Trying Lost Foam with Iron

Satanite, rapping on the bucket, dunno temp -- higher than I pour for greensand, quickly melted the steel stir rod to a point.

 

This shows some of the turbulence Kelly describes when I poured the bench socket for Hardy hole anvils. The pouring cup was just a dished basin, but when it was choked and started to vaporize the sprue it replicated the foam accurately.

 

That all sounds likely to me. As an additional problem this attempt, looking at the resulting scattered pieces in the sprue, I do think that the Satanite was not permeable. I think there was a violent breakup at the beaded section. I also think the sand had very poor permeability.

It was the fine bagged brand sand Kelly used, but I do think that likely varies by location. Iowa and VT are far apart for shipping bags of sand. The stuff I had was very non-uniform with a LOT of fines, so much so that when poured into a container there was a small dust storm. Compaction produced poor permeability. At the time I was tempted to go with medium sand which I also have and is uniform grain size quartz, very different than the dust-like Fine. I've used both for greensand cores in the past and the Fine sand is useless.

Overall, I'm glad I tried this experiment. I do think I've got the answer to my own personal question about the practicality of casting iron with lost foam. It would take a lot more effort and experimentation, as well as more equipment to achieve a decent casting for even a short tube shape. By comparison, I can cast the same piece in greensand tomorrow with a simple core and little wasted material. That's just a practical reality.

I do have to laugh at the "casting" I managed to produce, particularly in the light of the superb aluminum manifolds, etc you've done Kelly. Absolutely this is the worst casting I've ever made, or even seen anyone else make, in any material! The only positive things I can think of were, yes, ESC, it was gray iron, and the pouring cup slipping out of a tube form was kind of cool. It did hold up well considering, I think, also. I don't use pouring cups with greensand, so probably won't ever make more. But anyway, for those who do, that seemed to work.

 

Kelly, I'd be extremely interested if you do try iron with LF, and you do have the equipment and experience to start way ahead of the game. I of course understand it isn't a priority, but if you do some day I think it would be fantastic. And yes LF does enable amazing possibilities in molding, which you've clearly taken full advantage of.

My interest in LF was only in small cylinder liners and pistons. Small sand cores are more difficult than larger to make, and there is a tendency to pick up sand in bores, which is hard on drills and boring tools. I thought LF might allow higher quality ID surfaces, and also making much longer stock which could be sawn off for individual use.

Long narrow cores become too fragile for handling in sand, but LF would seem to allow longer stock. Iron is preferable for steam and hot air engine cylinders, which is my interest, because of the graphite assistance to lubrication. So aluminum and other metals aren't applicable for what I need. Other than small cylinder stock, and possibly pistons I'm happy with sand casting and as you say, it's long familiar. I was hoping a small straight tube would be an easy subject in LF, compared to manifolds, etc. But I got taught a lesson there!

I think iron is a lot more difficult in LF. When I cast before -- the little I did in LF was a result of seeing the fabulous Build Your Idea website in 2005. Sadly it is no longer around. Back then I did do some aluminum castings just as he did, and they came out great. Here's an old pic from that time of a linear bearing casting intended for a CNC router lathe I wanted to build. Simple molding in a large tin can with coarse sand and an LF pattern coated in wallboard compound. A hinged, above surface split steel pouring cup. No compaction to speak of, little head height. Would it be possible in Iron? Uhhhhhhh.......no.

 

I would if I believed there was a small chance it would work. But to me, it won't, for all the reasons we already cited, including bad sand low compaction, and now with wallboard compound, exceeding refractory melt temp. Plus this one: when molten aluminum hits foam at what, 1500F? the reaction is fast. Raise that a thousand degrees, and it's explosive. I think the beaded foam and Satanite of the sprue blew up internally, and they were only 3-1/2" sub-surface because the cup was 4" deep. Even with good compaction I think the pressure pulse in that area yields an expanding ball of sand, metal, refractory and gas. I don't think shallow compaction can hold it. Maybe if the pouring cup was 10" long before the metal hit foam, compaction would hold.

