Pouring soft Cast Iron

I'll open this thread to pass along the method I use to pour machinable cast iron.

It may take a while, I'm rusty.

 

. I started carving the split pattern for the 1/3 scale smallblock in my thumbnail in 1997 and it first ran in 2004. The block, heads, water pump and oil pump are cast iron. This is casting number five and was the first casting to pour out completely, but it had some hard spots that were remedied by taking it to a heat treat outfit in Oakland. It took two attempts. My understanding of what they were able to do is heat it without destroying it to a temperature that allowed the combined carbon to convert to free carbon in the form of graphite. I spent a couple years building the short block and then it was time for the heads.




The patterns and coreboxes for the heads took a full year and since I needed to machine the valve seats and bore the head bolt holes soft iron was the goal. I was following C W Ammen’s advice and had salvaged a ton of steam radiators from a demolition project. This has a high Phosphor content so it is very fluid cast iron. The challenge with the miniature castings is that the thin sections do not throw off enough free carbon during solidification if there is not sufficient total carbon. Adding Ferro Silicon as an innoculant has the effect of displacing a portion of the total carbon and casting it out as free carbon in the form of graphite. This process is affected by pouring temperature and section thickness.


In my search for solutions to my casting problems I wandered into AB&I in Oakland and asked for some help. I was given a couple chill wedge molds from a pallet on the foundry floor and told to pour them to adjust my melts. During another foray to a local foundry I was given a coffee can of Desulco which is a carbon increaser in the form of granules. These are the chill wedge molds. I made some of my own with both lindseed oil bonded sand and waterglass. whenever I had left over core material I make a chill wedge mold. I later eliminated the pouring basin and just use a wedge.



On the left is the Budget Casting 75% Ferro Silicon and on the right Desulco carbon increaser. I believe it is the 9001 , but do not have the bag. They have a decent website.



The last piece of the puzzle was solved at a junk shop in the form of a cast iron optical pyrometer so I was able to confirm pouring temperature.


The procedure for melting in the oil fired furnace is to charge the crucible with the broken steam radiator chunks, Oyster shell, and the carbon increaser. I pull and slag at 2600 to 2700 degrees and pour a chill wedge. The crucible is returned to the furnace and the wedge is cooled and broken. The thickness of the wedge where the white iron starts indicates that areas of the casting are that thin and that cool at that speed will be hard. The casting might be saved by heat treating, but it is easier to continue the heat and add the ferro silicon innoculant to the crucible just prior to pouring. I pour another wedge when I pour the casting to determine if the additional temperature or the addition of the ferro silicon had the desired effect. The upper photo is all soft while the lower shows the white iron at that thickness.

 

Very interesting post, thanks

 

Yes, thank you!

How much Desulco do you add? It goes into the crucible first? If you add excess increases it just burn off? To be honest, I am surprised it does not just burn off before the iron melts since the iron can’t absorb it in a solid state and the temperature is so high prior to melting.

Same questions concerning the oyster shell except the burning comment.

If your wedge shows white iron, then you add FeSi. But if the wedge shows no white iron, you skip the FeSi? You do not add any Desulco late in your melt should you see white iron?

Denis
.

 

Does anyone know what problems might occur if on tried to use crushed common charcoal briquettes as a carbon increases? Too much sulfur or other contaminant. Burns off worse?

Denis

 

One other comment is that you seem to confirm my suspicion that higher pouring temps reduce white iron formation. That makes sense as raising the temperature of the metal/mold to a higher initial temperature should result in a slower transit through the critical transition temperature where chill occurs. One downside to the higher pour temp, at least in my green sand might be poorer surface finish.

Denis

 

I add about a half cup to a #8 crucible that I fill to the rim loosely. The oyster shell is mixed in with the broken cast iron. Some may burn off, but it forms a cover that I plunge added charge pieces through. At our temperatures graphite does not melt and the cover stays granular.I have used charcoal briquettes crushed with the oyster shell in the past and that was my standard carbon additive for years. I slag the melt with Borax and that turns the slag cover to a gummy mass that is pulled over the lip of the crucible.
When I poured the Indian heads I used the ferro silicon because the fins were so thin.

 

Denis, I've been rethinking the carbon additive. I'm convinced it helps and the chill wedges prove that I am able to reduce the white iron. The primary use seems to be adding carbon to arc furnaces that are charged with steel scrap. Some of the sites I have looked at mention that graphite does not melt, but rather goes from a solid to a gas at a temperature more likely to be seen in that type of furnace.
My main benefit has been machineable iron so I'll take that. More research may be needed.

 

White iron is a killer in our foundry. If we fall off on a couple of processes we have made scrap. Our cleaning process usually insures a thorough beating and tumbling and slamming of our delicate castings so only the toughest survive. White iron usually does not make it through our process. We have tried many different types of inoculants and the in stream works the best for us...however I do suspect that there are times it causes inclusions when used improperly or if its contaminating the sand system.

 

I am still trying, please bear with me, to figure out how added carbon in the form of charcoal, nearly pure carbon in the form of Desulco, or graphite powder does not combust if it is sitting in the bottom of a crucible that necessarily reaches at least 2550F. By reading above the oyster shell evidently is not spread as a cover over the layer of graphite. So, why doesn’t the graphite, whose combustion temp is in the neighborhood of 1200F, just burn away long before iron starts melting? To actually increase carbon content must the increases be used as a late-stage inoculant rather than an in-the-crucible additive?

