Cast Metal Goblets
Optic Molds
(and Investment)

2003-07-20 Rev: 2003-08-29, -09-14, -09-23
2004-04-18 tune 2004-05-04 edits, 2005-03-14, -04-18; 2006-03-21, -10-03,
2007-01-01, -06-04, -08-09, -10-04, -12-06, -12-12 
2008-02-18, 2009-03-12 (minor), -08-02
2009-12-18, 2010-11-21
[Search on date pattern to find latest changes, more than one may be found.]
Split from MeltMetl

Return to Sitemap

I got into metal casting out of curiosity and to make forms to support glassblowing including molds and making metal shapes into which to blow glass.
Clay, Wax, & Plaster Molding

 Basic Lost Wax Example


WAX WELDING Spruing & Venting Step By Step


Latex Mold Compound HEAD MOLD
Fired Clay Core for casting Poured cast forms clay & wax Sand fill clay form lost wax
Casting Results & Projects

Casting a Goblet Stem & Foot

Second Round of Casting

Casting Results


Bad Casting


Hand & Foot Goblet

Large Brass Stems

Large Goblet Projects
Large Alum. Stem

Aluminum Blow Mold

Optic Molds 1 & 2 & 3 & 4



Related pages
Metal Center
Bronze Casting Overview
Excellent short video on bronze casting


BASIC STEPS -These are dribbled molding wax, put over shapes for the purpose of making cast brass goblet stems (to be added before brass casting) into which glass shall be blown. 2002-02


Wax originals for brass goblet forms.
These 3 images show the lost wax positive for the goblets to be cast in brass. The goblets proper were made in two parts, the foot and stem was molded in latex molds from an original clay model.
The bowls were made by dribbling casting wax over rounded shapes - clay and wood.   The sprues were cast in grooves in plaster, the pouring cone was poured in oiled paper.  Cones were since shaped in a plaster model for future use.
Silicone sealant was used to glue the wax to the aluminum foil which was taped around the outside of the tube made from conduit to keep it from leaking.. 
Wax positives of goblet stems with risers
Sleeve filled with investmentThis is the sleeve after filling with investment.  It is jewelry investment material. Investment mixes up quickly, using less water than plaster..  When heated to high heat (1350F), it becomes porous and water soluble.  It is still relatively weak and will leak out the side if not supported.  Usually stainless steel is used, but I find it too expensive in this size.  Steel conduit is used instead; rust has to be cleaned off between uses.  Wax positives of goblet stems with risers

Top 4

INVESTMENT - is a high temp casting material used in the jewelry industry.  It is a white powder that looks like plaster of Paris, but has less complications to its mixing.  When used for jewelry investment, it is placed under a vacuum while mixing or before and after pouring to remove air from the container and conform to the model, but I have not been doing this as I do not have the equipment and do not need this level of detail.  I am currently using KERR Satin Cast 20 investment, the most "forgiving".  [ ]
The directions which come with the investment are a curious alternation of metric and American measure.
Mixing Investment

  Investment is a white plaster-like material for making high temperature molds for casting metal.  Contrary to Plaster of Paris, after heating, it will dissolve in water which is used to clean the cast material; from my experience, it is much easier to clean up when it has been heated to the common mold temps (900-1100F) than when not (when I cast lead at 400 and did not heat the mold.)
Investment is purchased in substantial quantities (25-33-50 pounds) from jewelry supply places, Roseco being one.
  In my experience, investment requires much less babysitting than plaster, but rewards careful measurement. I always weigh the powder, using a small dietary scale with a pan.  Like plaster it looses volume when wetted down.  The result of mixing is a very thick cream which sets up only at a moderate pace, but flows more easily than plaster.  Unlike the plaster I have used, the investment does not seem to shrink. [Actually, it apparently expands somewhat while setting - cracking a thin cornered plaster mold - and shrinks a bit while heated, per Kerr.]

Mixing Plaster

Plaster of Paris is commonly used for molds because it gives good results, it cheap, and water tolerant.  It requires some care in mixing for best results. Plaster of Paris is available at most hardware, hobby and craft stores as well as pottery supply places.  It does not keep for long periods, taking moisture from the air and losing the ability to set, so buying more than will be used in a few months will waste it.  There are other forms of plaster, used on walls, etc., which are not discussed here - the word plaster means Plaster of Paris.

Plaster is commonly referred to as being mixed 2 parts plaster powder to 1 part water.  This is by weight, although it turns out that volume works fairly well.  Plaster gets hot when setting and should not be used on human skin, especially not faces.  Once mixed, plaster sets up firm very quickly (20 minutes or less), but it takes overnight or 24 hours to dry and get strong.  It will feel damp and cool to the touch until then, warm and chalky when dry. 

Determining how much to make because adding more to a mold is usually difficult because of the time it takes to mix more.  Extra plaster should be dumped onto paper or in a can or bottle rather than being dumped down the drain as it will set under water, in the drain pipes, possibly blocking them.  Plaster on mixing bowls and tools should be washed off in a basin before setting, otherwise it will have to be scraped off.

I have found that if the mold can be filled with water which can be poured off and measured, then that amount of water can be DIVIDED by 0.8 to get the amount of plaster powder by volume (in other words, plaster powder volume is greater and when mixed with water it collapses to 4/5ths its previous volume.)  Then water is half the amount of powder.  Art plasters available from pottery supply companies will have special ratios to support special features - low shrinkage, higher strength - and should be mixed as directed, usually by weight.

