Bob Puhakka on Bifilm theory

As I've mentioned before, I think gas bubbles in an ingot will escape upon melting so in remelting scrap castings and sprues and runners I count on entrained gas coming off, and in the case of hydrogen, burning in the hot nearly neutral atmosphere. For lost foam I heat to a minimum of 1,650F then let it continue heating while I lay aside the pyrometer and get ready to pour. Last step is a quick skim, usually with the heat still on and then cut the heat open and pour. I can easily be at 1,750F. My porosity is very fine and now I'm wondering if it is small bits of oxides rather than gas. I don't do anything to remove gas bubbles but am guilty of pouring from too high into open cones and giant runners and entrain a lot of air. I want to do better but I'm getting pretty clean sand castings. Obviously my castings don't require the integrity yours do, but I'm sure not seeing large porosity. I have done sand castings really hot which have a rough surface but again porosity does not seem much different.

An 8 minute pour time is likely from a very large crucible or ladle which will hold it's temperature well. I know I lose a lot of heat quickly when transferring to a pouring shank with a wind. It really doesn't matter much, by the way, if the ambient is 0F or 100F if you're cooling a 1,600F crucible, 1,500F versus 1,600F delta T.

 

You just delight in complicating my simple ideas.

Good points. I think my first basins will have to be silica bound sand. Now you've got me thinking about slicing up firebrick and gluing something together with a coating.

I just see no point in tapering sprues for lost foam in the size pours I can do.

 

I've been cranking through the math and I found a couple of interesting things.
First, Bob calculated the sprue geometries incorrectly. Using conservation of energy, he determined the velocity of the fluid at the top of the sprue. Then he calculated the velocity at the bottom of the sprue assuming all the energy at the top is converted to kinetic energy. This would mean the pressure at the top and bottom is the same. The problem is there are friction losses so, the pressure at the bottom would be less than at the top, i.e. air will be sucked in. BTW, using his method, the sprue taper is 4.92 degrees inclusive.

Second, the Bernoulli equation has a gravitational acceleration variable in the pressure energy term. Turns out, that variable drops out of the equation during unit conversions when expressing energy in units of ft. or in. Pressure Head (ft. or in.)= pressure/density. In SI units Pressure Head = pressure/(density x gravitational acceleration).

 

Sure. What—don’t you want to be sure you can compute proper gating for metalcasting on Mars or the Moon?

 

I am not sure if Bob Puhakka has run across this forum or not, but just for kicks, I emailed him and told him there was a discussion going on about bifilms.

Dang if he didn't send me photos of his castings and xrays.
He is quite a nice guy, and very funny, but does not beat around the bush, and minces no words.
I have not asked him any casting questions and don't intend to do so.
I figure he is busy enough running his foundry not to have little people asking him beginner questions.
Perhaps after I finish reading John Campbell's book I could have some intelligent questions.
I don't know enough yet to know what to ask, and so would come across as a total dipshoot (even more than I normally do).

Bob said there is a blemish on the left side in one photo, which is a defect in the film, not a casting defect.

This stuff is art work as far as I am concerned.
I am definitely a believer in the bifilm theory.
It is unreal what this guy is doing with metal casting, and this quality is not the exception, it is the every day rule.
No inclusions, no voids, no porosity, just beautiful clean castings; tons and tons of beautiful clean flawless castings.
He makes huge castings too. One of his recent videos show him inside the casting!

This dude has set the bar very very high in the industry.

 

Here is a Bob meme I made, and also a screen capture of him standing in front of one of his furnaces full of molten aluminum.
Bob very much admonishes others to get their casting systems straightened out.
He has posted some cool videos showing tours of his foundry.

 

I wonder if those are gamma ray or x-ray. Gamma ray is usually used in pipe welding and had a large (relatively) physical source of radiation so the images are blurry. It is intended that they are blurry enough to not show acceptable small defects.

X-ray, on the other hand, generated electronically, and has a true point source. We had a problem one time taking some big stainless steel pipe to a company who had an x-ray machine. Starry starry night! But acceptable under code and hard to convince the purchaser's inspector. All codes have an acceptable level of porosity which is deemed to be small enough to not affect weld integrity. Is the same true of castings? Can a clean casting have 0.0005" porosity?

