Background: I had the pleasure of visiting my mentors at a commercial foundry in Western Washington that casts grey and ductile iron, steels and stainless in amounts up to 10,000 pounds. The owner together with his metallurgist and foundry foreman, in the course of discussion about I was having trouble with a small casting, provided me some information about their use of ferrosilicon. I am sure they mentioned this a few years ago as I was beginning my small foundry quest, but I was so preoccupied with a hundred other questions that I failed to pay close enough attention. I also think that back then I missed the crucial part about inoculating the melt only a very few minutes (5 or less) prior to pouring. Anyway, I fumbled around and ended up using much more ferro than was needed for quite a while, but gradually decreasing and decreasing amounts used. Most recently I've been using about an ounce per ten pounds of iron melted and for quite a while have been meticulous about adding it within 2 minutes of actual pour. I ahve gotten good results that way, but may have had more shrinkage than was necessary though I have been able to overcome that by use of risers.so for 50 pounds of iron I should be using The recommendation is that I use what they use---a 1:800 ratio of FeSi to iron. So for 50 pounds of iron I need only use 1 ounce of FeSi. I will be doing just that later today along with repositioning my pattern in this deceptively "simple" casting that has been giving me fits. More on that in a separate thread when I have more time to write and post pics. How does this ratio of 1:800 square with others' experience? Incidentally the spectrometry they did on my failed casting showed spot-on analysis except for Si which was 3 times the amount called for. That amount of silicon is not likely to cause any problems except possibly excessive shrinkage. But machinability, strength, and stability should be unaffected. Denis
If that's the case then I've been using 8 times the required amount by putting one percent ferrosilicon into the iron. The iron bar I cast with the leftovers was machined into a pulley with the comment that it seemed slightly harder than regular grey iron by someone who has literally machined tons of the stuff over the years. My limited reading on using silicon carbide instead of ferrosilicon is that it's effect on iron lasts for a period of hours rather than minutes but I'd struggle to find that quote again.
Don’t feel bad, Mark. I was using about 5 times more than needed, assuming the information I got is correct. I think it should be as I repeated it back and wrote it down. The idea of needing to inoculate only minutes prior to pouring is what I have read from good papers on several occasions and I discussed that point with the metallurgist yesterday. He emphasized that timing as being important. Some foundries inoculate the ladle and some even use ferrosilicon in the mold itself. The ferro in the mold seems crazy to me, but it is true. Silicon does persist in the metal once added, but its fluidity and anti-hardening effects are short-lived. That seems strange, I know. But it was consistent as measured in my sample yesterday, for instance. Denis Added: For a while early in my iron endeavors I was using way too much FeSi based on some misinformation I came across here and based on my experience of having some hardness issues despite using 20 to 30 times as much as I now know is required. And that was because I was still experiencing some hardness with large amounts of FeSi added to the melt. But, in retrospect, I know I was having trouble because I was adding it 20 to 30 mins prior to a pour and by then it had lost almost all of its effect. That experience squares with the timing information I learned 12 to 18 months ago. Since then, no hardness issues as I have been gradually dialing back the amount added and will further reduce it to the 1:800 ratio recommended. You can be sure I will be Rockwell hardness testing the results as I have been routinely doing since starting to pour iron a few years ago.