The proper method of mixing plaster for the most strength, fineness of detail, etc., is to measure the ingredients (cold water and plaster) by weight, not volume.  Then the plaster is sifted into the water to wet all the particles evenly.  Eventually, the plaster powder will start building up a mound on the surface which is about as much as you are supposed to put in.  The material is supposed to be let stand for 5 minutes, to "slake" - absorb water, before stirring smoothly and without introducing bubbles that will disturb the detail of the mold. The measure of readiness is dragging a tool across the top surface of the mix - when it leaves a groove, the mix is ready.  Depending on exactly how much water is added, the mix will stand up like soft frosting or slump flat.  If properly mixed, it will always slump flat when the mold container is tapped. 2003-02-06

Top 5

Neither plaster nor investment discussions commonly give a way of going from a water volume of the mold to the amount of solid material needed to mix.  Investment is to be mixed 38-40 ml/gm water to 100 gm.  By experimentation, I have found that if I measure the water volume of the mold (fill it with water and pour it into a measuring cup) then calculate by dividing by 0.8 that will give me the grams of powder and take that times 0.4 to get the water to match the powder - example: 100 ml volume, divide by 0.8 to get 125 gm powder, times 0.4 to get 50 ml to go with powder.  The result is about 100 ml volume of set investment. [Or since two of the calculations cancel: Water in mold divide by 2 for water in mix, that times 2.5 for powder in mix.  So 400ml water in mold, 200ml/gm water for mix, 500gm powder for mix, yields 400ml mix. Per Kerr's directions, the weight in pounds needed  is the number of cubic inches divided by 20.  For the circular flask I use, the volume is 0.7854 x dia. x dia. x height (Pi R2 Ht > Pi/4=0.7854)  A pound of investment (454 gm) with 6.4 oz (182 ml) water yields 22 (359 cc)  [a ratio of .79 to my 0.80] [16.39 cc = 1]
   Burnout is to be done on a cycle based on the volume of the flask. [During the last 1 to 2 hours of burnout, the temperature must be adjusted so that the flasks are at correct temperature for casting.]

5 Hour

8 Hour

12 Hour

2.5D x 2.5"H
(63 x 63 mm)
0.61 lb.
(201 cc, 277 gm)

3.5D x 4"H
(89 x 100 mm)
1.92 lb
(630 cc, 872 gm)

4D x 8"H
(100 x 200 mm)
5.02 lb
(1647 cc 2279 gm)

1 hr 300F
1 hr 700F
2 hr 1350F
1 hr Cast.Temp
2 hr 300F
2 hr 700F
3 hr 1350F
1 hr Cast.Temp
2 hr 300F
2 hr 700F
2 hr 900F
4 hr 1350F
2 hr Cast.Temp

Top 7


This is the result of the casting process, a merge of 4 images to show before and after cutting the risers free.  The upper row shows the goblet base; the lower the attempted cast of a brass hammer head and door handle stem.  The handle stem is hollow and useless.  The upper end of the hammer is bubbly and contains crud.  It seems obvious more metal should have been poured, perhaps hotter.  The goblet came out quite nice. [As to shape; further work showed it was porous and broke when used.  Note that good castings seem to be brass colored, while dark indicates problems.]  The saw marks will be filed down, the bottom drilled for a guide rod for the punty and the foot and top will be mildly worked to better sit and to hold the glass. 2002-10-10. [Further work suggests a major problem was failure to burn out all the wax in the mold and thus having it too cool.  Rather than too cool metal, the metal may have been too hot.] Castings, good and bad, after removal
This is the second casting, which came out in fantastic form out of the mold this bright.  Tonight I blew into it and it worked well. The flatter bowl of the one above caused problems in holding the glass.  Will see in the morning. [It came out great.]  2002-10-11

I have since learned that preheating the casting in the glory hole to expand the metal will result in the glass wobbling less during the working and the more eccentric the casting, the better the glass is gripped, very round adds problems.  2003-08-29

Successful casting after removal
This is the cast goblet frame after sawing off the sprue and vents and tapping for 8-32 thread.  A short piece of threaded rod screws into the end of a special punty built for the purpose, keeping it centered while working the glass blown into the shape. Goblet casting after cleanup and threaded rod added

And here are two views of the completed goblet, still needing some cleaning an other work, but more than what I wanted from a first effort.

The leaf effect on the bowl is accidental, because it was built up by dribbling and draining wax for coverage rather than intentional design.  The stem was more intentional, having been molded in soft clay and a latex mold made.  The overall goal was to make a metal shape and work the glass after blowing into it, making the flared lip as shown here and not sawing off the glass.  The glass moves a bit in the metal and I am asking around for solutions since shrinkage is likely in future pieces..
CraftWeb suggested that plaster was used in past in provided interior holes and silicone adhesive would work.  In querying CMOG  about Lalique, their mention of jewelry background made me realize that if design includes a couple of bendable features, the glass can be more crimped in place.


Goblet with glass blow into it and lip worked.
Problems noted: it is much easier to change the wax than the brass, had to grind off the bottom of the foot to make it sit without rocking. Tried hammering the "feet" to bend and no luck so far  Probably should have worked the form more before putting glass inside.  The brass is very heavy, not bad, but needs refining, perhaps.  Where I cut off the sprues will require grinding or shaping - better attention to how they are attached will reduce work. [2002-10-23]  A look on a forging site indicates that some brass forges well, but only over a narrow range of temps not far below melting and the best ones have no lead (and this stuff does) which melts first and leaves a crumbly result. [2003-08-29] Goblet with glass blown into it, lip worked       Top 8


Second Round
The goblets above were made with fairly standard jewelry wax technique: making up sections and heat welding them together.  Because of the distortions that occurred in making the wax on a clay mold and taking it off, I decided on a variation for round 2.

The mold, in the background at right, is plaster of Paris, made from a wooden egg sold in craft shops.  It was made using standard plaster mold techniques.

