Avoiding Problems when Casting Aluminum

[GreenTwin]

Base Pattern in Steel - 2014:

The first pattern for the base was made from balsawood, with the support webs 3D printed.
I made a new base pattern from steel.
Patterns are removed from resin-bound sand once the sand has set and has become hard.
A small slide hammer is used with a light impact to break the pattern away from the mold.
Balsa patterns won't last long with resin-bound sand.
.

 

[GreenTwin]

13th Aluminum Pour - 2014:

I went back and tried the long mold with aluminum, since aluminum can be superheated quite a bit above its pour temperature, and gray iron cannot when using an oil burner.

I was able to get a complete fill of the parts with aluminum.
I cast 1/2 of the flywheel to make a pattern, but in reality, the pattern used to cast another pattern needs double shrinkage, and in this cast my cast metal pattern is too small.

The large flask and pattern layout was so cumbersome that I went back to the drawing board to look for a more manageable approach.
These parts were usable and had good surface finish.

Lots of runout onto the concrete.
This large flask size is pretty harde to manage, and produced an excessive amount of flash.
.

 

[GreenTwin]

14th Aluminum Pour - 2015:

I set my kiln up for heat treating aluminum, and neglected to buy a solid state temperature controller.
I overheated my two aluminum base castings, so I cast two more of them.
These both turned out perfectly, and I did not even weight the top of the molds (I got lucky that the cement held).

Things were becomming more routine at this point.
23 lbs of aluminum melted in 12 minutes, and no casting defects in either casting.
Very little flash either.
.

 

[GreenTwin]

Cast Iron Pour No.04 - 2015:

I forget exactly why I melted this gray iron.
I think I was testing my new dual burner arrangement.
I cast some ingots, and then cut them in section, to verify that they were solid, with no defects or inclusions.

I was still experimenting with oil burner types/arrangements at this point, and did not have an exact oil burner setting figured out yet.
Some things I learned are that you can melt gray iron that is extremely rusty, and with paint on it, and still get a perfectly clean casting.
The paint and rust all comes out as slag.
I think for aluminum, you want that scrap to be as clean as possible, but aluminum does not rust, so that is not a problem.

I started using a clock to time the melt, and recorded fuel usage, and time to reach pour temperature.
I started using ferrosilicon in the iron melts, but was using too much, and getting excessive shrinkage and hot tears.

I was using a 1/2" steel rod to stir the melt (you can stir iron melts, but don't stir aluminum melts).
I did not realize that molten iron will dissolve steel within about 30 seconds, when you are at pour temperature.

I added a damper on my combustion air blower, to control the amount of combustion air entering the furnace.

Note that you generally don't ever melt iron just for the sake of making ingots, since it takes a lot of energy to melt iron.
.

 

[GreenTwin]

Cast Iron Pour No.05 - 2015:

This was an attempt to cast the green twin cylinders in gray iron.
I used a ceramic filter at the base of the sprue, and I did not have the superheat I needed in the melt to pass through that filter, so I got a partial mold fill.

I only had one cylinder pattern, so I made twin mold halves, and cemented them together.

I rebuilt my pouring shank crucible retainer, since that one slipped one time and let an full #10 crucible of iron spill out onto the concrete (no explosion, but some brown stuff in the pants).
I added a heat shield to the pouring shank, and added a pouring handle facing straight down.
Having the pouring handle close to the crucible gives good fine pour control, and the downwards handle puts the wrist in a more natural position for the entire pour.
I control the pour with my right hand on the handle.
.

 

[GreenTwin]

Cast Iron Pour No.06 - 2015:

This was a second attempt to pour the cylinders in gray iron.
This was another failure.
I did not have the cast iron technique figured out yet.
I don't remember exactly what went wrong with the pour, but it did not work.
.

 

[GreenTwin]

Boat Shaft Bronze Cylinder Pour - 2015:

I attempted to cast the cylinders in boat shaft bronze.
I was ignorant of most metal alloys at the time, and I thought all bronze alloys would have low stiction.
I was lucky this attempt failed because I found out later that this boat shaft bronze machines like steel, and has an extremely high stiction factor, so it would not have made usable cylinder material.

