The Cussed Vagaries of Casting

Discussion in 'Sand Casting' started by Melterskelter, Dec 25, 2023.

  1. Melterskelter

    Melterskelter Gold Banner Member

    I make a little 8" prism casting for sale. I must have cast it a hundred or more times. It is a somewhat chunky casting and does require a riser to prevent shrink defects. Using a riser as shown below hasa been very effective in preventing defects.

    Here is an overview of the casting with the riser having been removed from the back side.

    Here is a view of the blind riser with a spike of metal from a vent that I poke to the surface of the cope on a pair of castings set up the wayn I usually cast them.

    For some unknown reason recently one of the castings in a pair decided to have a severe shrink defect extending all the way through the casting. The other of the pair was perfectly sound!


    As seen from the front.

    It boggles my mind how one casting was perfectly sound and the other was severely defective.

    I guess this is more of a rant than anything else. But it is a good demonstration of just how seemingly arbitrary the casting gods can be.

    Tobho Mott likes this.
  2. Could one casting have temporarily fed the other casting through the common runner?. Just as an example of weird fluid flow effects, an old foundry guy was telling me how he connected two propane gas cylinders to a common manifold expecting twice the gas flow rate: the flow rate ended up being the same as one cylinder, with one cylinder preferentially supplying gas over the other cylinder due to slight differences in plumbing.

    If your runners to each casting were a bit longer so the junction was under the pouring spout it would make it harder for one casting to flow back to the other (in theory).
    Last edited: Dec 25, 2023
  3. DavidF

    DavidF Administrator Staff Member Banner Member

    Looks to me like you forgot to say the magic words before the pour....
    Ironsides likes this.
  4. Melterskelter

    Melterskelter Gold Banner Member

    With respect to the guy that hooked up 2 cylinders, hoping to increase flow rate, He could only have expected increased flow if he had been drawing the single cylinder strongly enough that it cooled so much that the gas would not boil off to equal demands. So what he did is, more or less equivalent to hooking up batteries in parallel. He increased the potential total amount of gas (amp-hours) that he could provide but he did nothing to increase the pressure (voltage) to drive gas through the system faster.

    Regarding the possibility that one filled long before another. That is possible, I suppose.. However, I do incline the mold so that metallostatic pressure forces each cavity to fill more or less symmetrically. Had they been not inclined, then certainly one side could preferentially draw metal for a while, then the other one and back and forth. And that can cause a mess. So I continued to be mystified as to what exactly happened here. But since it's so rare, I'm not going to worry about it too much right now.

    Concerning David's suggestion that I may have forgotten to do the, incantation before pouring. that I think is as good an explanation as I'm likely to come up with. ;-)

  5. Still it would be nice if the castings can be improved by changing the runner design, in photo number two showing both castings, is it the upper or lower casting that has the defect?. You could almost have separate pouring holes at opposite ends to avoid iron spills getting into an unpoured hole (been there done that).
  6. Melterskelter

    Melterskelter Gold Banner Member

    Neither one has the defect! I did not have a photo of the actual pair of castings that contained the defective casting. In fact, it was not until machining the defective casting that I noticed a persistent tiny 1mm spot on the face of the casting that did not seem to clean up on subsequent 30 thou cuts. I kept expecting the next cut to produce a smooth shiny surface. Nope! And another cut showed the divot to actually be larger. That's when I tumbled to the fact that I was dealing with a vacuum defect. So, just to see how extensive the vacuum defect was, I cut right through the area. Again, I was sort of surprised to see how extensive the voids were. I have seen that "tip of the iceberg" effect in other castings at other times.

    From that same batch of pours all the other casting are just fine. Go figure...

  7. Al2O3

    Al2O3 Administrator Staff Member Banner Member

    At a defect rate of 1/100 I don't think my corrective action would go beyond profanity. The solution could be worse ;)

    Tops likes this.
  8. Melterskelter

    Melterskelter Gold Banner Member

    My thoughts exactly. This is my most reliable casting pattern/molding combo.