In front is a white casting from the mold partly covered with green wax.  The white is investment.  Instead of taking the wax off this core and putting investment all about it, I intend to build on the stem and then invest the outside, leaving the core in place.  The core being nicely cast round will be centered, etc. and of good shape.


Goblet stem molding
While I was on a roll, I continued with my optic mold development, extending the goblet technique above.  I made a clay outside shape on a wooden disk (as below) and cast that in plaster.  I then cast an investment core in the rubber mold made above.  After wetting down the plaster mold, I melted the wax and first poured the disk at the bottom, letting it set until it would support the core.  I used pencils of wax I had cast previously to center the core and then poured wax around to fill the mold.  Thus I have an outside shape I can use with various innards  [This mold and object are 2-3 times as big as the one above.] Optic mold outside casting

Top 9

This is the mold building process for the optic casting.  A core is investment cast in a rubber mold made on a carefully shaped plaster model. The most obvious problem in hindsight is the lack of a good key on the core to lock it to the shell. [Later, I am molding keys from investment in latex molds and sticking them into the wet cord casting.]  It broke free during melting the wax out. The line on the tube is to locate the cutoff point to make it shorter so it wouldn't require excess castable to support it from the bottom. [Later - using sand to support mold would allow partially filled tube to be used.]
  The wax was built up around the core, a long, tedious process interesting only for future reference.  Before I do it again, I am going to make an outside plaster mold for the optic to cast the wax into, holding the core inside. [2003-09-14 Having cast a couple of failures, I believe these must be cast the other way up.  Air got trapped inside the last casting, that was otherwise nice, and left a offset blob below the ribs.] More
Wax positive of optic with risers
Wax positive of optic with can, showing investment core Wax positive of optic mold - side view              Top 10
First partially successful cast optic.

The result of casting the form shown above is at left.  This is a poor casting.  The dark appearance is probably due to the wax not totally being burned out (not up over 1100F) and perhaps being too hot when poured.  The bottom of the inside is very sloppy and the casting uneven at the bottom.  When I tried plaster and later lead on the inside, both interfered with good punty work. 2003-05-21

I have learned that the core must be anchored to the body, so I now add a "key" to the back of cores to lock to the surrounding investment.  I have to work more on aligning the core and I messed up one casting by running out of aluminum because I rushed the melting. 2004-05-04 More


Top 11

Optic wax sprued in mistaken way to make pouring area Aluminum casting of optic
Optic casting interior showing result of investment core cracking 3
Optic casting showing wires and side view On a recent casting run, I made some flat wax plates instead of trying to glue the wax to the aluminum foil (1).  I also used brass wire to pin the core instead of an investment tail piece.  The pins are more standard, sort of. I drilled through the wax into the core leaving enough outside so the investment would probably support the core. Unfortunately, by the time I did this, the investment core was very, very dry.  The result being that in the multi-grooved optic (2) one the ridges broke away and in the tri-lobe optic (3) the core split and twisted letting aluminum leak into the crack  With very good eyes you may be able to see the wire pin just above the leak at 3 o'clock.  Wire pins show in other pictures.  They pull out easily.  2005-03-14

Top 12

In the ongoing process of learning aluminum casting and making usable optics, this shows the carved wood master (made by taking a shaped cylinder, putting it in a jig and using a router to carve grooves), the latex rubber mold built up around it and the plaster support that surrounds it.  An investment core is cast in the latex.  If the plaster is not used, the rubber mold sags under the weight of the investment. Not really visible is a very important marking line that runs across the top of the wood, onto the rubber and then onto the plaster - it allows lineup of the parts to prevent the small odd variations in wood and rubber to become major distortions when the parts are turned a few positions out of line.  The cast core is positioned in a plaster mold (above) and wax poured in to make the optic shape.  Sprues and vents are added and the whole invested.  What I learned from this pour was that the aluminum should be sprued with the opening up, because an air bubble formed inside, so the nice ribs end in a blob. 2003-09-23                           
Optic mold pictures

Top 13

Aluminum Blow Mold - The sketch shows a pot I saw at an exhibition of ancient Chinese ceramics at the Crow Collection of Asian Art in downtown Dallas.  I decided to make a mold to let me blow this and practice a larger casting.
 MAKE FOAM MODEL - I began with a plywood disk with a hole drilled at the center with a T-nut mounted in it.  A 1/4" threaded rod placed in that received roughly cut 1" bead board foam disks stacked with yellow wood glue between.  A nut and fender washer clamped the stack over night.  The hole left by the rod shows in the white object left of center.
  The rod was mounted in a drill press with a piece of wood clamped to the press bed to guide the other end.  A rasp was used on the spinning stack to even and shape it, an aluminum profile having been cut from the drawing.  Rather than the whole form, the shape was a blunt cone to permit removal of the glass straight up from a one piece mold.  The bottom was marked off in thirds.  A rasp was again used this time to carve down into the bottom to form the 3 legs and the curves between.  Bead foam makes a rough surface and beads tear out of it.  At a suggestion, latex paint was used on the surface, which helped, with sanding, but not completely.  The whole surface was coated with Crisco vegetable shortening for casting.
 MAKE PLASTER MOLD - Metal pieces were pushed into the foam and plaster was poured in a metal sleeve used for investment casting. When hard, a hacksaw blade cut grooves along side the metal parting and across the bottom and the plaster broken free.  The plaster mold was further smoothed and fastened with heavy rubber bands from blown bicycle tires.
Home made clay slip was poured in the mold.  The first time too little slip was used and the result was too short (top left), so more slip was made and a new one poured with a weighted bowl to push the slip up the sides to a deeper level. The slip was too thick or had too much sodium silicate in it so it was very jelled and extra was massaged to flow and scooped out with a spoon. (bottom image) The result was shaped while damp, dried and fired. 2008-02-18  It will be filled with sand and sand with sodium silicate to firm it.
Blow mold aluminum casting, preliminary work, clay slip bowl

Top 14

2003-04-17  These are some results of a good session on Sunday, 4/13.  The piece on the left is the first good one out of the second version of the home cast optic, after carbon coating it with acetylene flame.  Will get a picture of the optic.  The casting was not perfect because did not do a total burnout of the wax (to 1100F, etc.) before pouring. Two pieces from session 2003-04-13, including first successful from cast optic.

