Bob Puhakka on Bifilm theory

Discussion in 'Links to useful information' started by Gippeto, Feb 2, 2019.

  1. Gippeto

    Gippeto Silver

    Came across a link to this video and found it incredibly instructional. Might be grasping at straws to solve some of my casting issues...or maybe, just maybe... mis diagnosing my issues in the first place. At any rate, IMHO...well worth the time to watch....may just watch it a few more times. ;)


  2. PatJ

    PatJ Silver Banner Member

    He talks very fast, and puts forth a lot of information in a short time, but what he says seems to mesh with what I am reading elsewhere, which is:

    1. Turbulence in the mold cavity is a very bad thing.
    2. Entrained air into the metal is a very bad thing.
    3. High velocity metal tends to splash around in the mold cavity and entrain a lot of air.
    4. The system needs to be pressurized; ie: the gates (or sometimes a choke point in the runner) need to constrain and control the flow of molten metal to some extent.
    5. The basin under the sprue needs to have a flat bottom to prevent the inverted mushroom shape of metal flow which entrains air into the flow.
    6. A square basin is preferred.

    One of the white papers that I am using now as a go-by for sprue/runner/gate/riser design talks about introducing metal into the bottom of the mold cavity, so that you don't have a waterfall of molten metal splashing down into the mold, and the video above also touches on this.

    The fill simulations are very informative, and give a good visual indication of what is happening to the metal as it fills the mold at different velocities.

    The white paper discusses fill rates, and generally those are quite fast for a 20 lb casting, along the order of 5 seconds.
    The balance is between mold fill rate that is fast enough to fill the mold without having cold joints, but slow enough to prevent air from getting entrained into the melt.

    The stopper in the sprue seems like a good idea, and it seems like I have seen that before, but I am not sure where.
    It would almost require two people to use a stopper, although with a clever arrangement of levers and foot pedal, one person could maybe do that while pouring iron (yikes!).

    I very much agree with what he says about committees, and I have found that to be true over a period of 18 years working in corporate.

    It was a good video.
    I will try to dig out some fill simulation videos that I found.

    Last edited: Feb 2, 2019
  3. PatJ

    PatJ Silver Banner Member

    Here is another paper, apparently on the same topic.
    It seems to make sense, and the paper seems to indicate that testing of castings proves the point.


    Stating a theory is good.
    Actually getting the defects to drop to nearly zero in your casting process is the "proof-is-in-the-pudding" thing, and the thing that really gets my attention.
    Talk is cheap.
    Repeatable results are golden.

  4. PatJ

    PatJ Silver Banner Member

    Here are a few fill simulations I have run across.
    There are many of these out there if you search for them.

  5. HT1

    HT1 Silver

    I was taught all this By the Navy in 91, not particularly new, what is new is the Methods of execution. , one of the interesting things he mentions is Binders as Fuel, this shows clearly the use of resin bonded sand, perhaps Petrobond, because traditional green sand will have little to no flammables in it , starch, flours, seacoal, which are all trace binders mostly added to add collapsability or peel. resin sands allow gating systems that would be impossible in greensand or petrobond ( Basically your mold can be a large group of what we think of as cores assembled)
    . True bottom gating is all but functionally impossible without some serious equipment . the simplest way to do it is to design your mold so it will be poured on it's side ... the Cope and drag become the right and left side . this allows you to get the ingates under the casting . this technique requires your molding to be dead on. the smallest mistake has your metal pouring onto the ground , and for any reasonable sized mold will require specialized handling equipment.

    But as has always been my issue with Campbell. the Formulas are no where to be found, if you watched til the end , the First question was where are the Formulas. you noticed he even glossed over the correct angle for making sprues ... no specifics

    V/r HT1
  6. Melterskelter

    Melterskelter Silver

    I was very sorry to see that a velocity limiting sprue/gate, though mentioned by him several times and even flashed on the screen a time or two, was not discussed. It looks like he used to have a blog and other videos posted, but for some reason all that information has been taken down. Is that because he is now selling rather than giving away the information? Has anyone found information on the velocity limiting sprue/gating he alludes to? It looks like he has some potentially very useful information. How to access more of it from him or other folks working on similar theories?

    Indeed, I have a couple of patterns that would have to be poured with the part line vertical in order to fill from the bottom. That would be a challenge. Hmmm. Thinking, thinking....

  7. ESC

    ESC Silver Banner Member

    I didn't finish watching the video, but as to your questions, HT is there a specific formula you are looking for?
    Denis, from what I can see in his 2011 handbook, the velocity limiting sprue/ gate has to do with the fact that if you can produce an elbow at the base of the sprue that joins the gate/runner the surface roughness of the greensand at the rear of the elbow reduces the velocity by about 20% in tests.
  8. Gippeto

    Gippeto Silver

    Being basically new to casting/foundry work, the only thing I know for sure is that I don't know much at all. lol. I found it very interesting just how many common casting flaws could be related to the oxides...eliminating them might prove impossible for the home gamer, but finding practical methods to limit them seems like a worthwhile thing to dig into.