  9. HT1

    HT1 Gold Banner Member

    blind risers can "vacuum lock" and fail to feed just venting them is often not enough especially if the vents fill with metal , a vent that fills with metal let gas out of the mold cavity but is not letting it back in. A s the riser feeds, if the sand is not allowing atmosphere back in, or if the entire riser top solidifies it can create a vapor lock, and the riser will stop feeding, that is why Cracker Cores are often used to keep the blind riser open to the atmosphere,

    additionally one of the risers is not centered over the part, that can affect how much the riser feeds .

    V/r HT1

    P.S Cast Fe can just be a PITA sometimes
  10. Tops

    Tops Silver Banner Member

    Cracker Cores?
    @HT1 , what kind of crackers- Ritz, Saltine, Chicken in a Biskit... ?



    Can you explain the term a little deeper for someone like me who is unfamiliar please?


    (edit added photo of cracker box)
    Last edited: Dec 27, 2023
  11. Kriss

    Kriss Copper

    Does the riser is located only at one end of the casting? Does the defect appeared bottom of riser or other location?
  12. Melterskelter

    Melterskelter Gold Banner Member

    It appears at the base of the riser and persists all the way through the casting.

  13. Kriss

    Kriss Copper

    Generally speaking the riser would solidify first. The place where you see the defect is last place to solidify. Because of the geometry looks that it could be that liquid metal is still very late open to mold surface. Could be interaction between the moulding material at specific casting.
  14. Kriss

    Kriss Copper

  15. HT1

    HT1 Gold Banner Member

    Last edited: Dec 29, 2023
    Tops and Melterskelter like this.
  16. Melterskelter

    Melterskelter Gold Banner Member

    Concerning the USN Foundry Manual cited by HT1 it is amazing how much useful foundry information is concentrated in that single resource.

    If you look at the section on risers, you will see that they talk about a thin skin forming on the casting and riser as they cool. And if you look at the pics of the riser and casting above you will see clear evidence of that skin. You can also see that the center vent I poked into the the riser filled with iron and cooled tenting up and partially inhibiting the collapse of the riser. That happens to be a picture several years old and I no longer make those vents in the center of the riser. I poke the top of the riser at the at its upper corner. That corner may still tent up. But the center collapses nicely. I also make the riser with a depressed center to promote collapse and chamfer its edge where it meets the casting to avoid the classic sharp inside corner defect where the rider meets the casting. (I actually think I might have grabbed the wrong riser pattern without that chamfer in the case of this recent failed casting!)

    Here is what the proper current riser pattern for this casting looks like. It is about 1.5" diameter at its base.:


    Tops likes this.
  17. Tops

    Tops Silver Banner Member

    Thanks for earmarking the page HT1 !

    from page 101:
    The most successful method of introducing atmospheric pressure into blind risers involves the use of either a small-diameter sand core, or a graphite rod, placed in the riser cavity as shown in figure 147.

    It goes on to explain the diameter of riser to core and the mechanism and size of riser to graphite rod.
  18. Tobho Mott

    Tobho Mott Administrator Staff Member Banner Member

    I think "cracker cores" came up before, in Denis's "saggy riser" thread. I had. No luck finding that entry in the navy manual then, so it's good to have a page number reference.
    For what it's worth, here's that fig. 147 from several pages later, showing the core in the top of the blind riser. Doesn't look like how I had imagined it before.

    Tops likes this.
  19. Kriss

    Kriss Copper

    This sand core for venting the riser must be something used very long time ago. the most simple way is to make notch / V shape or cone shape like @Melterskelter shows on top of feeder. Like in this picture. The notch makes a hot spot at that place and makes the liquid metal to hold longer open to atmosphere to avoid the vacuum. and let to feed the casting. This also helps to create the piping we usually see at the top of feeders.
    Tobho Mott, Melterskelter and Tops like this.
  20. Melterskelter

    Melterskelter Gold Banner Member

    Here ae a couple pics of a blind riser in green sand I poured Thursday. The casting and basin plus riser weight 40 pounds total.

    You can see that there is piping and then at the end of solidification there is a little pip of molten metal that gets squeezed out as the riser contracts. This is a very common pattern of piping and pip formation that I see in my casting work. I have never used a cracker or felt the need for one. But that may be a function of the sand, binder and washes one might use. If a wash were applied (I don't use them) to the mold that might reduce permeability to gases and a cracker might in those circumstances prove helpful.


Share This Page