Cast iron porosity.

[airmodel]

I know that air vents are used in sand molds to prevent porosity so in this video plenty of air vents were used in the sand molds. The sand molds were made in a backyard setting and sent to a professional iron foundry to be poured. Watch from 47:39 to 49:34, this is a hour long video. The castings are the worst I have ever seen! Clearly the air vents never prevented the swiss cheese castings from happening so it looks like a metal-mold reaction where excess gas bubbled through the casting.
<a href="https://www.youtube.com/watch?v=L3gzFhcXMWo" target="_blank">http://www.youtube.com/watch?v=L3gzFhcXMWo</a>

[CI]

I use identical sand and resin binder that she uses.

I see several things that could cause problems.

Those type molds should be flamed with a soft propane flame to burn off an uncured resin that is on the surface.
I did not see any mold flaming in the video.

I spray on an alchol-based ceramic mold coat, light that to let it burn off, and then flame that too with propane.
I did not see any mold flaming after they applied the graphite.
Not flaming the graphite can cause terrible gassing/bubbles in the casting (don't ask me how I know that, but I have photos that look like the surface of the moon, from the early days).

She mentions the foundry doing a terrible casting job, but to be fair, she is the one who laid out the spure, runners and gates, and from experience, using a sprue/runner/gate system like that can cause those problems.
I use a sprue that has a smooth transition into the runner(s), with spin traps at the ends of the runners, and thin rectangular gates at the top of the runner.

When the sprue is initially filling, there is a point where both air and iron are flowing down the spure.
Once the spure is full, and if the pour is not interrupted, you have only iron flowing down the spure.
If you don't use spin traps at the ends of the runners, then that initial mix of air and iron can flow anywhere, and really cause a lot of damage.

With the spin traps open to the top of the cope, the initial flow will go into the spin trap along with the initial aspirated air, and any loose sand.
When the level of metal rises to the top of the runner, then and only then should the clean iron start flowing into the mold cavity.
If you use dead-end non-vented runners, the initial air/iron mix will strike the end of the runner, and bounce back into the runner system, churning more air into the iron.

The thin rectangular gates on top the runner also help skim an slag that may be floating on top of the iron, as iron flows into the mold cavity.
She does not use thin rectangular gates.

I use a small small vent hole at the highpoints of each mold cavity, with the vent extending out the top of the mold.
I forgot these vents on one mold, and had two very large bubbles in the casting at the top of the mold.
Resin bound sand is not very permeable, and the high points of the mold need small vent holes.

I think with spin traps at the end of the runners, these defects could be avoided.
I don't have defects like this, using the exact same sand and resin systems.

I don't think this was a venting problem, I think it was a runner with no spin trap problem.
You can also cause a lot of defects if you even slightly interrupt the pour, since that will aspirate air, which can cause defects like this.

That molding sand and resin are great to use, and make very accurate molds that are durable when handled.
I use a slide hammer just like the one they use, and that is why resin-bound molds are so accurate, because the pattern pulls straight out of the sand, and there is no gap between the pattern and the sand.

That is my slant on what went wrong with these otherwise very nice molds.

.

[vtsteam]

Thanks airmodel! 

[airmodel]

I spray on an alchol-based ceramic mold coat, light that to let it burn off, and then flame that too with propane.

For many years I used S.S. sand on thin sectioned castings, an alcohol based mold paint called "MOLDCOTE" by foseco, it was a zircon flour with a binder. Never had any problems with it and all I did as the instuctions said was use a match to burn the alcohol off and the mold is ready to use. Some time later I had to cast in iron a headstock for a small lathe in greensand. It needed two cores where the casting is very thick (30-35mm). Cast iron does burn on in thick sections so I used some of that mold paint so the cores could be removed a lot easier. Machining those bores revealed plenty of blowholes close to where the paint was. If more was machined the blowholes disappeared. To me it was puzzling as to why it worked for thin castings but on thick castings it did not. Thick walled castings take a lot time to solidify than thin castings and has plenty of time for the metal mold reaction to happen.

As to why cast iron gypsy castings failed I will give my two cents worth.

