Bell Casting

Petrobond. You can either buy the sand pre-mixed with it, or buy just the oil to mix with your own sand. My buddy who does casting swears by it and won't use anything else.
When he first tried it, he cast a Jefferson Nickel in aluminum and gave it to me. The detail is good enough for me to read (barely) the word "Liberty" and the year on it.
Alan

My formula by wt: 84 pct sand, 7 pct powdered clay, 7 pct water.
Make the powdered clay by ball-milling bentonite kitty litter (cf Dan Williams ball mill).
Blasting sand from Home Depot works well.
Saturation of water into clay may take hours or overnight.
==========================================================
From 5/2004:
Baked core formula 1(Mark Fowler's formula) 10 parts molding sand. 2 parts clean dry sand 1 part molasses water (which is 1 part molasses to 8 parts water). 1 part boiled linseed oil. 1 part flour (may be left out).
Baked core formula 2 (tested successfully by Rupert Wenig) 20 parts clean, dry silica sand. 1 part wheat flour. Molasses water to temper ( 1 part molasses to 10 parts water). Mix dry ingredients, then temper with the molasses water.
Sodium silicate formula clean dry silica sand. 3%-5% sodium silicate (waterglass) by weight Gas with C02 to set (about 10 sec. at about 3 psi).

There are many recipies for green sand and you will probably get several here. You do want to start with a fine grade of sand. I use AFS 60 Sieve and add about 10% bentonite. Bentonite is a very fine clay. It is used in kitty litter to make it clump together. It is also called drillers mud. Beg some from a well drilling company. The combination leaves a nice smooth surface. Plastic binders like LinoCure are easier to use and make a stronger mold but are basically single use unless you have some expensive sand reclaiming equipment. Petrobond is a very popular binder with small casters. For the baked core inside the bell the baked molasses recipies work well.
In either case the mixing (mulling) is the most important step. Spread the mix on the floor and rake, hoe and shovel it around until it is a very even consistancy. That can take way more time than you would expect.

I think it's better to buy bags of powdered bentonite than to bother with ball-milling kitty litter. (Plain unscented kitty litter is getting hard to find, anyway.) I paid around $7 for a 50# bag of bentonite, IIRC, at a local agricultural supply place. (It's used in feed supplements and for pelletizing.) Also available at industrial materials stores and foundry suppliers. ... ... ...
As a note to original poster, those core formulas apply if your part has a hollow in it where you can't withdraw a model when making your mold. For example, cooling passages in the walls of an engine block. The sand of a core is stuck together much more firmly than the usual green sand mold. I think core material isn't suitable for the main body of the mold because of not being porous enough but don't know for sure. -jiw

Here's some good links on sandcasting--
Sand:
Petrobond supplier--ready-mixed, or ingredients plus lots of other foundry goodies:
K-bond, a cheaper alternative to Petrobond--but Bentone (NOT Bentonite), it's main ingredient, is hard to find in small quantities:
Ken Grunke P.S.--Kitty litter sucks for sand mixes.

Hit the Lindsay web site. He not only has lots of books on casting, pattern making, etc., he also has stuff on old technology, such as would have been used to cast bells a couple of hundred years ago.
That includes a book on casting bronze (IIRC). You might also order Dave Gringery's book on making and using a small aluminum foundry. It doesn't apply directly to what you're doing but if you've never done sand casting before, it is a great place to start learning.
--RC
If I weren't interested in gardening and Ireland, I'd automatically killfile any messages mentioning 'bush' or 'Kerry'

I've had pretty good luck with natural bonded sand. I dug it out of a hillside on a power line right-of-way. It has enough clay to bind the sand together, and the sand if fine enough for detail in the casting. This is a reddish colored sand from piney woods hills.
-- Gary Brady Austin, TX

Not round here.
Sometimes they cleverly disguise it as Oil-Dri.

