For a while now I have been trying to figure out the right proportions of FeSi to add to cast iron to improve fluidity and improve machinability. I saw suggestions from folks to add perhaps 18 ounces or more to a 50# melt. My mentor at a large commercial iron foundry suggested only adding 2 ounces per 50. But he was basing this on his experience melting iron with an induction furnace which I subsequently learned depletes existing FeSi in iron less than when subjecting iron to the rigors of a diesel furnace as research done in India (where small to moderate commercial operations commonly use diesel) demonstrated---the longer the iron was heated and the more it was exposed to a non-inert atmosphere the more various components of cast iron are depleted. and consistently when I read papers on FeSi I saw cautions about not using too much FeSi as, in excess, it can cause a bunch of problems---like excess shrinkage. What I had a hard time getting was specific recommendations for how much to add. Positively every writer and my mentor emphasized the need to add FeSi only a few minutes prior to actually pouring. Curiously, the beneficial effects seem to be very time-dependent. Added more than ten minutes prior to pouring could be expected to have little or no benefit. The castings I make are used in such a way that even mild hardness is unacceptable and I was experiencing some Rockwell 10-15 areas in thin areas of my castings with addition of low levels (2-3 oz / 50#) of melted iron. BUT, early on when I was having such problems, I had not yet learned the importance of timing and I did not have a very good method of stirring the iron after adding FeSi. I stirred it, alright, but with a steel rod that rapidly melted away in the hot iron. Now I use a 1/2" graphite rod for stirring which holds up just fine and allows excellent stirring. So, I went to adding pretty large amounts of 75% FeSi to my melts---on the order of 16 ounces per 50 pounds---and seemed to overcome the hardness issue (by brute force, in retrospect). BUT, I started having shrinkage problems which not an issue at low levels of FeSi. I overcame those with risers and gating adjustments. Of late, however, I have be throttling back on the FeSi and have noted dramatic decreases in shrinkage. Today I added only 6 ounces to 50 pounds and had almost no shrinkage as the 2" dia risers, instead of being deeply sucked down in the middle were hardly sucked down at all. And on my second melt today, I added only 2 ounces to 30 pounds and again saw virtually no shrinkage. I will be testing for hardness in a day or two and let the forum know results. But, I am pretty sure there will be no hardness issues and clearly no shrinkage issues. By no means have I got this all figured out yet, but it is really nice to see the shrinkage which had reared its ugly head now receding. Clearly, I would expect some shrinkage and need for risers if I were casting patterns that were quite chunky as that is pretty much a given. But my patterns are much thinner in section and so riser additions are less likely to be needed if shrinkage can be corralled. On another thread where a windmill gear is to be cast, I would expect some shrinkage with relatively low FeSi additions and a lot of trouble, since it is so chunky, if high levels of FeSi are used. I am interested in rules of thumb other cast iron casters may use and what their observations are concerning shrinkage, softness etc. I am still trying to wrap my head around this and any outside help is appreciated. I think I still be quite a bit higher than needed at 6/800 or 0.75% FeSi 75% concentration added. Denis
Thanks for the info Denis. That is very good to know. I have said before, but it bears repeating. Before I discovered ferrosilicon, I did not use it at all when casting flywheels. The flywheels are perhaps 1.25" thick at the rim, and they machine easily. I cast two of these flywheels, and had no problems machining either of them. The material used was true purchased class 40 gray iron bar stock. I was using a multi-part mold when I cast one of the flywheels, and some of the parts were thin, so the same metal that filled on of the flywheels also filled the thin parts, and any part that was in the 1/4" to 3/8" thick had the hardness of tool steel, and could only be ground, not cut with any blade, and not cut with carbide. So I would venture to say from experience that parts that are at least 1" thick do not need ferrosilicon added for machinability. At some point at the parts get thinner, obviously some amount of ferrosilicon is needed. Scavenger used a considerable amount of ferrosilicon in his mufflers (I will dig out the exact amount), apparently with impunity, and he did not report problems, but that is not to say he did not have problems, he may have just not posted about them. I am aware of one other person who has experienced some shrinkage issues with iron, with some tears visible as the part cooled. I am not sure how much ferro this person was using, but I will inquire. And ironsides alluded the other day to the fact that some iron alloys shrink a lot, and others do not, but I wonder if it is the ferro that is contributing to this in addition to any alloy differences. See his recent angle plate casting. For the gear casting, if it were cast without ferro, I would anticipate that the hub part of the gear will be machinable, but the teeth will not be machinable. I will add some minimal amount of ferro to the gear casting, but I am not sure what. I have been jotting down reference numbers for the amount of ferro that various people have used. I am also casting a surface plate, and it has a top that is1/2" thick, so I can probably get away with a minimal amount of ferro. .
I am not sure which video or thread contains the information on Ironsides' angle plate. I did come across this video b y Luckygen1001 on his casting of an angle plate. There were two interesting points. At 7:24 he add Fe Si to 14 pounds of iron and you can see that he adds only a very small amount. He does not say how much (really wish he had) but it looked like only an ounce or so. And then near the end you can see the trouble he experienced with shrinkage which he intends to remedy by changing the orientation of his pattern. The results of his nameplate casting is very nice and in that very flat and thin pattern he had no trouble with shrinkage even with the same pouring metal. Denis
I asked ironsides if the angle plate defect could be caused from air trapped at the top of the mold, since he does not vent his molds. He said no, he was sure it was shrinkage, and had seen it before. I have had defects that looked exactly like that (bubble-shaped) that were indeed from when I forgot to drill vent holes in the top of the mold, so it can happen. I was using bound sand, and bound sand may be more air-tight than the greensand that ironsides uses. I wish he would try that again with vents at the top, but regardless, I always vent the high points of my molds because I have had what were obvious trapped air defects at the tops of non-vented molds. .
The strong clues to it being shrinkage rather than an air/gas bubble are the meniscus you can see on the near side of the casting where the center portion of the casting froze off last and tried to feed the end. In addition he points out the bits of metal showing initial complete mold filling. So, I do think he is right about that. You can also see quite a bit of shrinkage in the cap of his sprue suggesting that the alloy he poured had a propensity to shrink. As to whether vents are categorically unnecessary, that is a more difficult question. For him, with his particular sand, binder, sea coal, and moisture, he evidently does not see a need. I do always vent the high points of my molds feeling it is a simple and harmless measure and makes sense. But I have to admit to not being sure it is necessary. Denis
