EGroups discussing metal casting
Hobbiecast - Good information, lots of side chatter
Metalcasters - Folks who got tired of the Hobbiecast side chatter
The idea for this blast furnace originated with two books purchased from Lindsay Publishing. They are the Charcoal Foundry by Dave Gingery and the The Flowerpot Crucible Furnace by Lionel Oliver II. I also had a copy of Procedures in Experimental Physics by John Strong which contains two chapters that were useful for this project. The first is on small furnaces, the second on molding and casting items. This book is also available from Lindsay Publishing.
The following is a log of my experiments with casting metal using the information in the above books.
I stopped by Harbor Freight Tools in Richardson at lunch and purchased a 3½ gallon galvanized bucket to use on the furnace. The package from Lindsay arrived containing the Charcoal Foundry and The Flowerpot Crucible Furnace.
I cut the handle off of the bucket. I removed the tabs that the handle was attached to. I am keeping the handle to use with the lid.
After reading the books I decided to use the design from the Flowerpot Crucible Furnace since it could be built for less than $40 from common materials. On the way home I stopped by Home Depot and purchased 90lb bag of Masonry Cement (TXI Type N), 60lb bag of Masonry Sand, 10 ft of 10 inch galvanized flashing, 5 gallon plastic bucket, and a small trowel. I drilled a hole in the bucket at the right height and covered it with duct tape.
I cut a strip from the flashing that was 3" wide by 38" long. This strip was formed into a circle that was 11½ inches in diameter. This was secured with two small 8-32 machine screws. I cut out a circle of ¼" grid metal cloth to fit this circle with several tabs on it to secure it to the outer circle with more 8-32 machine screws. This serves the same purpose as the crossed wire discussed in the above books--it reinforces the concrete mixture. I already had the metal cloth on hand from the Furnace in a Thermos that I had previously built. I think it will make the lid stronger than the wire used in the books. Before finishing up for the day I made two tabs from the flashing to use to attach the handle to the lid when its finished.
I mixed up a batch of home-made refractory using 2 parts of Masonry Sand to 1 part masonry cement with enough water to make a fairly wet consistency to the mixture. I dumped this in the bucket and pressed the flower pot into the mixture. This is illustrated in the photo above. I set a small plastic flower pot filled with stones and two cans filled with lead on a 2x3 to hold the flower pot down while the mixture set. I then started mixing another batch of refractory to fill the lid. I added some broken terra cotta pot to the mix to function as grog. The lid is shown in the photo below. This was allowed to set up and cure.
I purchase a cheap 1600 watt blower style hair dryer. I took the dryer apart to remove the heating element. I notice that the motor was powered by a full wave bridge rectifier. This means that it is a DC motor, I checked and it works from 12VDC great. I removed two of the three switches, added a 5x2.5mm jack where the power cord would enter, wired it up and reassembled it. Great a working blower that run of 12VDC. I really wanted the 12V supply because I thought that I might want to take the furnace camping. I received the Lodge 16oz Cast Iron pot from Chef's catalog to use as a crucible.
I picked up some charcoal and a cast iron breadstick pan on the way home and prepared to fire up the furnace and melt 72 coke cans and some aluminum scrap left of from the Kite Aerial Photography rig that I built. I loaded the charcoal, crucible, and about 7 crushed cans. Lit the charcoal and turned on the blower. About fifteen minutes later I lifted the lid off and took the picture below. At first I didn't think that the cans had melted; however, when I poked them I noticed the pool of silvery metal underneath the can bottom in the photo. The can bottom looked like tissue paper. I then proceeded to add metal until the pot was 2/3 full.
I then poured the metal into the ingot mold shown in the next picture. I added more charcoal to the furnace and replaced the crucible and kept reloading it until I had melted all of the scrap.
When all was said and done I had a little over six ingots. They averaged about 5¼ ounces each for a total of about 2.2 pounds of aluminum. These ingots are shown in the following photograph. The next step is to make some flasks and prepare some molding sand.
I tested the modified hair dryer to determine the air flow it could provide. At 13.6VDC the velocity of air at the muzzle was approximately 6.9 meters per second. The cross section of the nozzle is 11.4 square centimeters. This results in an airflow of 0.5 cubic meter per minute. This equates to approximately 16.6 cubic feet per minute.