Your wasteing your time with home brew greensand mixes especially those utilizing kitty litter for the binding agent..it just plain sucks.......
Regular bentonite for foundry use is not exactly the same as driller mud. It is close and 110% better than kitty litter cold ever dream of being.....
There also is no get by on proper sand either........although mason sand, sand blasting sand etc all will work, they are just not uniform in grain size as a true foundry sand. Real foundry sand is chap, about 8 to 10 bucks for a 100# bag of Olivine which is hard to beat. Same thing with Hydrobond......(water based binder / bentonite) which is about 12 to 15 bucks for a 100 pound bag.........Mix bentonite to sand at a 100# sand 10 pounds hydrobond, and approx 3 % water.
Petro bond is better yet, but much more expensive especially if you buyt it already premixed, but it works super and you can get much finer detail and finishes with Petro Bond II resin binder than you can water bondd sands.......
A core is not needed if you have a bell pattern with the center of it already shaped and hollow.......but a core is much easier to use. It all depends on how your pattern is made or if your going to use a precast bell as the pattern then you can ram up the inside of the previously cast bell and use a greensand or petrobond sand core....
True kitty litter is comprised of bentonite, but what proportion it is in the litter is anyones guess, and other additives that have been added can also cause problems. It will take considerable time to grind and pulverize this kitty litter to a fine enough medium and then couple that with regular old run of the mill sand of grit that is not uniform in size and you sure do not get a quality greensand, or a greensand that will impart a decent finish by any stretch of the imagination. For under $30 bucks you can get the proper bentonite and foundry sand, which is not a lot of money when you look at it as having the proper stuff and save on headaches and waste by going home brew mix and match..............you just don;t get anywhere thats worthwhile.
Visit my website: expressed are those of my wife, I had no input whatsoever. Remove "nospam" from email addy.

I bought a 40# bag of oil-dry stuff expecting to find bentonite in the bag, but it doesn't seem to be. So I have to use it up on oil spills rather than as a green-sand binder, it looks like. -jiw

Oil absorber is usually vermiculite with a light dusting of Bentonite. But Bentinite is not all that hard to find unless you live in a big metropolitan area. Call a well driller and ask for drillers mud.

My cousin, in the alum and bronze casting business, uses powered walnut shells to provide the very fine cast surface. R. Wink

Any sand will work, but finer the better. Silica is typical (sandblasting sand should work, though it could be finer) but olivine has become popular, though more expensive.
Any clay will work, stickier (fine-grained like bentonite) or refractory (ball clay) the better. Fireclay is very refractory but not generally very sticky. Bentonite has historically been the primary medium.
I happen to use ball clay because that's all I can get at the local ceramic store. My sand is standard masonry sand, 50 pounds a bag from Home Despot. Pretty rough but that's what files are for... I recently mixed a batch of maybe 80 grit sandblasting sand and haven't tried it yet, but should work nicely.
Lay out the sand in a pile, toss on a few pounds of clay powder (don't bother with kitty litter products, it's been fired and as such loses most of its stickiness without extremely fine milling), mix it with a shovel and moisten. You'll know when it's right... it sticks together without leaving moisture on your hand, feels kinda soft when you break it (if it crushes suddenly, it needs more clay) and isn't excessively sticky (too much clay).
Too much clay and/or water reduces porosity and increases gas and can lead to steam voids in or on the casting, or even worse, bubbling of the metal.
Read up on sand casting (check my website and a slew of others on the webring and ABYMC links pages) and start with white metals, then aluminum, then try bronze. Just some nonsense castings, plaques or something for instance. Then try molding the bell.
You can also attempt it in lost wax, but you lose your pattern if you screw up. (You can make copies with the help of a silicone mold, but you need gallons of RTV and that's a lot of dollarsigns.) Here you take maybe 2 parts sand and 1 part plaster of paris, mix it with water and fill in around the pattern. Then fire in a kiln to 1200 degrees F over a few hours until it is uniformly heated and burned out. Then you can pour metal in it (mind that aluminum will cool VERY slowly if you pour into a mold this temperature).
Tim
-- "I've got more trophies than Wayne Gretsky and the Pope combined!" - Homer Simpson Website @

[I think you need to do more homework, Eric. I seriously doubt that Paul Revere used the sand-casting process to produce bells. It was only developed in the 19th century, and really didn't become common until the 20th, with the invention of matchplates and bentonite-bonded sands. It's much more likely that he used the traditional lost wax method that's still being used (with some modifications) to make large bells.] [Horse manure doesn't make sand stick together. But it is used to temper the clay used in traditional lost-wax casting techniques - here's a link describing how some people tried to recreate ancient casting methods: . If you don't want to use it, sawdust or powdered charcoal can be substituted.] [While there are plenty of resources on sandcasting, I really don't think this is a historically correct approach, even without the kitty litter. You might as well use ceramic shell. Here are a few links to historic bell-foundries, describing how they do it. They mention that this is the way it's always been done....]