The third item is developing a core for a squat stemmed bowl.  The shape was worked up with plasticine clay  centered on a wooden disk with a bolt for an axle that fit in a jig to guide blades for shaping the clay.  The disk was turned by hand.

A cylinder was made of galvanized flashing held in shape with a large band clamp (hose clamp) barely visible in the right upper portion of the picture.  The disk was positioned inside the cylinder so the tip of the clay was just below the rim and the band clamp moved and tightened to cinch everything in place.  A layer of adhesive tape helped seal around the disk.  Metal duct tape was positioned at the diameter and stuck down to the cleaned metal.  Plaster was mixed and (rather sloppily) poured in.  When it was set, guide holes were drilled, slop was removed, spray fat applied as a separator and the cylinder stood on end to cast the other half.

Bowl stem mold casting

Here is the investment core made in the mold described just above, with wax added.  The mold is behind.  A foot will have to be developed.

2002-04-17  The foot was cast from a separate 3 legged mold, rather thick, and added and blended, producing the casting shown below.

Bowl stem molding
This larger cast goblet base was blown 2003-04-13 and clearly needs a bunch of refining.  The glass was not trimmed enough above the goblet and an attempt to work the longer material obviously twisted the glass and worked toward uneven.  I tried harder to get the glass to stay in place on its own, but it still pulled in from the sides and wobbled.  I will probably break the glass out of this one and try again. [Did, looked good, gave away for goblet grab at GAS Conference 2003.  Found that preheating the metal helped with the grip situation.] First blown attempt with larger cast brass goblet base
This is the first attempt at blowing into a cast aluminum base, in April 2003. Everything was done to help the process - the aluminum was made about as thick as I would like to see it, the glass was small (too small for utility) and it was worked quickly.  At first I thought it was blown to fit nicely, but when I began to grind base and polish the aluminum, the glass broke free of being stuck to the aluminum and it wobbles.  The mold for the aluminum was made by building 1/3 of it in Plasticine clay on a core of a wooden egg shape with a screw for the stem core and a thin wooden disk (from Michael's as I recall) on which to model the base.  Latex rubber was painted on repeatedly to build layers and when that was done, a plaster backup was added.  The clay was removed from the wood core and left in the latex and an inner plaster support was cast.  Casting wax was poured in the mold between the plaster backing and after repeating 3 times, the three parts were welded/melted together to make the lost wax positive. [See support and methods for larger version.]
   The egg was too small a choice.  I will probably spread the upper arms after molding a second lost wax positive. [Done on later mold.]
First aluminum cast goblet base, blown into in April 2003

This is almost a text book example of a bad casting. (The image is rotated 90° from how it was taken so it is shown as in the investment. The intention was to cast a tree structure - roots, trunk and branches - the tree being upside down on the right side of the casting.)  The black, I think, is due to wax still being in the mold, although the investment web site says the polished stuff at the bottom is the wrong stuff from too much wax and the top is correct before pickling.  However, the incomplete cast is all my fault. Somewhere in the process of making up the wax, I forgot the basics of spruing, so I have both the vents and fill linked together, so brass fills from the top and down to the bottom and into the branches and where the air gets compressed between them, no casting.  Also, the roots are incomplete because no vent sprues were added.  More correctly done, the rod coming down the left side should stop at the middle cross bar and a separate riser should run from vents at the bottom, up the right side, pickup the missing vents from the roots and come up the top. Having visited the Kerr site, I think I will try pickling this. [Didn't]  2003-06-01

Bad casting example             
House Key Alloy from MSDS (note range of components. ="leaded brass")
(file sent from Ilco-Unicam which makes keys & locks)
Copper7440-50-61.0 1 .0059 to 64.5   [mid 61.75]
Zinc7440-66-65.0 5.0034 to 44.0   [mid 39.0]
Lead7439-92-10.05 0.151.3 to 2.5    [mid  1.9]
Iron7439-69-610.0 5.00Less than 0.15
Nickel7440-02-01.0 1.00Less than 1 .0
Tin7440-31-52.02.00Less than .1
Aluminum7429-90-1 10.00Less than .05
Silicon7440-31-51.00 2.00Less than .01

 Top 16


Information on Latex Molding Compound moved to here

     ONE THIRD MODEL - These photos illustrate my technique for modeling and building a 3 way symmetrical goblet base. Just below is a mold in the style of drapery, covered with latex rubber. (more details below)  After three were made, they were heat welded together as shown right and further formed.  Not yet cast.  2007-06-04 Cast in Aluminum 2007-11  [Several additional models are shown further down the page, including plaster backing technique.]
Clay 1/3 original with some latex of drapery goblet Wax original of drapery goblet from 3 pieces        Top 17
 Below a wood board has been saw grooved at a 120 angle (B) and a piece of sheet metal folded to match and the latter glued in the former with Gorilla Glue (A). A bar of wood under the angle keeps it level.  Aluminum foil was formed around a foam shape from a craft store and plaster was cast and shaped on the back to make the white 1/3 form shown.  This was stuck in the V-shape with soft clay and plasticine clay was used to model one-third of the goblet (C), in this case trying to model the fall of drapery as seen in a bronze sculpture and in samples of light cloth. Making the two edges so they fit together is important.
When the modeling was done, a latex was painted on to make the negative mold.  A plaster backing was poured and the whole was removed and inverted with the clay still in place and the plaster core removed. (D) The edges were masked with blue tape (E) and thick plaster applied to replace the core and fit the angle of the metal.(F)  The pieces were separated and the clay removed from the latex. Upon reassembly, wax was poured between the two forms three times to make a set to be assembled as above. 2010-11-21  (Another example)
Drapery goblet construction sequence