Plus, I want to do some sand casting to clear the palate.

 

Like Ironsides I coat cores with plumbago which does a good job of separating core and casting for cores over 1". So basically a dry core wash. Sometimes traditionally that was mixed with alcohol and sprayed on. There's still an occasional few grains on the inner surface and that's what I was talking about dulling tools, I wasn't talking about sand inclusions in the deeper metal. Surface grains are not so bad with carbide tipped boring tools but problematic on twist drills, which are often favored first to reduce boring time. They are easier to remove mechanically or avoid on larger ID tubes, but start getting difficult below 1". That's where I though LF might be an advantage.

I've never had a core act buoyantly or move if horizontal, and any superficial bonded sand is not found preferentially on the upper surface -- it's just a contact thing all around. Buoyant slag on the other hand does favor the high points, but stopping it has to happen before reaching the mold cavity. If present and cast vertical it would move to the top of the casting. But there may be some other advantages to vertical core casting with a vertical parting -- I've never tried it, so I will take your suggestion some day just to see what happens. I have cast a fair number of solid bars vertically in simple open face casting for stock.

No need for venting or chaplets in simple baked sand (or cement sand) cores of the sizes I use, like ironsides points out in vids. I like relatively low tech, so avoid resins generally and try to get good results with basic materials, if at all possible. I've used many gallons of epoxy over the years as a professional boatbuilder. I avoid it for the most part these days. When I look at the products of casting a century or more ago, I am continually amazed at what can be achieved with basic materials -- skill and knowledge are what made those possible, and the the rarest materials that I admire the most.

 

I did the palate cleansing today and took that junk LF casting and a few old sprues, threw them in an A6, popped it in the furnace and put the WVO to it. Melting started in 12 minutes, and I added stock up to 10 pounds. Got it good and hot liquified all the included junk, and skimmed. 16 grams of ferro added. Did a simple greensand open top vertical pour using two 1-1/4" PVC water pipe sections as patterns and cast a couple of 1.7" dia. bars 7" long out of that dirty metal. The single leftover ingot was pure gray when split. No waste.

Felt good to salvage that last LF failure.

 

Kelly, I never suggested that anyone else should not try iron with lost foam, in fact I feel; the opposite. I'd like to see more people try it. As clearly laid out, my own very specific interest in LF was ONLY in possibly improving interior finish and longer castable aspect ratio for conventional simple cores in casting small long cylinders ie. <1" and say 12" long. I started out with trial cylinders with 1" inner diameter and only 5" long in order to simplify things, though I had initially cut a full single 12" length in foam.

I can easily cast a more challenging 6" x 3/4" dia. core (ie. 8 to 1 aspect ratio for small diameters) in greensand as a conventional baked sand core using wire reinforcement and no venting. Using plumbago produces an acceptable interior finish -- I was hoping that LF might improve on that finish or length to produce the finishes I see on first class aluminum LF casting, right out of the mold. Nothing I did proved this was either possible or impossible, nor have I drawn either conclusion.

Yes, if simple handling breaks a sand core, then it's pointless to try to cast longer or even talk about core shifting or breakup in the mold. That was my point. A 3/4" x 6" long core even with wire reinforcing is delicate, but do-able. As you get small the problems increase geometrically. Just placing wire in a small core is difficult from a practical standpoint. I cannot imagine placing a wire in a 12" x 3/4" baked sand core. How would it be inserted, or the sand packed, or the core released? Maybe someone can, but I can't. I have nothing against sand casting a long tube vertically -- if one could produce a core that didn't break before placement in the mold. I can't, maybe someone else here can. That's great, and I'd like to see that, and I'd be happy to adopt that method.

And for personal reasons I don't want to use epoxy. That shouldn't be a problem for anyone else, that's just my preference. We all have specific personal preferences in casting. What I CAN imagine is that LF holds the possibility of casting small diameter tubes of aspect ratio's limited by the depth of metal penetration. Core handling strength is no longer a limiting factor.

