Bob Puhakka on Bifilm theory

Based on my understanding of bifilm theory it should have, which was my point. Even though I made it incorrectly and expanded the size of the runner at that point there is possibly enough leeway in the physics to allow backyard casters to realize improvement in our castings. Or maybe I was just lucky this time. The expanded area that was greater than the sessile drop height could have entrained air breaking up the solid oxide surface film and creating disjointed pieces that could be folded in the turbulence and create what we view as porosity in the casting. I don't think it did. I couldn't see anything even under the microscope other than the inclusion I noted.

I think you are correct about the offset basin in that it just provides a smoother feed to the sprue. This is seen in Oldfoundrymans video, but directly from the crucible.

I just got the microscope, I'll didn't see a magnification, but I'll check.

 

You can say that again, but with one big difference, Bob is pretty much on the genius level in my opinion with his casting knowledge, and he can back up everything he says with real world castings and multiple large castings.
Throwing out a bunch of theory is one thing, but putting that theory into use and producing mass quantities of flawless castings is an entirely different thing.
John Campbell credits Bob Puhakka with proving that his theory actually works in a production environment.

The introduction of John Campbell's book says:

To Robert Puhakka
For his dedication to the 10 rules and the living proof that they work.

But if you watched Bob's foundry tour videos, he conceptualized every single aspect of every part of the process, done all the calculations, written computer simulation programs, etc.
Bob has beaten the foundry guys at their own game.
And like Bob says, he comes from very humble beginnings, such as cleaning up shit in a mine.
It is really a Cinderella story.
The words of the song "In my own little corner" are quite prophetic for Bob, as follows:

But I know of a spot in my house
where no one can stand in my way.
In my own little corner in my own little chair
I can be whatever I want to be.
On the wings of my fancy I can fly anywhere

And indeed, Bob did his own thing, and would never been able to accomplish that under a corporate structure.
Bob built his own organically-grown company from the ground up.
If you have every tried to build even a little one-man company using just one computer (as I have done), you realize the magnitude of what Bob has built.

.

 

An interesting part of the story about Bob Puhakka is that he apparently was in the testing business (according to Bob), and he saw first hand how screwed up the cast products were, and realized that he could use John Campbell's 10 rules in order to build a modern casting foundry, and he did just that by going out and buying a foundry and rebuilding it from the ground up with both equipment and people.

Bob mentions sending out flawless castings to a customer, and the customer remarking "We don't even have to repair the castings after we get them", ie: the customer does not have to weld up all the bifold cracks.
Bob says that customers have always accepted sub-standard castings because they were told that was the best that any foundry can do under the circumstances.
That is why all the foundry guys hate Bob; because he has proven that you don't have to produce crap castings; you can actually make flawless castings, every single time, just as John Campbell did with the Cosworth process that he invented.

Edit:
Bob Puhakka has not only rocked the casting world boat, he has turned the boat over, and the existing foundry guys are swimming around like rats that have just jumped off the Titanic.
The question is will the rats swim to Bob's boat and climb up onto it, or just swim around and eventually sink to the bottom.
From my experience in corporate, the rats much prefer to go to the bottom with their ignorance intact. Corporate guys will never admit they could be wrong about something. They would literally rather be dead; that is a fact.

.

 

Here is one of the reasons I subscribe to his theories.
This is one of the wheels I cast for the '41. Not my best castings. The risers and sprues I put in the melt yesterday looked this bad. The spots are not hydrogen or they would be spehical. I believe they are bifolds. So where did these bifilms go when I remelted the gates and risers yesterday. Could a higher temperature melt them, or a reducing atmosphere convert them back to alloy? I didn't do anything to remove hydrogen from my melt, so it is still there, just not visible to the naked eye.

 

Bob does have quite an attitude, and it can take a while to get some sort of understanding of it all.
Bob is basically an "outside-the-box" thinker, and he has succeeded and prospered greatly by doing this.