HAND & FOOT GOBLET --  I am working on another cast brass goblet stem to blow glass into. As you can see, the upper part where the glass will go is three left hands, while the foot is three left feet.  The hand was sculpted in oil base clay on a fairly good sized wooden egg bought for the purpose.  Latex rubber mold material was painted on to copy the clay, a plaster backup cast around the latex and green casting was melted into the mold.  The curved inside was preserved by making a plaster inside plate that is pushed into the molten wax to get it into the fingers and wrist. Three copies were made and carved away at the wrists to fit together, then melt welded.
  The feet were also sculpted in oil clay, actually made one left foot and a smaller scale pair.  They were also copied with the latex, making 3 copies in wax, and melted/welded together at the heels.  After looking at the relative sizes and busy-ness of the two sets, I decided to use the single feet on the goblet and welded them in place and smoothed  the join, preserving effects that look like muscles (not too smooth.)  After I had done the assembly, I realized I would have been served well if I had made a rubber mold of the assembled feet to save the assembly step of the feet foot next time.                    
Goblet casting wax - 3 hands, 3 feet Goblet casting wax - 3 hands, 3 feet - view 2

Top 18

Wax tools and techniques are now here

In the past,  I have tried several tricks for holding wax forms inside metal tubes to keep them from floating, instead of buying (or trying to buy) rubber sprue bases for my conduit tubes.  Each has failed in some way at some time, including using silicone adhesive to the aluminum foil cover.  This last batch, I made flat wax plates and welded them to the form.  When they were smaller than the tube, a wire over the plate and under the edges could keep them from floating, but better is big enough to let the tube rest on the edges, which are then trimmed and the foil wrapped.  Looks to be working good. In hindsight, this is the normal way of doing it, but with the rubber cap (sprue base) having the rods mounted to it.
The best way to make flat wax sheet is to melt it on water and let it cool there.  Pouring on to water produces uneven thickness unless it is melted after.
Goblet casting wax - 3 hands, 3 feet - with sprues and vents  To the left is the wax original with the sprues and vents attached.  After this picture was taken, a flat plate of wax was added to hold the piece down.
 To the right, are four castings after burn out.  The flat plate added is visible as a recess.  The two with wire across them were assembled from two pieces of tubing to get sufficient height.  During investment pouring, they were held together with aluminum duct tape, but it will not survive full burnout heating, thus the wire which is twisted tight.. 2005-06-03.
Invested flasks after melt out and before burnout
  Dual feet?

  Here is the casting of the goblet above after the first blow of glass (it is loose in the hands) and after cleaning in diluted sulfuric acid - the brass had turned very dark during heating.  The glass will be broken out and the casting be more preheated before blowing into it so the metal has more expansion to close down around the glass. 2006-10-03


Finished goblet casting with first blown glass
  The wax working model for two hands holding a 4" sphere with the base being the cuff of a shirt with a button on it - actually a button edge for each hand on opposite sides of the cuff. 2006-10-03 Goblet wax - 2 hands in cuff

Top 19


Large Goblet Projects - I like the books of Dick Francis and one, To The Hilt, features a goblet of gold and gems made about 1860 in the style of King Alfred's time.  The piece has a quote from Bede's Death Song and is described as heavy and knobby with rough stones in it.  While reading this again last year, two goblet images occurred to me.  One, the more obvious, was to cast metal in the knobby form with holes for blown glass to show through, perhaps picking up color (which I rarely do) on the glass.  The other came from describing the rim as crenellated like the castle in which it is being viewed.  This led me to the fantasy of the castle in the air myth and fiction tales which sometimes becomes a castle on a rock.  I conceived a goblet with the stone walls of a castle floating on a pillar with an entry to give access to a tunnel path inside and around the rock to reach the traditional castle gate - not too practical since the tunnel could easily enter inside, but perhaps the castle had the gate and was moved to the rock. Still working on details, but progress reported here. 2007-06-04
Clay form coated with wax on left, wax shell on right, sitting on halves of mold.I finished off the old year working with wax for lost wax casting.  I have been making slip cast clay forms to put wax on and reduce use of investment cores.  It occurred to me that I might be able to use the same plaster mold to make a wax shell for greater creativity when working the wax not backed by the core.  So, having made another clay core after dropping and breaking a previous one, after dipping the clay core in wax, I wet down the plaster mold and poured wax inside and turned the mold to uniformly coat the inside and came up with a lovely wax shell. Clay form coated with wax on left, wax shell on right, sitting on halves of mold. 2007-01-01
I decided to use the hollow shell for the Bede goblet since I could probably push the wax out from the inside.  Looking at some of the stems I had already molded, I decided I needed one that was more formal, shorter and symmetrical.  I used plasticine clay as I usually do but the thin plywood disk was marked off in thirds and with various circles to insure symmetry.  After modeling it, I coated it with latex mold material.  I used heavy duty aluminum foil to form a cup the diameter of the disk in which to pour backing plaster.  The result is on the left. Note the black mark to line up the mold with the backing.  In the center is the latex pulled from the plaster.  The white coating is plaster pulled by the latex which should have been cured longer and coated with baby powder or other separator.  Fresh latex tends to absorb water and turn whitish, when it is weaker.  In this case, the damage is inconsequential.  The right picture shows a wax positive poured in the mold beside the raw wax bowl it will support. 2007-06-04
3 lobe foot mold making and example
These three pictures (below) show a rescue attempt which worked and a rough shod way of handling the plaster backing.  The problem here was that no plaster backing was made originally so it deformed on use when hung from the flange in the suggested manner.  Wax copies were distorted, leaning to one side.  Because I liked the shape, I decided to try this rescue.  After melting wax, I wrapped the rubber negative in a towel so my hand could stand the heat and to protect my arm from spills and added baby powder to ease removal.  I shaped the mold appropriately and poured in the wax, holding and shaping while it cooled.  When it set, I shaped a cup of two layers of aluminum foil over an appropriate sized cup and pushed in the sides to reduce the plaster needed. I poured in plaster and pushed the mold down into it, holding it with a plate on top.  Now the "proper" method of casting is to find dividing lines and pour in sections.  But the latex gives me flexibility and does not need detailed support.  So after the plaster was well set and no longer damp to the touch, I sawed grooves up the side and across the bottom and to a chisel and hammer to break the backing into three sections shown in the middle.  It is fitted back together to hold the rubber and tied with cotton string.  The wax inside while fixing was fractured while being pushed on while cooling so it was removed and melted.  The wax on the right is a casting from the mold representing a heavy tree trunk with bold roots.  2007-06-04
Wax latex mold repair support and example
These two pictures (below) show the other shell, the one to be a castle on a rock. On the left is the stem with a rocky surface and a built up column showing grooves for large stones.  On the right, is the dipped ceramic form that will make the bowl/castle shown beside the stem and several of the purple wax pieces modeled to be landings and entries for the path way.  The purpose of the white bowl is to not have to fill the inside with investment. I don't know if the hassle is worth it now that I can make the same shape in wax, but I will find out. Although the edge is very thin, the body is thick for carving.  At least I have learned about slip casting and may blow glass in clay.  2007-06-04