    And heck, we're heading for -35C in the next day or's warm here by the confuser. ;)

    HT, he did (sort of) give one formula for the sprues...F = A*V where F is volumetric flow rate, A is cross sectional area of the sprue and V is the velocity of the metal at that point.

    It's been ~30yrs since I did any physics so dig out the salt shaker.

    Given acceleration due to gravity is a constant, I crunched some numbers for a random sprue, threw that into CAD and came up with 8 degrees included angle.

    On a 100mm (4") drop and starting at zero velocity from a hypothetical pouring basin, I calculate the metal as doing ~55in/s

    As it seems he works with John Campbell, I would think the information is in Johns book...bit pricey but maybe worthwhile.

  9. Al Puddle

    Al Puddle Silver Banner Member

    babbling removed
    Last edited: Feb 4, 2019
  10. Petee716

    Petee716 Silver Banner Member

    Yes and yes. Some of the slides shown in the video are in Campbell's book as well. The video is more like a lecture and the Campbell book is the textbook.

  11. HT1

    HT1 Silver

    .F = A*V is not particularly valuable since F is a variable that is set by the designer of the gating system... My problem with Campbell is that everyone very quickly ends up at a Multimillion dollar computer simulation. where is the Math Behind that . just for the Simple stuff say Billets. to be completely honest with you everyone is protecting trade secrets, at this point even the American Foundry association. which used to try and educate the entire industry. so what we hear is "you are doing it wrong" with no guidance on how to do it right... which is total BS. the few practical things we actually learned :
    1) "Pouring cup stoppers" all but impossible for the hobbiest
    2) "correct sprue taper" No specifics
    3) "bifilm is bad" well no shit Sherlock
    4) not mentioned, but implied fill molds slowly ( pouring temps???) we need that imformation, along with the heat loss rate of the various sand systems ( this number is very complicated,) will be in degrees/sec / amount of surface area to volume ratio.

    anyone here ever calculated the surface area to volume ration of one of your castings????

    in the nicest way possible bifilm and just about anything advocated by Campbell is out of the hobbiest's realm, we are talking serious complicated math... Thus the computer simulations

    V/r HT1

    P.S. just a quick summary:
    1) avoid anything that causes turbulence in your gating systems
    2) the AFS gating system does a poor job of limiting turbulence unless you properly bottom gate (which is impractical at best )
    3) your sprue should have a very specific taper, but it is a secret
    4) you do not want your pouring cup to be a funnel,
    5) you do not want your pouring cup to "glug glug" pump air into the metal
    6) you want the mold cavity to fill slowly (related to 1.) but fast enough not to cold shut , but the specifics on how to do that are a secret
  12. Melterskelter

    Melterskelter Silver

    Exactly! Even more frustrating was looking for other Pruhakka videos which might have given clues to some of those points and finding they had been online at one time but the links are now deadends since the information has been taken down.

    At the same time, he has done a lot of work and can make a living from his intellectual property. So, I guess I can see why he is protecting it. Frustrating though.

  13. Gippeto

    Gippeto Silver

    Continuing to look into this and I agree hard numbers are few and far between. The closest I've found was on one of Martins (Olfoundrymans) videos where in the comments he gives the dimension of the small end (8mm) and the angle (1.3degrees per side) of the sprue former he uses. There's some history on pouring basins as well..was well worth reading for me.

    Melterskelter likes this.
  14. HT1

    HT1 Silver

    if you purchase John Campbell's books Starting with Castings, the information is there, but it is a complicated read, and does not lend it'self to easy solution of complicated problems. Most people will be completely overwhelmed by it. common, we have people that have to be smacked on the head to not Just pour molten metal into risers, most gating systems are better then that. we have known sincve the 50s that the AFS gating system had serious flaws (that was when the first Xrayed Pours of molten metal was done, and molten metal doesnt act the way the AFS system expected it to). I mean when I posted the AFS Math and Showed that No Hobbiest should be using larger then a 1/2 inch Sprue cutter there was doubt. Backed up by the fact that a 1/2 sprue cutter has to be special ordered and there are no smaller ones. I do agree with alot that the video says, but I know it is not implementable by the hobbiest community

    V/r HT1
  15. PatJ

    PatJ Silver Banner Member

    The castings that the backyard crowd make are generally on the small side, and it can be tricky to scale down equations that are applied to castings from 10 lbs to 1,000 lbs.

    I think a few things can be learned from what is in the video though even without exact calculations, such as keeping the sprue full, etc.

    As far as a design/layout of the sprue/basin/runner(s), gates, risers, I think the backyard folks have to experiment with what works and then stick with that if possible.
    I have had success with very rudimentary sprue/gate layouts, and also had success with more complex sprue/basin/runners/dead ends/gates/riser arrangements, so luckily there appears to be quite a bit of latitude with what actually works in a backyard setting.

    I have run across some who discover methods that work well for them, but these methods are far from what is considered conventional or accepted in standard foundry practice.
    To each his own and more power to you if you have a non-conventional method that works for you, and this is why this is a hobby to most.
    But some are also producing commercial castings, and so there is definitely a need/desire to find an optimized arrangement/method that minimizes miscasts and defects, and maximizes casting strength and machinability.