I melted down some aluminum angle iron scrap and more coke cans. I added small amounts of yellow brass scrap to the mix. I mixed up a batch of home-made green sand using the left over masonry sand and cat litter. I didn't wait for it to temper so when I poured the mold the cast didn't work out very well. The pattern and the casting are shown in the photograph below. I ended up with another five ingots in addition to the casting.
I finished melting the scrap and the remainder of the 96 coke cans, I also added about an ounce of scrap yellow brass. This poured out as another six ingots. I used up the remainder of the hardwood charcoal yesterday and melted today with cheap store brand charcoal briquettes. These actually work much better. The hardwood charcoal emitted a large quantity of sparks during the melting process. This cheap charcoal hardly sparked at all.
I am attempting to develop a formula for home-made investment plaster. My initial formula is 1 part plaster of Paris, 3 parts of Pumice, equal weight of water. I used this mixture to cast a mold of a small plastic gear.
The mold using the previous formula had very little integrity. It crumbled when I tried to remove it from the form. I am trying an equal mixture of Plaster and Pumice by weight. This is again mixed with an equal weight of water. I am attempting to use this to cast the same plastic gear.
I made a sand mold of a small pattern containing my initials on a 1/2" block. This sand mold worked better; however, the surface still shows bubbles and there was a small amount of dropping. The investment material held up to the aluminum very well; however, the mold itself did fill very well. There was almost no detail. I wonder if this is caused by the fact that it was an open faced mold.
After pouring the above molds I returned the pot to the furnace about half full. I added 10 3/4" copper pipe caps, 5 1/2" brass balls, and 1 1" brass ball to the pot. The Brass balls were alloy 360 yellow brass. I left the furnace with the cover on for about 10 minutes. The pot was glowing reddish yellow in bright light. The copper and brass was melted into the mix. I poured four ingots of this allow.
I mixed up a larger batch of molding sand earlier in the week. I now have about four gallons. I created a new flask that measures about 8.5x8.5x5". I also created a new pattern to serve as the base of another project. The pattern is a block 7.3" x 2.5" x 0.5". I put several coats of a wax finish on the pattern. I also fashioned two rammers to facilitate the molding process.
The mold went together much better. The pattern was much easier to remove. I cut a 3/4" sprue down to the pattern. I didn't use a gate, instead the sprue connected to the flat of the pattern. I melted 3 ingots of aluminum and 2 ingots of my aluminum, copper, brass alloy. I filled the mold and had enough left over to create two ingots.
The piece turned out much better. There was still a little dropping and the surface had a few bubble defects. The biggest problem was the sprue connecting to the surface plate. There was some deformation of the plate here. I need to start using a sprue and gate combination.
I have accumulated 6 cases of Coke cans. I started melting them down. Accumulated almost a full crucible with a little over three cases. I am using the Asbury Aluminum flux that I obtained from Budget Casting Supply I had to answer the phone just as I was preparing to pour ingots. When I returned the aluminum had mostly solidified in the crucible. I added more charcoal to the furnace and replace the crucible and lid. After about fifteen minutes it was clear that the furnace wasn't generating enough heat. I dumped a gelatinous mass of aluminum on the ingot mold and shut everything down.
After examining the cooled down furnace it was clear that accumulated ashes had restricted the air flow. I broke up the spongy mass of aluminum and remelted it. I noticed that this spongy mass looked a lot like the dross that I have been disposing of. This material melted down even better the original cans. The addition of more flux produced a black dross with some slight silver streaks that was of much less volume than I had been used to. I poured four ingots and finished melting the remaining three cases of Coke cans. I ended up with a little over eight ingots total.
I rammed up my small flask with a pattern for a flywheel. I am going to use this flywheel on a small electric beam engine. I started by ramming the drag completely fill with sand. I then added the Cope and dusted the mold with parting dust. I placed the pattern on the sand with the hub facing up. I rammed sand around and above the pattern. When this was complete I cut a gate directly down to the hub. I sized the gate at 3/8". I was able to gently remove the pattern with very little damage to the mold cavity. I repaired the mold cavity with some moistened sand. I reassembled the mold and set aside.