Top 20

Goblet stem, large aluminum #1
The clay shape is a trunk and several large branch shapes that will link around the cup.  The arms are reformed to align when the three pieces are combined.
While I am still working on the projects above, trying to settle the ideas in my mind and commit to doing something specific, to keep my hands and mind going, I shaped up a larger aluminum goblet stem.  Here I started with a plaster half molded in the same mold as the slip cast just above and carved away at the plaster to fit in the one-third holder.
  The one-third holder is a piece of stiff galvanized metal bent to 120 degrees and mounted in a matching pair of saw cuts in a 1x4 with Gorilla glue A 1x2 cut to fit and glued supports it level.  By modeling one-third of the object and cast 3 wax copies a full round can be assembled..
[All pictures can be seen larger by clicking on them.]
Goblet stem, large aluminum #1
This shows the V shape more clearly.  In the first two pictures the plaster with the clay form on it, has been covered repeatedly with a thin layer of latex rubber building it up to a thicker layer that is amber in color.
Goblet stem, large aluminum #1 I use metal tape intended for sealing ducts to make end plates to prevent the overflow of the plaster.  On the left, the mold has had plaster poured in, then as the batch thickened it was spooned and shaped to support the soft latex mold without breaking.
On the right, the project has been turned over and the latex, plaster core and green plasticine clay can be seen - the trunk at the left and the ends of the branches at upper center and center right - one is behind the little plaster blip mid-center.
Goblet stem, large aluminum #1
  Goblet stem, large aluminum #1 The back side of the mold is blocked off at the edges with blue painter's masking tape (very sticky, easy release) which is used on the other plaster for separation, the back of the core is scratched for better adhesion, the dry plaster wetted, and a thick coat of wet plaster is applied.  Then a gauze bandage square is opened up, a few tablespoons of a thinner mix is made, the gauze soaked in it and laid on the still setting thick coat and the remaining thin poured over and smoothed.  The gauze acts as reinforcement as in a cast for a broken arm.          Top 21
Goblet stem, large aluminum #1 When the second plaster coat is no longer warm and is drying, it is carefully pried free.  Here part of the clay model sticks to it and part stays in the mold.  When the clay is cleaned out, the two supports can be assembled with the latex mold between them and tough casting wax poured into the mold through the bottom of the trunk.
Goblet stem, large aluminum #1 The core plaster was damaged when it was made, so some clay is left in place to form the wax and save carving off the extra.  The latex mold is shown laid back on the support.  Notice that the inside surface is lighter than the amber shown in the first pictures.  This is because the plaster, still holding a lot of water, has wetted the latex giving it the milky color shown.  When this color, the latex is much weaker and allowance must be made in painting layers that tricky edges that might tear when damp during mold disassembly are made thick enough with repeated painting.
Goblet stem, large aluminum #1 This shows the outside of the latex and the inside of its support.  Besides the milky damp color of the latex the outside has pulled plaster from the surface of the support.  This can be reduced by applying baby powder (corn starch) or lubing the latex as when casting detail.. Since the support is mechanical and not involved in the detail of the wax, I don't do much. 2007-08-09

Top 22

Three pieces from the latex 1/3 mold are assembled and worked for smooth flow at the connections, shown here on the left to be cast in aluminum.  The other bowl shapes are to be cast in brass from junk keys trying various techniques.

Wax originals, various techniques

The goblet at right above is also shown to the right prepared for investment in an unusual way.  It is primarily a test piece for doing the one in the center above, which has taken much more work.  This was built up by dribbling wax on a thin shell of fired clay formed in a plaster mold from liquid clay slip.  Much easier to carry around than a solid investment cast form, I decided to try also to save investment. The bowl has been filled with dry sand nearly to the rim, then enough sand to make a 1/4" layer was mixed with sodium silicate (water glass) to make a sticky paste which was spread across and leveled.  The silicate reacts with carbon dioxide in the air to form a hard layer that holds the rest of the sand in place during investment - worked on test run, washed out nicely. 2007-12-06 Large goblet base built on fired clay slip bowl, packed with sand.
Wax cast bowl carved away and mounted on 3 lobe base On the left is the result of carving the hollow wax shell shown above and adding a cast wax base.  Carving was done with a flat brass tool mounted in a soldering iron and besides the shapes, the edges of the holes were thinned where the wax was thicker than desired.  This was vented (shown poorly several pictures down) from the inside through the holes to the bottom in an effort to preserve the outside and it should have been vented from the inside up and cast right side up with a sprue to the foot. That is came out okay is due partly to luck. 2007-12-12. 