I did not weight the top of the mold, and so the mold halves separated during the pour, and the metal in the cope ran out into the catch pan, and also onto the concrete (no explosion).
I have been told repeatedly that spilling molten metal on concrete will cause an explosion.
I have spilled aluminum, brass/bronze, and gray iron on my driveway numerous (unintentional) times, and never had an explosion.
That is not to say that your concrete driveway won't explode, but I can definitely say that my driveway has never exploded.
.

 

[GreenTwin]

15th Aluminum Pour - 2015:

I decided to cast the cylinders in aluminum, since I did not have the iron melting/pouring technique figured out enough to get a predictable pour with a defect free iron casting.
This cylinder casting went pretty well..............except, somehow I introduced a slight mold half misalignment, perhaps 1/8".
The casting looked good, but I could not live with a 1/8" misallignment, so these cylinders went into the scrap bin.
I had to diagnose exactly what was causing the alignment problem, and fix that (I was using some inserts that were suppose to provide accurate registration buttons in the mold, but they did not work out well).

The sprue was still excessively tall.
I think I had a ceramic filter at the bottom of the sprue.
.

 

[GreenTwin]

16th, 17th, 18th, 19th and 20th Aluminum Pours - 2015:

I redesigned my flask configurations, to use a radial pattern arrangement around a central sprue, and this worked well for the small miscellaneous parts. I did five back-to-back pours, at about 12 minutes per melt.
This was the first time I did multiple pours on the same day.
All of these parts turned out well, and were usable.
.

 

[GreenTwin]

21st Aluminum Pour - 2015:

I solved the cylinder mold alignment problem, and recast four more cylinders, which turned out well.
.

 

[GreenTwin]

22nd Aluminum Pour - 2015:

I was starting to feel more confident with aluminum pours, and was getting repeat high quality castings with aluminum, so this boosted my moral.
I was feeling pretty bummed about not being able to get consistent gray iron castings.
At this point, it was either finish the engine in aluminum, or not finish the engine, since I was unsure if I would ever figure out how to make consistent high quality iron castings.

I think thee were spare cylinders, and my buddy in Canada ended up needing one (I mailed him a few extra).
In these photos is one of the bases after it had been lightly buffed, but before the gates had been ground off.
.

 

[GreenTwin]

The Zamak27 Melt - 2016:

I had heard so many things about the alloy Zamak, especially the new alloys of it, such as Zamak 27.
Very strong metal, with good mass (much higher mass than aluminum), and supposedly good machinability.
I decided to try out this metal, since it has a much lower melting point than aluminum.

The results for this jumbo-coin casting were not good.
The coin is a replica that is about 3" in diameter.
A buddy of mine makes some fantastic aluminum castings of jumbo coin replicas.
I have not been able to match his work yet.

The zamak melt seemed to take about the same amount of time as an analogous aluminum melt, and the surface finish was not nearly as good as I have gotten with aluminum 356.
I tried drilling the zamak, and it just melted in front of the drill bit (I am told that coolant should be used).

I kicked my Zamak27 alloy ingot to the back of the shop, and have never used them again.
I much prefer just melting aluminum 356, and I could not even notice the lower melting point of zamak.
.

 

[GreenTwin]

23rd Aluminum Pour - 2016:

I got bored with making green twin castings, and wanted to try to make a jumbo Cretors flywheel.
The spokes were cast separately using a single hand-dremel-carved wood spoke.

This was a proof-of-concept thing that unexpectedly turned out very well.
I had some slight shrinkage in the rim, but I could easily correct that with a riser or two.
This rim I think it usable, but I will recast it in gray iron.
.

 

[GreenTwin]

23rd Aluminum Pour - 2016 (continued):

.

 

[GreenTwin]

Cast Iron Pour No.07 - New Low Mass Furnace - 2018:

My first iron furnace was very high mass, perhaps 400 (+) lbs of dense refractory.
I was clueless about efficient furnace design when I constructed my first iron furnace.

I made a new low-mass iron melting furnace, using 70 lbs. of Mizzou refractory for the hot face and lid lining, and then a layer of low density fire brick, followed by two layers of 1" ceramic blanket.

