Member postings for PatJ

Here was a 2nd attempt to cast the green twin base.

This time used petrobond in the cope and drag, with no core, and that worked pretty well.

I was unaware of the correct pour temperatures for metals, and so my melt was too cool (pehaps a little under 1,300 F).

Result: Failure, incomplete mold fill. ( I think the correct term is "cold shut"?).

Lesson learned: Measure your pour temperature, and know what the correct pour temperature is (1,350 F for aluminum).

Also note that there was no reason to use a follower board for this pattern.  That shows just how clueless I was at the time.

Edited By PatJ on 03/07/2022 09:57:48

Aluminum Casting Demonstration:

A friend of my brother's got all excited about my new casting abilities, and insisted on a demonstration.

I was a bit nervous, for good reason, because I still had very little idea of what exactly I was doing.

I discovered (after the pour) that if the same diameter pins were used on either side of cope and drag, then the cope could be installed in a position that was 180 degrees off, and thus the upper part of the casting is shifted over about 1" from the lower part of the casting.

My brother's friend was not impressed, and he never came around again for any more castings sessions,LOL.

One of those fair-weather casting friends I guess; always there for the glory; never to be seen for the failures, LOL.

Results: Abject Failure, but important lesson learned. Use a 1/4" pin on the left side of the flask, and a 3/8" pin on the right side of the flask. Orientation problem solved.

Edited By PatJ on 03/07/2022 09:50:40

I was trying to learn 3D modeling, and learn how to make 3D printed patterns, and so I printed the frame for one of my dad's small oscillators.

At the time, I didn't even know how to machine metal, and so it was sort of like a blind man hiking through a mountainous region full of cliffs and crevasses. If you fall down, you get up and keep hiking.

Result: These two castings actually turned out relatively well, with just some slight porosity on the top of the bearing supports. I never used these castings because at the time I didn't know how to machine or build engines.

I did learn how to fill a flat 3D printed pattern to give the surfaces a nice concave shape.

Edited By PatJ on 03/07/2022 09:42:16

I made a pattern for the base of the green twin engine I was building.

I discovered sodium silicate could be used for a core, and so this core was made from sodium silicate and hardware store sand.

The remainder of the mold was made using petrobond.

I had no idea that hydraulic forces can lift the cope up off the drag, and so I did not use weights on top of the flask.

The cope lifted, some of the aluminum ran out.

The core was not secure, but it did not actually float.

The hardware store sand produced a dreadful finish on the inside of the casting.

Result: Failure on many levels. (No Pain, No Gain, as they say.)

I made some long ingot molds, and melted a damaged aluminum car rim.

This casting is what I consider in the "disaster" category.

I was just blindly stumbling along, trying to figure out exactly what I was doing.

I call this my "CLUELESS" phase.

Edited By PatJ on 03/07/2022 09:22:08

Edited By PatJ on 03/07/2022 09:22:34

3rd Attempt to Cast Aluminum:

I tried to cast another green twin flywheel in alumimum, and this time I used straight petrobond.

This flywheel turned out pretty decent.

Result: First usable cast aluminum part that I made.

A bit rough in places, but I called this a SUCCESS, relatively speaking.

Edited By PatJ on 03/07/2022 09:18:40

Edited By PatJ on 03/07/2022 09:19:15

2nd Aluminum Casting Attempt:

I had a 3D printed pattern made for my green twin engine build, and so I started trying to see if I could cast that piece.

I got the wise idea that I could improve the surface finish of the casting by spraying the petrobond mold with a mixture of alcohol and graphite.

This was my first attempt at a two piece mold, using the cope and the drag, and vertical sprue to feed metal to the mold cavity.

Result: The alcohol flashed and created a "surface of the moon" casting.

MAJOR Failure. Learn from my mistakes, and don't repeat these blunders.

2nd Aluminum Casting Attempt:

I had a 3D printed pattern made for my green twin engine build, and so I started trying to see if I could cast that piece.

I got the wise idea that I could improve the surface finish of the casting by spraying the petrobond mold with a mixture of alcohol and graphite.