I melted 2 copper/brass/aluminum ingots, 3 aluminum ingots, and the previously cast plate. When this was melted in about 15 minutes I added some more flux and the Asbury Degasser (1/4 tablespoon). I mixed this in gently for about sixty seconds and skimmed the dross. I poured this melt into the mold and paused briefly when the aluminum appeared to fill the mold; however, I needed to add more when the level dropped.
When the flask was unmolded there were still problems with the piece; however, there did not appear to be any surface bubbles or the dropping that I have been experiencing. The hub was malformed, probably due to an insufficiently sized sprue and the pause in the pour. The biggest problem with the piece was the lack of uniformity in the thickness. I believe that this was probably caused by the molding procedure.
The crucible is showing a defect in the bottom where a hole is almost through. The hole is burning in from the outside. This crucible has lasted through 11-12 melts. It is clear that the fire is too hot for aluminum and probably oxidizing. Even so this crucible only cost ten bucks.
When the furnace cooled down I performed some repairs on it. The lid had developed several cavities ranging from pea to golf ball size. I patched these cavities with furnace cement. The lining of the furnace has developed numerous cracks and the top edge has developed several pea size cavities. I patched these by lining the inside with a thin layer of furnace cement and patching the top edge. When doing the top edge I also took the opportunity to level things up a bit.
Since the old crucible is showing signs that it might be about to fail I've decided to create another using the other Lodge Cast Iron pot I had purchased. The preceding photos show the pot being transformed into the new crucible. This is a very simple process and only requires a hack saw to remove the handle and a file to smooth the cut. I can get these pots for about $10.00. They are made of 1/4" thick cast iron. They hold about two pounds of aluminum comfortably and fit the 8" flower pot with room to spare.
Using the new crucible I've decided to make another attempt at casting the flywheel again. The following photo shows the sequence of events I performed to cast this flywheel. In the first image, I placed the pattern in the approximate center of the drag. I then molded it using standard procedures. This time I used 3/4" square sprue and riser. I cut a 1/2" gate between the sprue, gate and the pattern. The pattern unmolded with virtually no damage. I moistened the mold slightly to improved the adhesion. The casing is not perfect; however, it does appear good enough that with a little machining it should work fine.
I decided to try my hand at lost foam casting. I picked a 16oz styrofoam cup to serve as a pattern. I packed the core of this cup with molding sand and rammed it fairly well. I then put about two inches of rammed molding sand into the bottom of a steel bucket. I inverted the cup into the bucket so that the open area (packed with sand) faced down. I then packed molding sand around the cup filling up the bucket to a level of about one inch above the cup. I then cut a funnel down to the base of the cup with a bottom diameter of about 3/4" and a top diameter of about 4". After melting some additional aluminum, I poured the molten aluminum into the funnel. I didn't get much smoke and the puddle appeared to fill too quickly. I unmolded the bucket in about forty-five minutes to yield the cast piece shown below. The remainder of the styrofoam cup was intact. I'm not sure what the problem was.
I just received the Taig Microlathe that I purchased. Using this I cleaned up the flywheel casting. It turned out great! A picture of the final flywheel is shown below. Given all that I had heard about pure aluminum's poor machining charactistics I am very pleased with how well my alloy worked. I think that the addition of the copper has helped tremendously!
I needed some 1 1/4" aluminum round stock for a component I am making and didn't have any on hand so I rammed up a 1 1/4" dowel in a small bucket and poured the aluminum in this ad hock mold. It worked better than I expected, so I am going to make patterns for some of the standard sizes of round and square stock that I might want. It is sure quicker than ordering the pieces from the internet!
Given the problems with texture that I've been having with my home-made green sand I decided to cast the uprights with a plaster-of-paris mold. Using a tupperware container I made a mold of the upright pattern. After this mold had dried I removed the pattern--it was damaged in the process. I baked the mold at 250 degrees for 1 hour, 375 degrees for another hour, and then at five hundred degrees for yet another hour. I allowed the mold to cool for about 30 minutes while I waited for the furnace to heat to pouring temperature. I embeded the mold in a bucket filled with dry sand. The results of this project were much better than I expected; however, the plaster mold cracked badly in the process. The photo below shows the steps described above.