I found this vendor's writeup to be interesting. While they are real bottom feeders, it is a good writeup on bifilm from a different perspective. http://cminovacast.com/paper4.shtml

 

Bob has challenged others to design the system to cast a random part, and then pick any and all types of testing that is known to man, xray, sectioning, microscope, scanning electron microscope, whatever, and he will have a good casting.
He has taken his castings to Lloyd's and had them tested and independently certified, and he basically gets very high strength with an aluminum alloy (I think it is aluminum 205, or some alloy in the 200 family, and is stronger than high strength steel), and perhaps twice the ductility of what others are producing.
John Campbell states that the castings that Bob Puhakka produces are envied by the aerospace community, and are of that grade or better.
Its not rocket science; if you get rid of the bifilms in the metal, you get rid of the weak points that can crack and fail.

The industry is in denial about what they need to do to make better castings with high yields, and they don't want to admit that John Campbell and Bob Puhakka are correct in their methods.
I know from personal experience that there is nothing more difficult than trying to convince a bunch of corporate types that what they have been doing all their life is totally wrong.
The longer your boss has been doing something wrong, the more adamant he will be that he must be doing things right. It is like some law of physics.

John Campbell came up with the Cosworth Casting Process, and it was used to make race car engines that went on to win the majority of Formula 1 and Indy races for over 20 years.
The Cosworth method was adopted by Ford and General Motors, and is still used today.

So Bob Puhakka does not come from nowhere with his casting quality, he is building upon 20 years of experience John Campbell had with perfecting the Cosworth process, and other casting processes.

John and Bob's story is an interesting one, and I am glad I happened across it because I think it is some extremely valuable information, and it is information that I think for the most part can be incorporated into backyard castings.

I mean why not make great castings?

Edit:
There was a commercial on TV for insurance, and it said something like "So simple, even a cave man can get it.".
I think what frustrates John Campbell and Bob Puhakka to no end is that they have figured out this high yield/high quality method that is world beating, and the cave men don't get it, and nor do the supposedly very smart foundry owners, and the foundries won't use these methods for pride or other foolish reasons.

Edit02:
There was a similar situation at the end of WWII when the Allies went into Germany and discovered that advanced methods had been invented for making all sorts of things, such as ammonia, synthetic fuels, rocket engines, jet aircraft engines, etc. etc.
The US realized just how far behind they were technologically.
The technology that we borrowed from the Germans after the war, which was the only true reparations we received, was worth untold billions to the US and others, and most of that boiled down to efficiency.
If you find an efficient way to mass produce things, then that will save a proportional amount of money.

Today's foundries using obsolete methods will go out of business in the next 20 years or less because they won't be able to complete with those who use John Campbell's and Bob Puhakka's methods.
Adapt or perish; that is it in a nutshell.

Other countries around the world will adapt this technology and use it effectively, and many no doubt already do. The foundries in North America will wake up one day and find they can't complete with the foundries of the world anymore, and they will scratch their heads and ask out loud "How could this have happened?".
Ignorance is bliss I guess, until you go out of business that is.

Edit03:
It is analogous to a football game where your team sets you up for the perfect play, all the opponents are blocked, you have a clear path forward for an easy touchdown, you catch the ball, success is all but guaranteed, and then you fumble the ball, the opposing team recovers and scores a goal and wins the Superbowl.
I call it snatching defeat from the jaws of victory, and I witness it first hand all the time in the corporate world.
I go to great lengths to make sure I don't do that sort of thing with my business.
.

 

This is a long thread. I thought I would try to summarize some of the math that's been talked about.
First, calculation of the critical velocity. This velocity is the maximum you want entering the mold cavity.

Since most of us are using a cope height of 3.5", here is the math calculating the velocity at the bottom of the sprue and thus runner velocity.


Let me know if I made any errors (such as the units in gvcf- it should be 52 in/s - doh!).