[By the way, I came across this note in my research - is this Gellomer/Gillimore an ancestor? He probably shared the usual technique known to bell-founders at the time...]
Bell Founding "In 1769", says Benjamin Hobart, in his History of Abington (1866, p140-1) " a deserter from the Bristol Army, a bell founder by the name of Gillimore, was employed by Colonel Aaron Hobart". The town records on 8 January 1770 of Old Abington give his name as Thomas Tallamore, and records of Old North Bridgewater ( now Brockton) spell his name Gellomer. The reader can take his choice. Mr. Hobart says that his father sent representatives to Boston to teach Paul Revere to mould and cast bells, this being the beginning of the famous Revere bells that are now collector's items.
Andrew Werby

No way. Vermiculate is fluffy stuff; I've never seen a bag of Oil-Dri that even resembled that.
Also see the MSDS:
(NB: "fullers earth" today can mean various clays. "Fuller" means someone in the clothes-cleaning trade, where the absorbent clay is used to blot and absorb greasy stains.)

I don't know that there is any such thing as "true" kitty litter, but the cheap bentonite clay type is just the mineral as mined, which has perhaps 10 percent impurities like silica, not enough to matter.
Non-uniformity of sand grain size is not an issue for rough casting. If anything, it makes the packing stronger, as this is a general principle of logarithmically graded aggregate, e.g., in concrete, versus uniform sized. It can slightly affect surface finish. Blasting sand or especially pool filter sand is precisely graded in the sieve series, versus masons sand (ugh) or play sand, which aren't even qualified.
You're right that commercial casting products are better than improvising, but for many of us, we have no local sources, and it is more convenient, economical and satisfying to do it with what is at hand. I have done good work with it, definitely not "wasting your time".

This is just not true. All you need is a ball mill of intelligent design, cf Lloyd Sponenburgh ("sponenmill") and Dan Williams, a fun project in itself.

No matter how you cut the cake you have to have ball mill to do it right and thats gonna take time to make is it not....Those that don't have a ball mill have used everything from pieces of steel plat whopping the plates with a sledge hammer to crush up the KL, to blenders to coffee bean grinders to a steel bar in a bucket used like a mortar and pestel (sp?) Visit my website: expressed are those of my wife, I had no input whatsoever. Remove "nospam" from email addy.