I needed a 1" square piece of aluminum stock about 1.25" long. Like the round stock above I couldn't find a local source so I decided to make one from my scrap aluminum. Given the success of the plaster molding process I made a mold using a wood pattern about 8" long. To try and prevent the cracking problem I baked the mold at 250 degrees overnight. I then baked it at 450 degrees for another three hours. When I was ready to pour I bedded the mold immediately into a bucket with dry sand. The piece came out fine; however, the mold cracked in the oven--before I even had a chance to pour molten aluminum. A picture of the piece after a little clean-up is show below.
Given the difficulty I've had with the upright and the shape of the beam on my beam engine I decided to try a lost wax casting of the patterns. I purchased some RTV silicon rubber from Micro Mark to make master molds of the patterns. This worked out well and allowed me to make several paraffin versions of the patterns. I added sprues, gates, and risers in wax. Placed this inside a large food tin and filled it with my home made investment plaster discussed above.
After this mold air dried for about two hours I placed it, inverted over a tray, in the oven and baked it for about three hours at about 200 degrees. This removed most of the wax from the mold without smokin. I then set up a small steel plate on a electric hotplate outside. I inverted the mold on this plate with some steel stock to raise it off the surface and allow the remaining wax to drain. I turned the hotplate on medium. I placed two bricks on either side of the plate to support a terra cotta pot which I inverted over the whole thing. This serves as an impromptu burn out oven. I ran this for about two hours until there was no more smokin. I then cranked up the hotplate to high and ran for about six more hours.
The two castings of the uprights came out perfect; however, the beam was deformed slightly. This was probably caused when I filled the mold. I'll need to make another attempt on the beam. Other than that my first attempt at lost wax worked out very well.
I've been distracted a little from my electric beam engine while I try to build a dividing head for use on my lathe. The first component for the dividing head is a frame piece that I am attempting to cast as a solid piece. While pouring this casting I decided to add a block pattern to be used as a boring bar holder for the lathe. The casting as drawn from the sand is shown below.
The flowerpot furnace is starting to look a little worse for wear. The patching that I did with the furnace cement has crumbled off. I decided that it is time to build a replacement furnace. The new furnace has two design criteria. First, it should be built with commercial castable refractory. Second, it should be big enough for a #4 "A" shape Silicon Carbide crucible. I found a source for castable refractory in Fort Worth for $35/75lb bag. This should work well. I chose the #4 crucible because its the smallest cheapest Silicon Carbide crucible that Budget Casting Supply sells.
I located a metal bucket 12" tall, 13" wide that should work well. I started modifying the lid for holding the refractory. A while back I purchased a used blower from All Electronics. This is a pretty large blower; however, I created a reducer and flow control device that should allow it to work well. A photo of the blower as supplied is shown below. When I finish wiring the blower I will post some photos and measurements of its airflow.
I was unable to make it over to Ft. Worth to purchase the refractory for the furnace; however, I did make a little more progress on the blower. I installed the blower on a wood frame and have begun laying out some sheet metal to create a case to house the wiring, fuse, and switch for the device.
I received the order I placed with Budget Casting Supply for the #4 SiC crucible and the electric element with infinite range control. When I finish the new charcoal furnance I want to build a small electric furnace along the lines of Gingery's "Lil' Bertha". This will be used for heat treating and burnout. And maybe for small aluminum runs. The crucible is shown in the photo below.
I finished putting the new blower together that I obtained from Alltronics. It comes wired for two speeds, Hi and Lo. I fashioned a reducer and flow control with a 1 1/4" nozzle. With the flow control open all the way the low speed setting produces 41mph flow and the high setting produces a 63 mph flow. The flow control when closed all the way reduces the airflow to less than what I obtained from my modified hair dryer. This range of control should work much better with the new furnace I am constructing.
I've been getting a few requests for the name of the firm in Fort Worth that sells castable refractory. So here it is:
Wesco Refactories Inc
410 E Magnolia Ave, Fort Worth, TX 76104
Phone: (817) 429-4331
This page and all of its contents are copyright
by Walter Anderson
Return to Home Page