This was a quick test of the melting capabilities of the new furnace, and so I just poured a couple of ingots.
Around this time, I finally figured out exactly how to tune an oil burner for maximum heat output, and so I finally was able to get a consistent 1 hour melt time with a #10 crucible of gray iron, and consistently good/high pour temperatures.

The new furnace along with the knowledge of how to correctly tune an oil burner was a real game changer for me as far as being able to make consistent high-quality iron castings.

I tried measuring the temperature of the outside of the hot face with 2,500 F thermocouples, and I ended up melting all of the thermocouples.
The back of the 1" thick hot face gets a lot hotter than 2,500 F.
.

 

[GreenTwin]

Cast Iron Pour No.10, 11, 12 - 2018:

I did a series of demonstration pours for various family and friends in the art-iron casting community, and just poured ingots.

I decided to try some greensand with iron, and purchased some water-based greensand, which is sand and clay.
This greensand was supose to work well and give a good surface finish.
I cast the Cretors replica oversized flywheel again, and the surface finish can only be described as horrifying.
I don't think I have a photo of it, but the surface of the moon looked smoother.

The water-based greensand went into the dumpster.

The new furnace worked exceptionally well, with consistent melt times and very good pour temperatures.

.

 

[GreenTwin]

Cast Iron Pour No.13 - Metal Museum "FIRE" Exhibition - 2019:

The art iron folks requested that I demonstrate my furnace and oil burner at their annual FIRE art-iron event, and so I disassembled my furnace, hauled it across town, and reassembled it (glad I made the new furnace modular).

I cast a "FIRE" plaque, and also poured into a previously used art-deco resin-bound mold, just to have something to pour into.
I did not have time to make a mold other than the plaque.

The furnace and burner performed flawlessly, and the pour went well.
The art iron folks use cupolets that are fired with coke, and most were not aware that you could melt iron with an oil burner.
None had ever seen an oil burner in operation with an iron furnace.

The FIRE exhibition was a multi-day event, attended by art-iron folks from around the country.
This was the most fun I have had to date pour iron, even though I did not cast anything useful.
I had never demonstrated before a crowd, and I was extremely nervous that something would malfunction, or I would drop a crucible full of iron, but it all went perfectly.

You can see the 1" thick Mizzou hot face, and the insulating fire bricks around it.
Two layers of 1" ceramic blanket then were wrapped around the insulating fire brick, and finally a stainless steel shell.
Most of the outer stainless shell remains cool to the touch during a melt, which is a testament to the efficiency of the furnace, as far as how much heat gets transmitted out the furnace wall.
The efficiency of an oil-fired furnace is about 10%, which is low, but not unreasonable as far as total fuel used for a melt, which is typically about 2.7 gallons of diesel.
.

 

[GreenTwin]

Cast Iron Pours No.14 - Straight Edges - Late 2019:

I made two pours late in 2019.
One was a straight edge and camelback for a buddy of mine in Ohio, and a second pour was to make a replacement bearing cap for a buffing machine.
Both of these castings turned out well.

Here are the straight edge flasks and molds.
I came up with a new snap flask design that hinges open, so that the wood flask can be removed from the mold prior to the pour, to avoid any damage to the wood from spilled molten metal.
I also purchases a used commercial kitchen mixer to mix the bound sand.
I tried a three-piece flask also.
.

 

[GreenTwin]

Cast Iron Pours No.14 - Straight Edges - Late 2019 (continued):

I mailed off the raw straight edge castings to my buddy in Ohio, and he said they machined easily with no hard spots, and only a few tiny inclusion spots that were small.
.

 

[GreenTwin]

Cast Iron Pours No.15 - Bearing Cap - Late 2019:
The bearing cap also turned out very well.
Someone mailed me the their broken cap, and I patched it up, and added some machining allowance by gluing on some thin plywood to some of the surfaces.
I did not have time to do a perfect fill job on the pattern, and so I ended up using some blue painter's tape, the outline of which can be buffed out of the casting.
The owner of the bearing cap wanted to preserve the original numbers/letters in the top of the cap, so I had to work around those.
The cap turned out well, and the owner said it machined easily.
The added wood material were tongue depressors from a craft store.
.