Most of those molds she made were thick castings and needed a coarser sand so the gases can escape through the sand instead of bubbling through the casting. It could also be that the pouring temperature was too hot making the thick sectioned castings take a lot longer to solidify and make the problem worse. In hindsight this is what should have been done using a coarser sand for the cores in my headstock casting.

[CI]

They use that same sand and resin at the local Metal Museum, and when they have a big event, they literally pour a LOT of iron.
The sand is OK85; not sure if that means mesh 85 or what, but it is pretty fine and apparently kiln dried sand, all round grain, and all the same grain diameter.
Very dry sand is required for the resin to work correctly (per Ask Chemical, the company that makes the resin Linocure).

I have been to a number of iron pours at the Metal Museum, and seen perhaps 100's of their resin-bound molds, which are OK85/Ask Chemical molds, including Laura's molds.
I have never seen a problem with having the sand too fine, reflected in the castings they make, at least it not obvious to me that it is an issue.

You can get a pinhole in a bag of OK85, and the entire bag will have emptied itself onto the floor by the next morning.

It sure would have been nice to see a video of the pour for those castings.
That would perhaps shed some light on things, with something obvious like an interrupted pour, or a pour that had slag floating on top, etc.

Laura casts some fantastic stuff in iron; she has talent.
She has a very large cupola tatooed on her upper leg in front, which is quite impressive, but my wife does not like me to notice things like that.
The art iron group people are very passionate about what they do, perhaps because many of them sell it as artwork.
And the art-iron folks are the nicest people I have ever met, and will stop what they are doing to answer questions or help you with what you are doing.
Those big multi-cupola iron pours are wild.  If you have not attended one of those, you should try to find one.

Our member "airmodel", aka "Luckygen" has done a lot of cupola iron pours.
He is the king of hobby iron in my opinion.

Edit:
The most impressive cupola tat shown below.
LOL, don't tell Laura I posted this; it can be our little secret.

.

[CI]

This is one of the castings made from the molds in the video above.
The surface irregularity is quite odd.
I will try to go dig out the photos of where I used graphite and alcohol, and did not get it all burned out.

This almost looks like what airmodel mentions, the pour was too hot, but OK85 is not prone to burn-in, at least not with cupola iron.
They do use an iron-rated pyrometer at the Metal Museum, to avoid poring too hot.

The thing about metal burning into the sand, is that it produces an additive to the castings.
These defects are negative, and into the face of the casting.
It is a bit odd that it does not occur on the perimeter.
Could have been the graphite alcohol gassing though.

.

.

[CI]

This is a bearing cap I cast using identical resin and sand.
I did not have time to do much work on the pattern.
He wanted to keep the original letters, so I patched up the broken casting, but left some blue tape on it.
You can see the blue tape mirrored into the sand very accurately.

This mold had a coat of "Velacote" (tm) ceramic mold coat, which was burned off prior to pouring.
Not any odd defects like the one's Laura experienced.

I don't see any vent holes in this mold.
Normally I vent the highpoint of the mold.
Edit: Actually I do see vent holes in the cope; last photo, mold on the right side.

.

[CI]

The bearing cap casting.
I did not know about spin traps when I cast this piece, but luckily it still turned out ok.

You can see the iron was quite fluid, since it rose up and out the vent holes (2nd photo).

.

[CI]

This was the 2nd pour I ever made, and I made a lot of mistakes.
I was trying to use oil-based sand, and sprayed a mixture of alcohol and graphite onto the drag mold, and did not get it all burned off.
I would not suggest spraying alcohol onto oil-based sand anyway, but I had no casting experience, and so learning as I went.

The alcohol flashed and created huge bubbles.
I see some sand inclusions too.
This is a great example of how NOT to cast metal.
Almost everything that could go wrong did go wrong on the casting.
One has to start somewhere, and mistakes will be made if you are going to learn the process.

Also are photos of an ingot that was poured right after this casting was poured.
Note: Tall risers like those in the photos are not necessary; don't copy that.
No bubbling on the top of the casting because the alcohol and graphite were not used on that side.