This was my first attempt at a two piece mold, using the cope and the drag, and vertical sprue to feed metal to the mold cavity.

Result: The alcohol flashed and created a "surface of the moon" casting.

MAJOR Failure. Learn from my mistakes, and don't repeat these blunders.

This was my first attempt at melting aluminum, and I think this was my first casting ever, and before I built the furnace above.

I stacked some bricks in a circle, welded up a steel crucible, and melted some aluminum.

I purchased some aluminum ingots.

I carved out a crude flywheel pattern from wood, and a circle pattern, just so I could try to melt and cast something/anything.

I purchased some petrobond sand.

This is called an open-face pour, where you just pour molten metal into a cavity in the sand.

1st Aluminum Casting Proof of Concept: Success

1st Metal I Ever Melted: Success

I succeeded for the first time in creating molten metal, making a sand mold, pouring molten metal into the sand, and not dying in the process.

I bought a crucible, and built a very heavy (ludicrously heavy) cast refractory furnace, and a naturally aspirated propane burner.

Bought a 100 lb. propane tank.

Tried to melt iron.

2nd Iron Attempt: Failure (no combustion air blower used; I had no idea one was required)

My first attempts to melt metal were very crude, so don't laugh too hard.

I first tried charcoal and coal in a steel pipe, with a leaf blower for combustion air.

I held an iron rod into the heat, but no melting.

1st Iron Attempt: Fail

Edited By PatJ on 03/07/2022 08:46:11

I will split my casting blunders and successes out into a separate thread, since this may take a while.

I am going to post my blunders because sometimes it is as important to understand what does not work, as it is to understand what does work.

And perhaps seeing the progression from blunders to consistent castings will help others see that things can be rough in the foundry world when one is trying to learn the methods.

.

I would not worry about surface cosmetics too much.

That can be resolved over time.

If you don't control the metal velocity as you pour, you will get splashing, air aspiration, mold degradation, sand washed into the molten metal stream, etc.

Think of pouring metal like pouring a beer.

If you pour a mug of beer from a tall height, from straight above, you get a multitude of splashing, foam, etc.

If you pour from the lowest height possible, sometimes resting the lip of the crucible on the top of the sand mold, then you get a much more smooth pour.

Scraped ways on a lathe have an introduced surface defect, and that does not affect the performance of the metal, so it is not out of the question to use castings with surface defects.

I have tried various methods to stop the lid leak, and found it impossible.

The lid leaking will not affect furnace performance.

The total amount of combustion air being forced into the furnace will equal the sum of the gasses exiting the furnace lid and the gasses exiting the lid joint.

No big deal; all lids leak. Just ignore that.

.

Noel-

I always like to discuss foundry topics with fellow casters, because no matter how much one learns about foundry work, there are many clever individuals who have better ideas and methods.

Learning foundry work is an ongoing process, and I think there is always room for improvement.

I know of one forum that has an informal "Foundry/Casting Manual".

It would be nice to have an open-sourced modern foundry manual that could be edited by multiple people, such as how wikipedia is handled.

I will look for a pdf of the foundry manual that was written by a ME while he was in school.

It is a somewhat decent overview of the casting/foundry process, but needs a few items added.

I would not take this as the final word on foundry information, but it is open source and free, and covers a lot of topics.

http://prometheus-foundry.com/tutorial.html

.

Seems like I have been using a factor of perhaps 1.015 for shrinkage.

My memory is not what it use to be.

So if you want to cast a part that is 6" long, make the pattern 6 x 1.015 = 6.09" long.

.

If I have an existing part, and want to make a casting of it, it should be noted that the casting will be smaller than the original part.

The way I work around the shrinkage issue is to glue tongue depressor sticks or thin plywood onto the surfaces that will be machined, to build them up a bit.

You can do that with your pattern to give a bit more machining allowance.

I generally use the same shrinkage factor for all the metals I use (aluminum 356, bronze, and iron), which is about (LOL, I forget the number; I will look that up).

The shrinkage varies a bit between these metals, but not enough to make a difference for any part that I cast.

.

And the last option that I am aware of is to do what you have done, which is make your own greensand.

