Making Ductile Iron in the Home Foundry

Recently I started chatting in another thread about making ductile iron. In hopes of
A) Making any information more easily found in someone’s future search and
B) Improving visibility of the this subject to more people
I am starting this stand-alone thread. I was not quite sure where it belonged. Mods please move it if it should be somewhere else.

I am interested in making some ductile iron to cast a lever for my metal working lathe. And over and above this simple project I have been interested in knowing how to make ductile iron and in doing it safely. I am hoping folks here can fill in a few blanks for me as there seems to be very little on the Internet concerning the specifics of inoculation of grey iron to convert it to ductile.

1) ,I think magnesium ferrosilicon is the nodulizer most commonly used. Is that true? If so, where can I get some. Pure magnesium bar is easy to come by, but I have found nothing about it being used to make ductile.

2) How much is used per pound of iron?

I know that some foundries us a steel “cover” method of innoculation in which nodulizer (MgFeSi?) is placed in an empty crucible, a layer of iron pellets or coins is poure over that to act as a temporary barrier, and finally grey iron is poured over the top. Then a significant reaction occurs with smoke and light and maybe#plashing iron and voila! Ductile Iron.

I have also seen video of a “dump method” of adding in occupant:


This would be best for my setup I think. I have a plan to do this somewhat remotely and safely.

I do not think that simply pushing metallic magnesium into the melt will do the trick as I have found no patent methods suggesting this nor any other reports of adding metallic magnesium.

Any help on any of the above points would be appreciated.

Denis

 

I watched the video I linked and noted there was some distance from the chute they used to the surface of the melt. Of course, there is no way to know if that was a desired feature or simply expedient and there is no way to know what material is being dumped. Since there was some drop from the chute to the melt I was theorizing that it might be desirable to have the inocculant being "thrown" into the pot rather than just skimming onto the surface. It sounds like, from a ofline conversation, just gently sliding it on is good enough.

Adding the inocculant without being branded by flying iron in the process would be a good thing. When I thought some "drop" was needed, I drew up the following crude outline of method to do that safely. If just sliding the NiMg3 onto the melt is adequate, then no "dog-leg" is needed and a shorter length of pipe or simply a tipping cup on a shaft could be used to add the mag. I do think for me to be 15 feet away and to be standing behind a shield when starting the reaction would make negligible the risk of being hit by molten iron.




Denis

 

The chief advantage(s) of the nickel material is the moderation of the reaction (?) and the adding of Nickel. This last supposedly helps the resulting iron be "tougher" - ductile, or otherwise.

I bought a kilo of nickel a while back. I could send a few chunks up your way.

 

Thanks for your kind offer, Dennis.

I think I need to work with a known inocculant to be added in an amount that has been previously worked out. It is unclear to me what the result would be if I added metallic Magnesium and metallic Nickel as compared to NiMg3. On the other hand, if you or someone else has knowledge of this being done and the proper proportions of ingredients, I am all ears.

Denis

 

I'd been doing some reading about ductile iron, the paper linked below compares silicon carbide and Zircinoc ductile iron innoculant and mixtures of the two and concludes that 0.3% silicon carbide can replace Zircinoc as a ductile iron additive for the first of two innoculations into the melt.

The strength of 0.3% SiC is comparable to ductile iron and the hardness is middle of the range but close to the other samples. The microstructure photos of the silicon carbide innoculant only (sample 1) show larger nodules but half the number of the zircinoc innoculant: 100 for SiC vs 200 for Zircinoc. These guys are melting a mix of 40% steel and 60% ductile iron scrap in an induction furnace and except for silicon carbide (sample 1) they are innoculating twice.