 

This is a great thread, you guys are doing Masters if not Phd level research here

If you wonder why there are some contradictions / confusion in the information you are seeing, it is because you have caught up to the absolute bleeding edge of research, anything from 25 years to today is just that

We have been casting for 4,000 or so years, tribal knowledge passed down as alchemy, then a skilled trade, science started to show up around the great war, this work is a mere eye blink in that timeline

It is hard to change anything never mind an industry

When I was a much younger man the Queen Mary II docked at Halifax harbour

Back before 911 you could just walk up the pier and lay your hands on it which I did

I could feel that with a long enough and hard enough push one person could actually move an ocean liner

Nice metaphor for forcing change

Unlike Bob and the good Doctor, I gave up pushing and just walked away, says a lot about me and more about them

Like Bob I did thousands of tensile tests on Aluminum wheels

Like him I noticed a few very rare test bars that would go to 18 or 20% elongation yet still have excellent tensile and yield strength (typical was 3 to 6 % elongation)

As a degreed metallurgist I was taught that you trade strength for ductility - you can't have both

I noticed the absence of defects in the fracture faces of these exceptional test bars

And that's where I left it, until I started seeing Campbell's work and it all clicked

So here is a thought if you are enjoying reading research at the curly crashy edge of the wave

For $ US 105 you can join the American Foundry Society

One new benefit this year is free online access to the AFS Library, the world’s largest collection of metal casting research with more than 15,000+ papers and articles

For me this is pig heaven....

https://www.afsinc.org/sites/default/files/inline-files/2019 AFS Individual Member Application_5.pdf

Here is a sample of what I found in a few minutes that helped me solve a severe scrap problem

The best $105 I ever spent - can you even buy a single golf club for $105?

Atmosphere at the Mold-Metal Interface Scott, W. D. (2001

BAUME COMPLETE COATING CONTROL AFS Molding Div. Mold-Metal Interface Reactions Committee 4-F 2003.pdf

Comparison of Gas Evolution Results from Chemically Bonded Cores In Contact with Magnesium and Aluminum Melts Winardi, L. , Scarber Jr, P. , Griffin, R. D. & Weiss, D. 2008.pdf

Determining_your_binder_worktime.pdf

EMI Core Venting WhitePaper-2_ENG.pdf

Experiences Related to Mold-Metal Reactions in Various Casting Alloys Fischer, R. B. 1998.pdf

FOUNDRYMAN’S GUIDE TO SAND COMPACTION Scott, W. D. , Easterly, W. C. , Lodge, P. & Blackburn, C. S. 2004.pdf

FUNCTION OF THE STEEL FOUNDRY FOREMAN IN PREVENTING POROSITY Bull, R. A. 1934.pdf

Fundamentals_of_Sand_Reclamation_Updated.pdf

Gas Evolution and Permeability of Shell Cores in Contact with Aluminum Winardi, L. , Griffin, R. D. , Griffin, J. A. , Onda, H. , Harada, S. & Yoshida, A. 2008.pdf

GAS GENERATION AT THE MOLD-METAL INTERFACE Scott, W. D. , Goodman, P. A. & Monroe, R. W. 1978.pdf

Layer Porosity Effects in Aluminum-10% Mg Alloy Castings Pollard, W. A. 1965.pdf

Metallurgical Properties inside a Tilt-Poured Permanent Mold Structural Aluminum AlSi7Mg03 A356 Casting Chiesa, F., Carignan, J., Levasseur, D., Marin, G., Jutras, M. 2018.pdf

MOLD BINDER DECOMPOSITION AND ITS RELATION TO GAS DEFECTS IN CASTINGS Bates, C. E. & Monroe, R. W. 1981.pdf

MOLD-METAL REACTIONS IN FERROUS AND NONFERROUS ALLOYS Flinn, R.A. , Van Vlack, L. H. & Colligan, G. A. 1986.pdf

NO BAKE THE BASICS MCDP Staff 2018.pdf

Palmer - Mold vent Articles.pdf

POROSITY IN IRON CASTINGS FROM MOLD-METAL INTERFACE REACTIONS Naro, R. L. (000.pdf

Rapid Loss on Ignition (LOI) Using Conventional Oven Taylor, S. & Windardi, L. 2016 - Copy.pdf

RELATIONSHIP BETWEEN LOSS ON IGNITION AND GAS EVOLUTION Hoyt, D. J. 1994.pdf

VARIABLES AFFECTING THE FORMATION OF POROSITY DEFECTS IN IRON CASTINGS PREPARED WITH URETHANE BINDER SYSTEMS Naro, R. L. 1974.pdf

 

That is very tempting, but I have not finished reading John Campbell's book yet, so I guess there is not need to acquire even more reading material.