I think Bob and John Campbell have been really bothered by the fact that most of the foundry industry have ignored them completely, and as Bob says, the industry is controlled by people who don't give a rats ass about casting quality, such as academia that are after grant money, sales guys who just want to sell a product with no regard as to whether it actually works, the lazy people who run foundries who don't want to take the time to change their process no matter how many improvements that may offer, the fat cats who offer "financial services", but can't even control the food that flows into their mouth (Bob's words), etc.

So unlike most people, Bob says exactly what he thinks, and what he thinks is pretty much right on the mark.
Bob also admits that he is using the over-the-top approach now because the easy going approach has not worked for so many years.
Bob said he does things (videos, presentations, etc) so deliberately over the top that nobody will ever forget them, and the pro wrestling analogy is a good one.
Nobody would ever remember pro wrestling if they were dainty and kind to each other. It is those absurd body slams followed up by several hits in the head with the nearest chair that we remember so vividly.
The casting industry desperately needs a few body slams and many many chair hits to the head.
The casting industry will be lifted by Bob and thrown completely out of the rink if they are not very careful. Its is already happening; Bob has shut down a number of his competitors already, and has plans to expand into the steel industry.
Bob does not appear to be a big guy, but I think he is a heavyweight in the casting industry, and he is not going away any time soon.
.

 

In one of his videos, Bob shows the ingots that he uses for his castings, and discusses the bifolds that are in the ingots already.
He mentions that he works with the ingot folks to minimize the bifolds and maximize casting quality.

I have also read that the aluminum in a large foundry was allowed to sit in a large crucible for a while before it was poured, and when they tried to speed up the process it stirred up the metal in the crucible to the extent that all sorts of defects began to happen in the castings (defects that they had never had before).
So allowing the metal to settle in the crucible may be important, as well as being sure not to stir the metal and thus churn bifilms into it.

Morgan also makes a bottom-feed crucible spout, and that is a consideration with aluminum since it would allow the metal to feed into the mold without bringing in the garbage that floats on top the melt.

.

 

Still, where did these oxides go? Is that the bit of crud I scraped off the surface?
As Bob mentions, you can't really see molten aluminum because it is under the aluminum oxide layer. When he skims that big furnace surface it immediately films over.

 

the man in the back asked "what is the taper angle of the sprue".



I have to perform a few sanity checks on this yet but, I think this is correct.

 

It appears what you are showing is a method to calculate the included angle given the top areas and bottom areas for round sprues.

The question I have is what angle would be appropriate given the bottom area and any top area required to give that ideal angle.

I arbitrarily chose 5 degrees per side through ignorance, primarily to allow easy withdrawal from the sand.

 

Still messing with the equations.
I'm working with cylindrical values because, for me, the math is easier.
When it comes time to calculate the sides of the sprue/runner/gates, etc. just remember
Area=pi r sq.=length x width

It's looking like the level in the basin has a big effect on the angle.
The angle varies slightly with the height of the sprue.
The area of the sprue has a somewhat linear effect on the angle but not a whole lot.

The area of the sprue bottom is going to be a function of the mold volume. Still working on that equation.

You know, all along it's been 5 deg. In the vague recesses of my mind this was the ideal declination for transporting water over long distances.
The math is showing me otherwise. We'll see.

 

I tried to upload a spreadsheet, no-go.
user inputs are columns A-C
calculated values are in columns D-I

Column
A sprue height
B basin melt depth
C sprue bottom area
D runner velocity =SQRT((A3+B3)*(772))
E sprue bottom radius =SQRT(C3/3.1415)
F sprue top area =D3*C3/H3
G sprue top radius =SQRT(F3/3.1415)
H sprue top velocity =SQRT((B3)*(772))
I sprue included angle =2*ATAN(E3*(SQRT(D3/H3)-1)/(A3+B3))*180/PI()

No guarantees this is exactly right. I'm just putting it out there for review and implementation.
It's a fine day for casting.