So it may be possible to make ductile iron with 0.3% silicon carbide added halfway through the melt so it has enough time to dissolve fully. The paper also mentions the desired particle size of the silicon carbide and how soon to add it to the melt to ensure it dissolves in time. Silicon carbide will be a lot less spectacular to add to the melt and much easier to source but gives a slightly different microstructure with comparable mechanical properties to normal ductile iron.

http://foundrygate.com/upload/artigos/cwvdHuZP6Rdfod5sCbaSgtTLXI35.pdf

 

I believe this paper concerns the post inoculation of Ductile Iron. They already have ductile iron and are adding the necessary Si to replace what was lost in the process. It also enhances the nodular graphite. It's just a cheaper substitution for FeSi. The only way to make iron ductile that I've seen is with Magnesium, Cerium or a Calcium Silicide addition, (Meehanite Process).

I think the main problem for backyard ductile iron is not knowing the chemistry of your scrap. You need a very low sulfur content, .030% or less. Your scrap may be .100% or higher. Sulfur will neutralize the action of the Magnesium. We are kind of shooting in the dark. Treatment with some form of Calcium, Lime, oyster shells, Dolomite lime etc.

One thing I read is that it's a good idea to raise the Carbon level as much as possible. Iron won't hold much more than 4% C then it becomes saturated so, with the lime, an addition of crushed anthracite coal will add Carbon and make a good slag to hold the impurities. Toss some vermiculite on top to make the slag easy to remove. Then add your inoculant, NiMg3 or what ever your using, then add FeSi or SiC to replenish the Silicon, skim and pour.

https://willmanind.com/what-ductile-iron/

 

Calcium Carbide is said to reduce/remove sulfur. Not sure how or when to use it, though.

 

In my case, I purchase returns from a commercial foundry. So, the chemistry is pretty well known. My goal is to obtain fracture-resistant castings. I should be able to test for fracture resistance pretty easily.

I will bear in mind your recommendations for improving carbon content and also lower sulphuric if needed.

I will get some vermiculite to improve slag removal. That alone would be a step ahead. Thanks for you input.

Denis

 

You can do an in mold process where you have a large basin at the top. It has to hold the iron for three seconds to homogenize before going into the sprue. You can use a reaction chamber to put the alloy in and have the stream run through it dissolving the alloy as it goes into the basin or you can just pour straight on top of the alloy into the basing without a chamber. You can use a metal disc around .040" thick to hold melt at top before melting out and introducing the melt into the sprue....or you can do like I was going to do and just use a core or something stopping up the sprue then pulling it out letting it drain into the sprue. Just needs 3 seconds to homogenize. Base iron must have very low sulfur level. Hard to get sulfur down without using calcium carbide. Calcium carbide produces lots of slag...but in hobby furnace setup no big deal. I may have some literature I can post if I can get good pics...my camera lens is scratched and takes horrible pics right now. I will find out how many grams of alloy we did per pound for sure...we also used rare earths to make it and did well. I know we used somewhere close to a gram a pound on some trials but do not think it was that high on most.

 

As in "mischmetal" / ferrocerium?

The latter isn't too hard to get ahold of - it's used in "fire-starting rods" - and in some cases, has a piece of Magnesium attached to it!

 

So I checked into some of the weights we used and they fluctuated drastically depending on which alloy you use. The closest thing I saw to NiMg3 we used about 75 grams on average of 30 pound pour weight. We had high sulfur and tried to over inoculate on a few trials and the alloy would not go into solution above a certain level and left unmelted alloy floating around the top of the pouring basin.
Im not sure about the rare earths but I definitely know that we had some with the ferrocium in it. We usually buy preblended mixes that are supposed to be mixed specifically for our iron chemistry. Even if you dont hit high nodularity ductile and get CGI you will see a huge performance gain for tensile strength. Our CGI skillets would bend the handles before they broke.

 

Namely, even stuff that isn't the full-on ductile will be less inclined to break. Toss in some nickel... that handle will last!

 

Try checking with Globe ....not sure what the whole name is...Globe metals I think.... they are who supplied our alloy and have lots of knowledgeable staff. Not sure how they would treat small order hobby level enthusiast but they may help or know of another company that will.
Are your returns that you purchase grey iron or ductile?