At least now I know the proper amount of ferrosilicon for next time. I've had a bit of a trawl online for papers describing the use of silicon carbide as an innoculant. There's one Czech republic paper that claims 0.35% silicon carbide with 0.15% Zircinoc will give you ductile iron, complete with microphotos of the grain structure and there's another Indian paper that claims silicon carbide has no effect on grey iron. This is at odds to just about every paper I've read about silicon carbide. They used ferrosilicon for a given silicon content to all test sample melts and a range of inocculants: they concluded that calcium silicide was the best innoculant. It would be great to be able to make ductile iron with something as common as silicon carbide but it still looks like other innoculants are needed, which is how I interpret the paper by the Czechs. Here's a link to the Czech paper where they are casting a suspension part for a Peugeot 405 car in ductile iron, they typically use Zircinoc which is silicon, zirconium, calcium, aluminium innoculant for ductile iron and they claim silicon carbide is a direct substitute for a large portion of Zircinoc (75%) when making ductile iron. https://foundrygate.com/upload/artigos/cwvdHuZP6Rdfod5sCbaSgtTLXI35.pdf Here's the link the Indian paper by a couple of Bangalore professors who claim silicon carbide does bugger all. They use ferrosilicon in all melts to adjust silicon content and also a fair amount of scrap steel in their melts too. They claim calcium silicide is the go...for making ductile iron. I think calcium silicide is what's used for Meehanite iron. https://www.irjet.net/archives/V4/i9/IRJET-V4I933.pdf A link to a description of Zircinoc inocculant: https://www.elkem.com/foundry/foundry-and-steel-products/inoculants2/zircinoc-inoculant/
So this paper about silicon carbide is interesting: https://www.researchgate.net/public...on_Based_on_Steel_Scrap_Using_Silicon_Carbide It mentions that silicon carbide is slow to dissolve in iron "melting process lasts dozen of minutes long" and the most efficient way is to add it to the solid charge of iron in the crucible rather than to the molten iron (for an induction furnace). The paper also mentions that silicon carbide as it dissolves will form graphite clusters and that: "The graphite clusters are thermodynamically metastable over some time period so the play a very important role in nucleation process and eutectic graphite creation. The ferrosilicon dissolution can also promote the graphite clusters appearing but due to its higher dissolution rate these cluster remain stable over a short time period only." So it looks like the silicon carbide inocculation effect does last longer than ferrosilicon but takes longer to dissolve too.
I am happy to just use the Ferro Silicon I can easily source, is easy to use, and yields excellent results. Perhaps in a large scale operation some other inoculants might be preferable for some reason. But, for now, I am content (perhaps ecstatic) to just get the basics right. If I were to make ductile iron I would simply use magnesium added shortly prior to pouring—-something I discussed with the same metallurgist who said that is the method they use at the commercial foundry and what I could do in my setting. Of course, I would devise a “remote” dumping method to make sure there was a safe distance between me and the expected spectacular reaction of the mag with white hot iron. That would be a fun one to do at night! Denis
Yes that would need a lid for the crucible for sure, like a long rod with a charge of magnesium and a round lid above it. I haven't asked yet but I'm pretty sure I can get calcium silicide from the place I got the ferrosilicon and I think it's less spectacular to add to the melt.
Here is followup on my 1:800 FeSi to Iron pour that I did yesterday. The iron showed excellent fluidity as it flowed 2.5 inches up the 1/16th" vent holes and filled the remainder of the mold very nicely. I tested it with my Wilson Junior calibrated hardness tester and got readings of between 0 and 12 indicating very soft grey iron----as desired. And I tried drilling it using a hand-held drill motor and bit and it drilled very easily. My conclusion is that 1:800 FeSi produced a very fine casting. I poured at 2550F and allowed about 90 secs from inoculation to initiation of pour. This is the resulting casting having been brushed off. Incidentally, this pattern had been giving me a lot of trouble with collapse of the angled prism face and the concave cove between the top rail and the prism. I had been trying to pour it pointy side up. But, when I flipped pointy side down I cast it with no riser and simple vent holes in the mold and got a very acceptable (for me) casting. The casting is about 8 inches long and weighs about 8 pounds each with a pair of castings poured together. I packed two molds of two castings each and poured them with 35 pounds of iron. I would have gated the mold on the ends but there was not room to do so in the flasks I was using. Denis
Here are a few pics of the machined casting with a very sensitive longitudinal vial pocket drilled and tapped and a cross vial hole made in it as well. Denis Denis
That looks great!, I have to make a few tools like that to scrape in the Graziano saddle and cross slide.
This tool is aimed at use on smaller dovetails and shorter flats like those found on a cross slide or saddle of a mill. The vial that it is designed for is a 10-arc-second vial from Geier and Bluhm 4-3080 (obviously the packet can be whatever you want). The cross vial is a 3-0507 the same manufacturer. The design shown is original as, to date, no one that I know of has made a 4-function Straight Edge/ Parallel/Prism/Level. It has been well received by the scraping community. Denis
I can see why, it combines several tools into one handy unit. An old lathe can be freshened up a lot by scraping the saddle, cross slide and compound and that tool lets you do the dovetails and 90 degree V ways while ensuring the bed is aligned without twist.