Sand casting is old technology. Sand casting has been done for hundreds, if not thousands of years.
Paul Revere was a silversmith as his main craft and did later turn to bell casting. As I read and understand your post your main goal is to cast bronze bells. Bells have been cast in bronze for hundreds of years also, probably even back to ancient Chinese who were some of the first foundrymen. "Lost Wax" or Investment casting is ONE technique used to make bells but only a small percentage of all bells cast throughout history. I would estimate 95% of all bronze bells ever cast were cast in sand.
For molding in sand there are two main items to consider:
1) The base (unbonded) sand and what you will use for bonding.
The sand selections can be silica sand (for example) which will range in fineness from AFS GFN 35 to 120 or even higher. These are numbers that describe the average size of individual sand grains. The numbers DO NOT relate to the same sizes of grain on sandpaper. The "AFS" stands for American Foundrymen's Society, and the "GFN" stands for Grain Fineness Number. Lower number = coarse sand and higher number = fine sand. Coarser sand leaves more "perm" (short for permeability) to the sand which allows gases to escape easier in the casting process. Finer sand will reproduce detail more easily. For a starting project I would recommend about AFS GFN 70. Also you might encounter the term "3-screen" or "4-screen" sand. This refers to the distribution of the sand. In other words the average would be "70" but there would be a certain percentage of 40, 50, 70, 100 and 140 sized grains in the mix. Also sand may be round grained or angular shaped. This describes what the grains will look like under a microscope. In normal situations round grain works best. Sometimes for special applications we will use angular shaped grains. There are also other types of sand, besides silica sand, such as chromite, zircon, olivine, and other minerals. At the foundry where I work, we use AFS GFN 38, 3 screen, for the chemical bonded foundries and AFS GFN 70 for the "green sand" foundry. We also use an AFS GFN 65 lake blend, and AFS GFN 55 chromite sand, and occassionally AFS GFN 70 zirconium sand.
2) What will be used for bonding.
Green sand means a sand that has clay in it for bonding. Water activates the clay. You might use about 3-4% clay and about 1-2% moisture. You will have to experiment a bit. You might try to find a local foundry who will sell you some of their sand already mixed, if you decide to use green sand. There are more modern methods you might use to bond your sand. If you are making just a few molds, and can afford to throw out your sand, I recommend using an air-set or chemical bonded sand. Your results will be much more likely to succeed on the first try. There are many standard chmicals used to make chemical bonded sands. For example there is the phenolic-urethane process, the furan process, the sodium-silicate process, and many others. While the sand can be recycled, and used over, after teh mold is poured and the casting shaken out, you have to re-crush the sand back to grain fineness level to be able to recycle it. A big advantage of chemical bonded sand is that the mold surface which is in contact with the molten metal can be coated with a refractory. This is like a high temperature paint. It will give a nice surface finish so the polishing of the bell can be done easily, if you want a shiny bell.
A few years ago Glenn Ashmore showed how the casting process was done to make a huge lead keel. You might find his website on line with a decription of how it's done. At the foundry where I work we use the furan process, to cast parts weighing from a few pounds to 30+ tons. We coat the sand mold molds with a water-based coating and dry it. The surface finish is exceptional. We reclaim 97% of the sand with ball mills to recrush the sand for use all over again. In Ohio, from 2001 to 2003, the Verdin Company set up a mobile foundry. They cast a 300 pound bronze bell in the county seat of every Ohio county (88 in all). The molds were made in phenolic-urethane, and coated with a refractory coating of graphite and zircon. They would do the whole operation on a Saturday and Sunday and move to the next county once a week. The world's largest singing bell is in Newport, Ky. In 1998 a group a very philanthropic businessman in Newport Kentucky commissioned the casting of the World Peace Bell. This huge bronze bell was cast in France, brought to the U.S., and rung for the first time at the change from 1999 to 2000. It was called the Millenium Bell. I was involved in the early stages of the Verdin Mobile foundry and Millenium Bell projects.
Already answered above, but try chemical bonded sand if your making just a few but use green sand if you're willing to keep it going as a long term project. Or try both.
That would not have worked as described. The horse dung would have been a carbon product and I doubt if it had any "glue-like" properties, even if you threw it against the wall to see if it would stick LOL. But, they might have added small amounts of horse dung to act as a carbon additive to the sand. While pouring the molten metal into a green sand (clay bonded) mold, the carbon products will form a soot-like gas which automatically coats the interface between the metal and sand, with a thin layer. This formed layer prevents the molten metal from easily "burning-in" to the sand. What foundries use today are "seacoal" which is finely ground coal, or other carbon type products. I have seen wood flour, crushed corn cobs, and other materials used as well. Most likely the people in Paul Revere's foundry used a natural bonded sand. This would be a sand that might have come from a river bank, and contained a lot of clay in the sand. The problem would have been that this sand would have contained an excess of free water (i.e. water not being utilized by the clay). During pouring the free water turns to steam, and "blowholes" and a fair amount of steam can develop. Even a slight excess of free water will cause burn-in on the casting so the carbon in the sand (about 1/4 to 1% is plenty) would offset any free water.
I doubt they gave you the full details of what they are doing.
The clay is bentonite. There are two types, Western and Southern. Southern works great for thin section metals, and lower melting point casting, because that clay develops a lot of strength and you will find it retains its shape easier. Western has better hot strength. You can blend the two clays to customize a mix that works best. If you're making say 10 pound bells 100% Southern bentonite would work OK. You could add a little western if you have access to it but it's not important unless you're pouring larger castings.
You've only scrathed the surface of what you need to know. Please ask more questions and I think you will find this list will provide the answers you need.
Mark