 

Ni-Mg ? Hard to alloy. I tried recntly a small gram quantity under a carbon arc to melt the nickel and add the Mg which immediately caught fire. The boiling point of Mg (1100 C) is much lower than the melting point of Ni (1450 C).

 

Billy,

I hugely apreciate your investigation into methods that work. I will check with Globe. But, I am pretty sure how that i s going to go....

How much calcium arbide is used on a per pound of melt basis. I suppose that depends on sulphur level. But just having a range might help.

I can use either ductile or grey returns. If I were using ductile, it sounds like low sulphur levels should be more or less guaranteed. Right? That would be a good place to start. Then just getting hold of some NiMg3 might do the trick. As you suggest, if I could even just get CGI, I'd be a lot better off for this project and likely that would be just fine for future needs when reduced brittleness was needed. By doing a sample fracture test I could be pretty confident that at least CGI was acheived.

NiMg3 going to be a significant barrier. I may be able to get a little now. But future needs???

I'd also like to get a Home-Foundry method for making CGI documented here for the "next guy" that wants to do this.

Denis

 

Your base iron determines almost all of the calculations. If you start with ductile returns you will have to do very little to get ductile and might be able to do straight melt with little to no alloy and get CGI. Adding too little will leave grey iron or CGI and adding too much will make ductile but will just not all go into solution and may stop up the sprue with globs of alloy. I would definitely go with ductile returns and make life easier with the lower sulfur.
1. Get total weight of pour and establish a pour weight to grams of alloy ratio and start experimenting. This may take some time to establish proper ratio for desired results. Having start with ductile you will have an almost given outcome of at least CGI.
2. Build pouring basing big enough to hold entire pour weight in the cope.
3. Take a .040" metal disc or a core shoved into the sprue to keep iron from going down it.
4. Put alloy in basin and pour iron on top. It will spark a little but nothing scary for us grown men.
5. Wait for disc to melt or count to three and pull the core and let it drain. That is all there is to it.
The hardest part to all of this is establishing a process that gets end results and being consistent enough to get the same results every time. Even with our most sophisticated technology we used basic methods that were easily reproduced.

 

Thanks for the above, Billy. So now we are getting closer. About what range of NiMg3 per pound to get (or shoot for at least) ductile would you expect when melting returns. I am sure that will vary. But just having a starting range may be helpful. I can recall using way way too much FeSi in my early pouring days partly related to incorrect information obtained from a no-longer member here and due to my incorrect timing of addition. I'd like to avoid a repeat of those false-starts if possible.

Do you think there is some reason not to dump the NiMg3 into the melt like they seem to be doing in the video I linked above? Seems like that would be simpler than using cover steel. It also avoids trapping air under the cover steel and then having it expand rapidly when the melt burns through the cover. The methods of having inocculant melting into the stream of iron sounds like asystem that would bve cool for a production setup once all the bugs were worked out. But, it sounds like it might be initially more prone to failure than a straight away addition.

Workin on it...

Denis

 

I'm trying to follow along, Denis. What are you referring to as "cover steel"? How is that different than what you have been doing lifting a plug?

 

Strange question: would ductile iron "work" for lathe tooling, e.g. quick change tool post and tool holders? (Better than Chin-Lee's "aluminium," most likely...)

 

Well, OIF, I’m trying to figure this whole ductile thing out too!

A commercial foundry I visit uses cover steel to make ductile. They pour their inoculant which I THINK is NiMg3 into their tapping ladle. Then they cover it with a layer of steel “coins” about 1/4 to 1/2 inch thick. Then they tap their pouring iron into the ladle. The hot iron melts through the steel and mixes with the inoculant. Now they have DCI. I am working on them to get the details straight and hope to get some inoculant from them. From bits and pieces I’ve found online a solid thin cover can also be used.

Denis