 

I edited a correction in post #252

D runner velocity =SQRT((A3+B3)*(772))
I sprue included angle =2*ATAN(E3*(SQRT(D3/H3)-1)/(A3+B3))*180/PI()

 

With the left over iron from the lost foam pour I rammed up the core plate using the same runner and gate setup, but added two vents out the far side. The result looks like cold iron, but the feed that broke off in the sprue was only about 3/16" in diameter. I either bobbled, or ended up with some sort of blow, because the surface at the base is smooth and sort of circular. Then what looks like tracks run from each gate across to the vent and between them the edge cold shut. If the last of the metal was running through that small stem, it is impressive fill for no more height than I had.
For cast iron the sprue and runner should be much smaller based on cast iron's sessile height which I measure at .315". That is 60% of the aluminum number.

 

That is a pretty interesting fill pattern.

 

I recreated a spreadsheet using the above formulae. I only changed pi to 3.1416 (3.14159) because I couldn't help myself, and changed 772 to 773 (772.8) and that does not make any difference. Looks like you did a good job.

They are pretty straight forward, and I assume they are from Puhakka or are they from Campbell?

So the formulae basically assume the velocity at the bottom of the sprue is the same as a body free falling from the top of the liquid. I guess it's a reasonable upper limit. In my straight sprue system I calculated 35 in/sec where the formula calculates 52 in/sec. They mention reducing flow rate by a friction correction factor. To me that is a significant difference in trying to limit air entrainment.

The pouring basin velocity at the top of the sprue is also calculated as though it were a falling body the depth of the liquid. Sensibly, the fluid will be flowing very slowly as it approaches the sprue, that's the design of the pouring basin. I'm trying to grasp why they are assuming it should be faster. If it is assumed to be slower then the top of the sprue will be larger, which seems to be the design to make sure the liquid stream can't be pulled in two by liquid falling faster below it. Any other ideas?

I guess it doesn't matter, the designers of the system have chosen these parameters and have found they work. Although I'm sticking with rectangular sprues and runners, I really appreciate you putting the equations down and I've already figured out my current tapered sprue is too small at the top, per the equations. It will be more that five degrees but I tapered it only in one plane, it is constant thickness the other direction.

Also, using a friction correction factor, 0.6 in my case, results in the runner velocity I observed and at that runner velocity the top of my sprue is the right size.

Very interesting.

 

You should list the units for each variable too; perhaps it is obvious to some.
That is how I check my answers.
If I get the wrong units in the answer, the numbers don't matter because they are wrong (EIT exam trick; solve for the units first, and if that turns out right, work backwards and get the numbers).

I guess it is as follows:

A = inches
B = inches
C = square inches
D = inches per second ?
E = square inches
F = square inches
G = inches
H = inches per second ?
I = degrees

 

I guess even if the equations are used to find the wrong values, those values could be used as a benchmark, and a reference point to tweek the dimensions of the delivery system.
So once you found a system that worked well, then look at the numbers for that system, and use that information as a relative benchmark?
Is this a reasonable assumption, or is it more reasonable to assume that if it is a little off, then it is way off?
.

 

PatJ: Your assessment of the units is correct. I used 32.2 (or 32.174 from memory) ft/sec2, which times 12 gives the 386 used, and worked through the units.

 

They pointed the way, I just derived the formulas from the Bernoulli equation then checked those formulas against what Puhakka and Campbell were saying.

Yes, I think the calculated velocities would be max. values since friction is ignored. There is a friction factor which varies from 0-1 that can be included in the velocity equation but, I haven't looked at it much. I'm assuming its 1.

Puhakka talked about a gate velocity correction factor that is used to determine gate area needed to slow the velocity down. It's the ratio of runner velocity/sessille drop velocity. So, there should be a taper from runner area to gate area.

BTW - I was tired during my last edit and made an error.
column I sprue included angle =2*ATAN(E3*(SQRT(D3/H3)-1)/(A3+B3))*180/PI() should be corrected to:
I sprue included angle =2*ATAN(E3*(SQRT(D3/H3)-1)/(A3))*180/PI()