    I think you can learn a lot by just pouring water out of a pitcher, and vary the height of the pour, and also pour into bowls that are round and have a round bottom, and then bowls that are square with a flat bottom. Its really not rocket science, and you can see a lot of what goes on.
    A very fast pour may erode the sand, and too slow a pour may cause cold joints or a partial mold fill.
    You can also experiment with pouring water into various sized sprues and see which pour rate keeps the sprue full.
    Generally speaking, it is the gates that provide the restriction to flow and thus keep the runners/basin/sprue full, but the diameter of the bottom of the sprue also seems to be critical.

    And I don't think a pour is necessarily linear, ie: I think I start out on the slow side until the sprue is filled, and then I increase the pour rate to pour as fast as the sprue will take metal without overflowing.

    I think a little experimentation is the key to finding an arrangement that works for us backyard guys.

    Last edited: Feb 4, 2019
  16. HT1

    HT1 Silver

    I'm not sure it is Possible/ likely anyone here can poor fast enough to cause erosion, unless their sand has critical issues
    In AFS the bottom of the Sprue NORMALLY restricts the Flow of Metal , Sometimes it is moved to the Beginning of the Runner. I think that is a Better idea, But harder to Implement.
    I know None of you follow me on facebook, But I have about 50 Different Patterns Now, and the vast Majority of them use EXACTLY the same sprue, splash well, and runner , I have a cast set I use in them, gates vary based on the Casting being gated. BUT the vast majority of my work is Plaques on a flatback. so a couple of pounds of metal going into a thin, under 3/8 inch section through gates at the parting line. Probably the easiest castings imaginable

    V/r HT1
  17. JoeC

    JoeC Copper

    I have corresponded with Bob and also Dr Campbell about what is now being called naturally pressurized gating systems

    These are some very seriously smart gentlemen

    I have been able to implement these concepts at my day foundry job, with just incredible defect reductions

    I agree Campbells books are a tough slog as is Bobs video

    There are no formulas anymore not because they want to sell it but because it is just assumed everyone you have simulation software....

    I don't have software and am too old to learn it, but the basic concept are simple, just go ahead and try whatever feels right based on what you have read/seen

    If it looks right it probably is...remember Junkyard Wars where they built and flew airplanes just by show ever (the British original that is)

    Bottom gating isn't that hard, even for hobby casting - I will leave it as an exercise to the student to come up with a simple design....if you need to know I can post one idea later

    Also consider these concepts are for engineering castings where you need the absolute best properties, SpaceX engine parts and oil rig leg node steel castings in subzero water subject to fatigue loads - Bobs claim to fame

    For a commercial casting (shape), just go ahead and pour it the easiest fastest way you can
  18. Melterskelter

    Melterskelter Silver

    I guess I did one recently that I think qualifies as bottom gating. Or maybe I don't fully understand the concept yet...

    This casting was one of two that I made to produce a two-step pulley. I filled it from the bottom hoping it would fill without defect. It did. But for such a simple casting I probably could have poured down the riser and gotten away with it.

    I will add that this does not have a vertical component to the fill point on the casting and I am not proud of my funnel-shaped pour basin and the sprue is likely much oversized. Maybe critique of this example this will serve as a starting point for discussion.

    Had I done the drag in two parts---with a, say, 2 inch "drag" of the drag and a 2" "cope" of the drag it would have been possible to lower the runner 2 inches and have a 2 inch vertical gate to fill from below.
    StepPulley (4).JPG IMG_5105.JPG StepPulley (3).JPG StepPulley (2).JPG StepPulley (1).JPG
    Last edited: Feb 4, 2019
  19. Melterskelter

    Melterskelter Silver

    Looking more at this casting and the vortex filling advocated at one point by Puhakka, I might have done better to have a single runner entering the cavity more tangentially. that might have allowed a less turbulent swirl of metal rather than the two rivers of metal colliding in the middle of the cavity. Don't ask me why I used two...cause I have no good answer.

  20. PatJ

    PatJ Silver Banner Member

    I think your photos prove the point that there is a wide range of configurations that may work quite well.

    I generally use two gates instead of one, and I do that because early in my foundry work, I only used one gate for some parts, and had a partial fill.

    I did a layout recently for cae2100's surface plate casting, which I plan on casting soon, and I went by a publication by Sorelmetal called "Ductile Iron - The Essentials of Gating and Risering System Design and a Few Practical Notes on Risering".

    The paper can be downloaded here:

    The calculations are pretty straightforward (if I understood and applied them correctly).

    A few things they recommend that I have not tried is putting the gates on the bottom of the runner, not the top.
    All the dimensions for this arrangement came from a spreadsheet that I made in accordance with the Sorelmetal paper.
    The choke points are the gates.
    The entire casting is in the cope.
    The risers will be blind, and I will add one riser in the center of the piece.

    Its not that I would spend this much time calculating every part that I cast, but it would be nice to have one layout by the book that either proves or disproves some of the theory.

    I guess if you have methods or a system that work well for you, then I would not change the method just for the sake of change, but rather would continue doing what works well until you find a situation that does not work well, then experiment.










    Last edited: Feb 4, 2019
    Tobho Mott likes this.

Share This Page