Denis

 

That is why we add in stream (as mold is being poured) as a late addition.

 

Very interesting.

That is some nice work there.

I have some doubts about this too. Not saying that it doesn't work, but hard to understand how it works. Do you mean that you literally sprinkle the graphite on the metal stream as it is going into the mold? For me this raises some questions. Can you be sure that all of the carbon will be dissolved and some of it not remain in the casting as an inclusion which would be regarded as a defect? Can the carbon really dissolve that fast and distribute evenly into the mass? If you used charcoal as an additive like this, there would probably be a lot of volatile compounds other than carbon, that will burn and create gas as it goes down which doesn't sound like a good idea.

My idea of increasing the carbon content was roughly that after getting the iron somewhat melted (as soon as the mass starts to resemble a pool), you would add in a bunch of charcoal or graphite, and keep adding it as it burns away. Some of it would then be dissolved into the melt as long as it doesn't burn all off. I have not tried it yet so I don't know if it works.

 

I told you guys I was rusty, and WD40 didn't help.

I charge the crucible with iron, oyster shell, and Desulco or charcoal. Heat to 2600 degrees adding iron to fill the crucible. Pull, slag, and pour a chill wedge. Return the crucible and continue to heat. Break the chill wedge, read it and decide if it will require Ferro Silicon added to the melt. This is done just before pouring. Wait 20 to 30 seconds to dissolve the chunks of FeSi that have been added, and pour.

Everything I have found regarding graphite is that it has no melting point, but rather turns to gas at 6250 F. This explains the ability to withstand iron pouring temperatures. I have a mask that I lay on the mold and spray with graphite to improve the surface finish of the heads.

So as I mentioned yesterday I am not sure how much of the carbon gets added from the Desulco, but I will continue to use it because something worked. I just cannot be certain as to why.

 

I had some charcoal and oyster shell mixed so I took a shot beside the Desulco. I break it down a little smaller, but that is the ratio I use. Ten to one?

.

 

You have to have oxygen to combust. I would think a bit of charcoal would burn but then CO and CO2 would keep new oxygen out, as well as the labyrinth of a full crucible.

 

Does carbon actually dissolve or does it (like in pack case hardening) diffuse into the melt?

 

The above are good questions and likely very hard to answer. Given the vigorous swirling the hot oxidizing atmosphere of a combustion furnace (as opposed to electric furnaces) I have to say I am skeptical of the labyrinth aspect of the charge effectively excluding oxygen from the region of hot Desulco or charcoal. BUT, one very favorable characteristic of charcoal, and likely Desulco or graphite, is a very slow burn rate. How else would my grandpa made those incredibly delicious slow roasted ribs that he cooked all day on certain very special Sundays.? So, my suspicion is the charcoal/Desulco/graphite persists through the first 30 to 40 minutes of melt despite the oxidizing atmosphere and then starts dissolving into the iron as it drips down onto the carbon source sitting in the bottom of the crucible. With my next melt I’m going to add in some charcoal (until I can buy some Desulco) plus oyster shell as outlined above by ESC.

Now, my next question is when is someone on the forum going to buy a spectrometer so we can actually measure what we are doing on this and similar questions?

Denis

 

I did write to Tom Cobett concerning use of charcoal/graphite for optimization of iron softness. He is pretty busy right now and has limited time for forum responses and gave me permission to quote his respone on the forum:

"For machinable cast iron you need:

Low copper and chromium content iron. No alloy iron. Thin pieces like manifolds will be low in copper and chromium.

High carbon content iron. Melt it WITH charcoal. Some of the charcoal will burn to create a reducing atmosphere (no oxygen near the iron). That way you do not burn out carbon and silicon already in the iron. Also, Carbon takes a long time to dissolve into iron. Put it in with the iron scrap.

Add a flux while melting to also reduce oxidation of the iron. Equal volumes of borax and crushed limestone will work well. The slag should be liquid and alkaline. The alkalinity will prevent the formation of iron silicate.

Inoculate with granular 50% ferrosilicon just before you pour. About 1% by weight of the iron should be enough. This will promote the formation of flake graphite. If you do not add the ferrosilicon, you are likely to get white iron with no flake graphite formation. Can’t machine white iron. It’s like glass.

Tom Cobett"

I then asked him about ferro silicon and he said in response to my question:
My note:
"Thank you, Tom! I worry that I will exhaust your patience with my seemingly endless questions. Your taking the time to answer is appreciated. Is it OK if I quote your comments to the forum? I think other folks might benefit as well.

Incidentally, I’ve been using a much small ratio of FeSi in my CI melts: 1 ounce per 50 pounds of iron. This was based on the practice in a local ferrous foundry.
Did I get good info from them or not? Are there any studies out there showing an optimal ratio?"



Tom Cobett
5:51 AM (2 hours ago)


to me:


"Using more FeSi will give you better fluidity from the increased Silicon.

Yes, feel free to post on my behalf.

Regards

Tom Cobett"

My thanks to Tom,
Denis