I've never suggested that my one limited trial was the end-all and be-all of LF iron small tube casting Just that what I used didn't work, and I offered my own theory (among many additional reasons) why not, in this specific case. Theories abound in metal casting, almost none of them proven, and a forum would seem a reasonable place to speculate on what failed in a specific experiment of my own. I did not condemn an entire branch of casting by doing that. Anyone is free to disagree, and I wouldn't be surprised or chagrined to find I was totally wrong. Newbies take note: believe nothing I say without a grain of salt. Experiment, learn, think, have fun. Casting is an open topic.

As for my giving up on casting small tubes in iron using LF with only one experiment, sorry, but I do what pleases me at this age. I have a lot more I want to do in casting and building engines this summer.

 

Sodium silicate cores can be quite strong. I have in the last 6 months increased my Na Sililate RU percentage in sand cores by maybe 20% with significant increases in strength while retaining excellent shakeout. I detailed my prior testing in

http://forums.thehomefoundry.org/in...temperature-and-sugar-content.794/#post-17878

I think I just scratched the surface in that thread. I no longer bake cores. I use about 7 to 8 percent silicate. I use the same silicate/water/sugar mix. I increased coal about 1/5 over my initial mix.

Other members (e.g. Chazza) have expertise in linseed cores which I believe can also be very strong.

Epoxy is great as a binder if you don't mind using it. I don't mind at all. It is used in such small percentages that it is not prohibitively expensive for one-offs.

Denis

 

Hi Dennis, thanks for replying. I just read your thread and it's definitely interesting. I do, of course know about sodium silicate and linseed oil cores in general, but not by test, as you've done. You mentioned a hollow prism in another thread -- can you give me a link? Were you able to cast it the way you wanted?

You mentioned in the thread 100 mesh olivine for the core sand and the need for venting. I don't know if it's of interest (or this is just old news) but I found that the finer the mesh, the more binder was required for equivalent strength. The more binder needed and the finer the mesh, the lower the gas permeability. The need for venting then becomes important.

When I mentioned a bunch of posts above that I don't use vents on baked sand cores (in the sizes I mold) for iron, the reason that became possible was that i moved over to using much coarser sharp quartz sand than I had previously. The cores were stronger, and more permeable, the only drawback was the resulting interior finish. In aluminum this might be objectionable for a visible non-machined surface.

In iron this can be controlled by rubbing in plumbago on the core, which smooths the core surface among other benefits. For a machined iron surface I find that large grit surface sand inclusion is more easily removed mechanically before machining than finer core sand.

 

Just to be more specific, I checked my core sand and it appears to be about 40 mesh.

The reason less binder is needed is that, if you imagine it, to make a solid, each grain needs to be coated all over with binder, but ideally, no more, leaving space for gas permeability. As you know surface area does not increase as fast as volume when you increase the size of something. If we take the total volume of a core as fixed, if we double the size of the grains to fill it, we reduce the number of grains needed by eight times, and increase the surface area each grain by only four times which gives a net reduction of surface area by about half for that core. Thus less binder needed to coat those grains. We also increase the size of the spaces between grains, by eight times. Meanwhile the binder thickness around each grain can remain the same, increasing the permeability. Less of the now enlarged space between grains is taken up by binder, proportionately.

You can also see the advantage for a core of having relatively uniform grain size vs lots of fines.

 

Yes, that is well known basic core mesh vs binder vs permeability information. Cobett covers it in his core-making monograph, for instance.

I discarded the triangular prism design for a better one without testing core making for it.

Besides variations in binder quantity and formulation, reinforcing them with a metal wire or tube armature is a another method to I use to strengthen cores.

Denis

 

Yup, I use wire in 3/4" dia cores, too. But 6" is roughly the limit for easy insertion/packing/handling even with wire on that size.

For cores greater than 1" dia things are easier, and I use Portland cement as a binder per Ironsides.

Where would I find that Corbett monograph?