball turner also filled completely although there was some distortion perhaps from the jolting that renders it unusable. There were some surface folds on the round portion, and whether there is enough material to machine the T nut portion needs to be determined. No bigge since I wasn't going to use that design any way. I'll clean them in the blast cabinet to see if any sand vitrified to it.
    Last edited: Feb 18, 2019 at 7:16 PM
  13. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    Inspired by the same classic spiral fluidity methods, I did a number of metal propagation rate tests with aluminum at various temps. However, I used a straight vertical coupon with a 1/4" x 1" cross section. I did the test behind glass and on video so the advancement rate in in/sec could be observed. Since most of my castings are relatively thin walled, and high surface area/volume, I thought this may be more representative. At pour temps below 1600F I seemed to max between 12-14" of total travel length, and that was also 12-14" of head.

    Coated or uncoated patterns didn't seem to make a great deal of difference, with uncoated patterns producing perhaps an inch or two more travel. I was surprised by this because if the insulating affect of the liquid/gas layer between the foam and the molten metal front was a factor in the propagation speed, I would have thought no coating would mean the gas would rapidly diffuse into the sand whereas that would occur much more slowly through a mud coating. The mud coating may have provided slightly better insulation offsetting the lower permeability. When I used vacuum assist, I did get up to 50% increase in total travel distance. Looking back, I think the 1/4" cross section may have been the limiting factor. I tired of the testing because I used plate instead of boro glass and the plate glass always failed shortly after the pour was completed and then I had to pick glass out of the sand.

    One thing you might try is video taping your test pours. I've found something as simple as watching the time to the initial hesitation and total duration of the pour and very closely approximated the propagation speed testing, and with such there was no need to view the metal advancing through glass.

    Bottom line on all this according to the experts, introducing metal too slowly can cause the pattern to evaporate faster than the mold can be filled, causing localized mold collapse. Too fast, folds and oxide defects arise. I've had best results with top fed patterns and always just tried to keep my gate contact area comparable or larger than the sprue area. With this approach I need very little sprue height/head, and besides an orientation that is conducive to packing, I like to gate into a massive section of the casting when possible. When not possible I make the gate more massive. I've had almost no instances of shrink even with large differences in cross sectional area and I think this is because you tend to get a natural degree of direction solidification with this approach.

    With iron, I would think the metal propagation speeds would be faster due to the significantly higher heat and density increasing the sprue pressure. The commercial videos have indicated lost foam requires 50-100C higher pour temps than traditional sand casting because that is the additional energy required to evaporate the pattern. Interestingly, it's been 50c for iron and 100c for aluminum. I think this is because even though iron is so much hotter, the difference in specific heat and the fact we're talking the difference between pour and freeze temp of the metal is what is in play not just the absolute temps.

    Not having poured LF iron, I'd probably go the same route until otherwise swayed.

  14. ESC

    ESC Silver Banner Member

    As I mentioned in the ram guide thread I was thinking Campbell sprues rather than lost foam. And bottom line the temperature was the difference. I checked the needle on the optical pyrometer and adjusted it for weaker batteries. I also take a shot of the exterior with the IR thermometer. When it stabilizes at about 500* at the chalk "X" I can push the melt to almost 2700* in the crucible.

Share This Page