Top 23

Vent failure or improper spruingThis is the brass version of the pink venting/spruing shown above with the sand and shows how close the casting (below) came to failing.  The brass was poured into the foot of the goblet, with a short sprue and poured to overflowing,  Because of the thin edge to the casting, the brass froze without rising fully up the vents.  The one on the left filled from bottom and top from overflow.  This ring was sort of fun to cut this off intact although I was prepared to cut through if I needed to.  A metal cutting blade in an inverted saber saw was used. 2007-12-12
This shows castings after first washing with hose to blast out most of the investment, before removal of vents and sprues, were cast upside down and both show vents failing to fill. Acceptable castings only by luck. Note the gray clay showing through the holes on the right.  I found it softer and less brittle than I expected.  I don't recall if I fired it to 1450F (likely) in the kiln or above 1800F in the glory hole in the mold run.  2007-12-12 Casting of carved wax bowl and dribbled showing sprues

These two brass castings below, shown mostly cleaned and cut free of sprues and vents, are the result of the two right hand wax bases in the large picture above.  The left hand one is the wax drizzled on the hollow fired clay bowl (also shown in the sand filled, pink sprued image.) The right hand one was carved from the thin wax shell. Since extra brass removal is not complete, the right is propped level with a scrap of glass. The white material in both images is the small remains of investment to be scrubbed with a wire brush.  2007-12-12

Brass goblet castings part way through final cleanup.

Below are three waxes for goblets which have been sprued with pink, more flexible, precast rods.  Each wax has been glued to heavy duty aluminum foil with Marine Goop to keep it from floating when investment is added.  Bent stainless steel sheet forms large flasks for each.  The flasks shown are made from 18 gauge stainless steel sheet from an odd piece left over from a commercial backsplash installation.  This gauge is a bear to cut and bend into a curve, but when I tried thinner non-stainless, it rusted to failure in one use.  The shape is made by bending to fit and is held with a stainless steel hose clamp. The aluminum foil has been folded up carefully and wrapped with a heavy rubber band cut from leaking bicycle tubes.  Burnout temps are higher than aluminum's melting temp and the metal is poured in from that end, so it is removed before burnout.  Investment is weak and will crack and leak molten metal, so the invested cans are set in sand for casting, not set on a hard surface.  Sand also catches any overflow.  If the waxes are weighed before before investing, the weight of the wax can be used calculate the weight of metal needed. 2009-08-02

Wax originals sprued and placed for investment pour in clamped sleeves

Below is a panorama of my glassblowing area outdoors setup to cast aluminum goblet stems.  The aluminum is melted in a cast iron dutch oven with short legs so the heat gets all around in the gloryhole to the left (the round ended unit.)  The flasks are burned out in the rectangular box to the far right which is my smaller annealer for glass. The large flasks are gradually heated first to 150C to melt out the wax into pans with water in them which are then removed, then to 200C to begin burn out, then to 500C then to 720C (1350F) for 4 hours and finally back down to casting temperature which is 200C (400F) for aluminum.  In the center, is a box of sand resting on two portable step shaped stands.  A flask is shown in the sand here and next below. 2007-10-04

Aluminum casting using kiln & glory - panorama

  These next four pictures show steps after the metal has been poured.  The long cuffed gloves are from a welding shop and only cost about $8 a pair.  The box is heavy 2x6 with a hinged lid and is used both for wet sand casting of ingots and supporting flask molds.  The smaller cylinder is a ring cut from 4" thin wall electrical conduit which I use for my mid-sized flasks.  It rusts, but doesn't lose so much metal as to make it worth me buying stainless steel flask, especially in this size.   As described above, the large flask is made by bending stainless sheet metal to fit and is held with a stainless steel hose clamp shown as a band half way down  The purpose of the conduit ring in this case was a failure on my part to put a riser on the wax to provide molten metal for fill as the casting shrunk on cooling.  I will have to cut the disk off.   The third picture shows the casting emerging from the investment as water soaking and blasting weakens the investment.  And the right picture shows the cleaned casting with the sprues and riser disk still in place. 2007-10-04

Aluminum tree goblet casting sequnce - 4 panel
Top 24

Spruing & Venting

Spruing and venting, that is adding sprues and vents to a wax for casting, is one part common sense, one part art, and one part practice.  A sprue is a way of getting metal into the hollow of the mold for the casting, while a vent lets air out.  A successful job results in a very nice casting while a bad job results in a useless one.

Common sense: The pathways for metal to get in have got to be generous so that the metal gets to all parts of the piece before getting too cool.  In getting into the mold, the metal has to drive the air out, so either the vents must connect to the lowest part of the mold with the metal flowing in from the top or the sprues must reach to the very bottom with the vents at the top and or the tips to let the air out as the metal rises.  The sprue receiving the pour must normally have a funnel shape at the top and hold enough extra metal so that the shrinkage that occurs when cooling will be taken up with more metal. (A flaw in casting immediately above.)  Recommended procedure, and I found out why, is to carry sprue path down and over to the bottom, with a riser beyond to hold further metal, which rises through the casting to venting at the top.  If metal is poured directly at a thick bottom in a unit invested upside down, the base may be bubbly or have insufficient metal.