Comparing this casting with Laura's shown above, it seems possible that her alcohol and graphite gassed.
I think the problems that showed up outside of her casting were from entrain air in the iron.
.

[CI]

A camelback and straight edge I cast.
No problems with sand, etc.
Same resin-bound sand.

There was some odd surface wrinkling, I am not sure why, but that was superficial, and the edges/surfaces machined well without any inclusions or hard spots.

The patterns were made by a buddy of mine, who needed them cast in gray iron.
.

[airmodel]

It sure would have been nice to see a video of the pour for those castings.
That would perhaps shed some light on things

I found two videos that shows what happens when a metal mold reaction happens.In the first video 6:08 to 6:48 the pouring cup backs up higher than when it was first poured and bubbles constantly until the copper solidifies. 10:26 to 10:50 Drag surface shows lots of wrinkles caused by the bubbles.
<a href="https://www.youtube.com/watch?v=aMejcz3hUFg" target="_blank">http://www.youtube.com/watch?v=aMejcz3hUFg</a>
The second video shows no bubbles when poured. If you thought the bubbles were bad in cast iron watch this video from 3:02 to 3:14 no bubbles at the pouring cup 4:16 to 4:26 plenty of bubbles in the casting!
<a href="https://www.youtube.com/watch?v=LAEyVTAtyJA" target="_blank">http://www.youtube.com/watch?v=LAEyVTAtyJA</a>

[airmodel]

There was some odd surface wrinkling, I am not sure why

If there is wrinkles on the surface on your castings it means that a small amount of metal mold reaction has happened and there may or maynot be bubbles in your casting.

[airmodel]

In the original video have a look at 49:33 to 49:38, the square pouring cup is full of bubbles but as the cast iron flows to the ends of the casting the surface is smooth with no bubbles. This means that that the cast iron has cooled enough by the time it gets to the ends of the casting. Classic case of too high a pouring temperature. There is also a possibility that she put too much of the three part resin which caused the bubbles to rise through the casting.

[Jasonb]

I don't know what foundries policies are when casting into other peoples mould? I would have thought most have a disclaimer as they have no idea of what is going on inside in relation to sprues, gates, runners etc. Also as they do not have the patterns they have no option to cast again if there is a problem.

I did see them flaming off the mould wash so don't think that can be blamed although brushing rather than spraying could leave heavy areas in some places.

Pat mentions excess resin causing more gas and I would imagine it also fills any spaces between the gains more, combined with glueing the two halves together does not leave much room for gas to escape. Bigger bulk of iron is also going to burn deeper into the sand causing more gas than a fine thin part so would flaming off the resin have gone very deep or just affected the surface.

It would have been interesting to see those parts saw through to see if it was right through or just the upper surface.

At the end of the day it may well have been a combination of several factors. Maybe she will mould them again in exactly the same way and get them  cast as per her usual source which would be the only real way to see if it was foundry or not.

[CI]

I don't think she used too much resin.
She has mixed hundreds of molds, and knows exactly what she it doing, so I don't suspect that.

Bob Puhakka has explained much of these problems in his videos about aluminum casting; much of the same items apply to iron castings.

In the photo below, the iron will strike the bottom of the pour basin and curl up and fold back onto itself before the spure and basin are full.
This churns lots of air into the metal, and is a very predictable problem.

Bob Puhakka mentions that the you must have a smooth transition at the bottom of the sprue into the runner system.
Bob uses various methods to eliminate the air that gets churned into the metal before the spure is full.
Spin traps at the ends of the runners(s) is one way to separate the initial air entrained in the pour.

It is easy to avoid these defects, but most people use old-school sprue/runner gate system, with no spin traps, and gates that are not at the top of the runner(s).
This is why so many castings have defects; the fundamentals that are required for good castings are ignored.
You can see the splashing that happens inside the basin and runners in some of the mold fill simulations that can be found on ytube.
I don't need a mold fill simulation; I understand what is happening and can envision it easily without a simulation program.

The molten metal flow must be laminar.
You cannot just dump the metal in and have it get churned like butter, and expect to have a good casting.
Stevie Wonder could see what will happen.
Garbage in; garbage out.

.