You could try naturally sourced sand and clay, or purchase sand and clay from a pottery supply house.

I tried some ready-made store-bought green sand with iron about 3 years ago, just to see what the results would be, and the surface finish was terrible, but otherwise the part (a flywheel) was solid.

Too much water in greensand can cause a steam explosion, so be cautious of that.

If you want to use greensand, and can machine off the surface, then you could cast the piece oversized, and trim off the rough surface.

It should be noted that all metal should be poured at the lowest temperature possible, while the metal is still hot enough to completely fill the mold.  Surface finish is directly related to metal pour temperature.

Overheated metal degrades the sand mold, and you can easily see this in action if you measure and record your pour temperatures.

I know of one individual who pours jumbo coins, and he has the most magnificent surface finish.

I noticed that he did not use a pyrometer, and so I asked him "How do you always have perfect pour temperature on your aluminum".

He said he observes the meniscus at the edge of the melt, and when the aluminum goes completely flat across its surface, then you should pour immediately.

I was shocked that this worked, but it did for him.  I have not tried this method.

.

Edited By PatJ on 02/07/2022 14:31:10

I looked for a source for sodium silicate, and found that it is used somehow in the pottery business, along with some fine grained sand that seems to be similar to OK85.

One source on this side of the lake for sodium silicate is Clay Planet.

https://shop.clay-planet.com/sodium-silicate-solution---pint-1.aspx

Clay planet also offers a catalyst that can be used with sodium silicate, so that the SS bound sand sets just as it does with resin. The catalyst eliminates the need for setting the SS sand with CO2.

JasonB mentioned a pottery material source on the other side of the lake, but I forget the name.

So purchasing some sodium silicate and sand from a pottery supply house is one option.

If you make your own CO2, which I have seen some do, be cautious with the pressures that can be generated.

One fellow put dry ice in a fire extinguisher container, and the pressure skyrocketed to something like 1,500 psi in a few minutes. Don't blow yourself up with CO2.

.

In summary, I tried oil-based Petrobond sand, with some success, but variable success due to spotty results trying to get the sand conditioned correctly.

I tried resin-bound sand using very dry commercial OK85 sand, and that worked well, and continues to work well.

A commercial chemical-rated respirator must be worn at all times when mixing resin-bound sand.

I then discovered a third type of binder with is sodium silicate.

The art-iron folks use this material as a binder with the same very dry OK85 sand.

Sodium silicate is not considered toxic like resin, although I would still wear nitrile gloves when handling it.

I am told that sodium silicate molds can be used with iron castings, and they have a surface finish almost as good as resin-bound sand.

.

I had the luck of having a local art-iron group in the City, and so I approached them for advice on foundry sand.

They use a lot of Petrobond for their art classes, and produce excellent castings with it.

I asked them about reusing burnt Petrobond, and they said yes, you can reuse the burnt sand; it does not have to be discarded. They have a larger commercial muller at the art-iron foundry.

What they do with their Petrobond is to use a thin layer of new Petrobond as facing sand, ie: the sand right around the pattern, and then they backfill around that with the used Petrobond. This method works well.

When they get too much burned sand in their Petrobond, they mix in some new Petrobond, and all is well again.

I still had problems getting my Petrobond to have good green strength, and so I want back to the art-iron foundry, and asked them what other foundry sand solutions the had.

Their molding sand of choice for iron sculptures is resin-bound sand, which is not reusable, but extremely flexible material, that sets into a hard mass, and can then be cut, cemented, drilled, almost like wood.

The sand that they use with the resin is called OK85, and it is a fine grain, and round grain sand, with a very low moisture content. I am pretty sure the OK85 is oven-baked, and its moisture content is extremely low.

Resin binder requires sand that has a very low moisture content.

I purchased some OK85 and resin binder, and overnight, all of my sand problems vanished.

Resin bound sand is not reusable (not easily anyway), and so my solution to that is to build custom wood flasks that wrap very closely around the pattern, thus minimizing the sand usage.

Resin bound molds can be very thin, and as thin as 1/2" thick in places, and so a I probably use 25% or less of the sand that would be used in a normal green sand flask.

.