Here is a quote from Chapter 13 of John Campbell's book "Complete Casting Handbook, 2nd Ed.):

When I am asked, 'What is the most important feature of the filling system design?' the reply is "Everything".
As has been stated before, all gravity filling systems run with such high velocities that they are hypersensitive to any small error or mismatch, sizing or geometry.
A small, innocent-looking expansion or ledge leads to a stream of entrainment defects.
Every part of the system needs to be correct.

These recommendations are per John's book:

1. For a two-piece mold, the mold cavity should be located mostly (or completely) in the cope, with gates at the lowest part of the casting.

2. The weight of the casting should be estimated, and also the weight of the filling and feeding system, to arrive at a total poured weight. Divide the total poured weight by the density of the metal to get the total poured volume.

3. The fill system should fill the mold using a laminar surface flow.

4. Critical veolocity of the metal.

5. A choke in the system should not be used since its constriction speeds up the flow and creates a jet of metal that does not expand to fill the downstream channel, and so air is entrained.
The entire length of the filling system should be considered the choke, and the metal should be in continuous contact with the walls of the filling system, with the metal gently pressurizing the walls.

6. Gating ratios assume the the sprue exit area is 1.
Comparing the relative size of the runners and gates, and to keep the system completely full and slightly pressurized, then the ratio should be perhaps 1 : 1.1 : 1.2, or perhaps even better a ratio of 1 : 1 : 1 could be tried.

7. The runner area is usually selected to equal the sprue exit area.

8. For the gate area, there are some variables due to the fact that additional devices are needed to further slow the metal velocity, so the ratio turns out to be perhaps 1 : 1 : n, where n can vary.
The velocity of the metal in the ingates needs to be reduced by either a filter, a runner extension such as flow-offs, a surge control system, a vertical fan gate at the end of the runner, or vortices which absorb energy while avoiding significant turbulence.

9.
p = the density of the melt (were p = greek letter rho)
V= local flow velocity

inertial pressure = p*(V squared)/2

surface tension = lambda (greek letter lambda)

The critical velocity V (crit) at which the surface becomes unstable and starts to experience surface turbulence is 2*sqrt (lambda/r*p)

10. A sessile (stationary) drop of metal sitting on a non-wetted surface will have a height of h, which is defined by the balance between gravity and surface tension.

11. Most of the metals we cast have a critical velocity of close to 0.5 m/second-squared, or 19.7 inches/second-squared.

12. Critical meniscus velocity = V is a function of the head height of the pour.
V = the velocity of the metal in any point in the runner system.

V = sqrt(2*g*h)

where g is a constant equal to 300 inches/second-squared
and h is the head height of the casting
and this assumes a friction-free filling system.

13. The fill time (time required to fill the mold with metal) should be determined by the design of the metal delivery system with the pour basin full at all times.
The person pouring the metal should not have any control over the fill time, and their only task is to make sure the pour basin remains full while the mold is filling.


.

 

?
Stated another way, don't put the pattern in the drag and the runner and gates in the cope, else the metal will run downhill like a waterfall when it fills the cavity, and churn up a bunch of air and bifilms.

Fill the mold cavity from the bottom upwards.
.

 

Ignore bifilms at your peril.
Its your castings, and if you ignore it and get perfect castings, then its all a bunch of bunk.
But I think I will at least use John's calcs as a starting point, not that they guarantee the casting will turn out well, but as base line approach, and a point of departure for determining what to tweek to get it working right.
Otherwise it is just a bunch of random stuff, and no two castings will necessarily be the same, which will be perfectly fine for most of us here, but definitely not a wise approach for some of us.

.