Art: There are two aspects to the art - one is that the vents and sprues have got to work - letting the air out and the metal in.  The second is that they have got to be removable.  After the casting is done, each sprue and vent is attached to the casting and to other paths and these have to be cut off and then the cut surface made to look like the casting.  Without damaging the casting.

In the case of the tree structure immediately above, all the vents were attached on the inside so when they are cut off and ground flush, any further work will be hidden by the glass blown inside.  A side view (below left) shows that a cross vent not only supports the wax but may rescue the pour - because of the disk of metal, one vent was blocked as shown by the gap between the rod and the disk.

In the case of the 2 hands casting, the vents needed to not damage the details of the finger nails and knuckle skin.  The right hand goblet is intended to look like drapery hanging and fallen on the floor.  This was vented on the outside where it can be cut off and cleaned easily.  It might well have been sprued to the bottom and vented from the inside up which would have been harder to cut off because the blade would hit the inside of the cup. 

Alumnium castings after clean up showing spruing

Practice: Examination of the various castings, shown most clearly in the 2 hands, will reveal that almost all the joins to the casting are smaller than the rod of the vent or sprue.  It is much easier when working in a confined space to start a saw blade in the slight groove formed by the reduction thus avoiding a slip that might damage the casting.  And in fact, practice trains the worker to put the vents and sprues in a location where it is easier to cut them off.
  Further, after working for too many minutes in a confined space with a short stroke of a hacksaw blade, one quickly learns to give access to whatever blade is to be used.  I use a portable jig saw with a blade about 2" long moving up and down from a platform.  Someone with a band saw will use a different plan as will someone cutting by hand hacksaw.  Other choices include a Dremel-type high speed tool or a Fordham-type flexible shaft tool. 2007-10-04
  Note that I don't have a riser and that I am casting with a large base up, and small parts down from which vents are attached.  I am adding a pouring funnel to my base.  If my castings were less round and I wasn't pouring in the middle, this would be poor practice (it may be anyway.)  The recommendation for casting is that metal come in at the bottom and push air up ahead of it, that a good side sprue exist for taking metal from the pour point to the casting and that a riser be placed opposite the pour point so that metal can be seen to rise in it and so it and the pour point act as reservoirs inside the flask to provide molten fill metal as the casting cools and shrinks.  The two hands base shows the biggest problems with my method and that did not have the pouring cup on the base, so reservoir problems occurred. 2007-12-06 [Photos of damage.]

Top 25

Step By Step
  Three choices of wax form making.  
Model is section of whole Wax Model Drip or apply wax to core
Make or find interior model shape (wood, plaster, plastic) for half/third/fourth build of core. As shown, a plaster open mold can be used to pour a thin wax shell for further working.  
Make fractional model in clay on core Build a model entirely in wax Make a core on which to build wax
Coat the model with many layers of latex. Optionally, make latex mold to preserve for repeat wax Core may be solid wax shape, or investment core, or clay slip shell core
Make a plaster backing to support the flexible latex mold in proper shape (latex gives detail, plaster gives strength   Add wax to core to build original. Dipping, dribbling, welding add-ons. Cut away if that is useful.
If necessary, make a plaster core backing to permit pouring wax in special shape between core and latex mold. May require reshaping clay and latex against mold backing plaster if distorted when interior support is removed.    
Remove latex from clay and clean up plaster.  Let plaster dry.    
Assemble plaster and latex parts and tie or rubber band together.  I use wide rubber bands cut lengthwise from blown bicycle tires.    
Pour casting wax at as low a temperature as possible, to reduce latex damage.  Remove result.  Repeat as needed for full wax original.    
Heat weld parts into full wax original    
Clean the wax original and fix details Clean the wax original and fix details Complete all details
    If fired clay slip, may add sand and water glass rigidizer to take space.
Add sprues, riser and venting making allowance for container size and metal and air flow.
Glue wax to heavy duty aluminum foil with Marine Goop or weld to flat piece of wax larger than container.
Place container around model. Fold up foil and tape to secure from leaks.
If stacked ring container, tape joint and wire rings together firmly.
Mix and pour investment, filling from edge with allowance for shape and to reduce bubbles.
Allow setting time.  Remove tape, foil, and any Goop on investment, tighten adjustable band if used.
Remove additional investment around pouring hole if needed.
Arrange bread pans with water and wire grid to support containers and heat upside down to drain wax.  Heat to 220F (105C) first, then to 240F (115C) if needed.  Check for drain of wax in pan and in investment.  Remove drain pans and return to upright.  Tighten expandable bands. Clean out fill hole if needed.
Run burnout cycle, venting kiln to remove wax fumes and moisture. Lower to casting temperature and hold.
Fire up metal melting furnace, preloading some metal and adding flux or antioxidant suitable to metal.
Add additional metal if needed.
Prepare sand bed to receive containers (flasks). Clear path for moving flasks and molten metal.
Don protective gear. Turn on hose.
When metal is to temp, move flask to sand bed and level.
Open furnace and lift crucible to sand bed in pouring ring.
Break flux clear to pour.
Lift crucible and pour briskly to fill mold.
Set crucible down, clear pouring ring, return crucible to furnace and close furnace.
Allow flask to cool as required by metal.
Add metal to furnace to start further melting
or pour off excess metal in sand ingot mold and cool in shutdown furnace
Lift flask into water bucket when appropriate to boil off investment.
Slosh as needed to aid cooling and boiling.
Monitor next casting as flask cools
Remove flask to cleaning area and use blast nozzle on hose to strip investment.
Remove flask shell and continue cleaning of investment.
Examine casting and begin removing flash, vents and sprues
Continue cleaning to final, doing repairs if needed.
2007-12-06, -08-02


Contact Mike Firth