Ball Hopper Monitor - Casting Project

[GreenTwin]

I am about to start the first corebox half 3D print.
.

 

[GreenTwin]

This print is a little slower since I added supports under the domed and flat spots to try and get a slightly better interior finish.
Looks like I forgot to change the infill setting, and am using the default 20%, which is way too much infill for this application.
No wonder it is running slowly.
.

 

[GreenTwin]

20% fill is way too slow.
Nothing to do but wait it out.
So annoying; this is taking a very long time.
Uses a lot of filament too.
.
Edit:
Looks like the hopper will be about 6 lbs aluminum, or 18 lbs in gray iron.
This engine will need to be on a cart; I can start to sense that.
.

 

[GreenTwin]

Print No.04 has been started, which is corebox #2.
.

 

[Jasonb]

If it is going onto a cart you probably won't want that sub base that you drew up. Let the flywheels hang below the cart sodes as it is quite a top heavy engine anyway.

 

[GreenTwin]

I want to be able to set it on a table, on the base.
We will have to see what the total weight is though.
If I can't lift it, then that is a bit of a problem.
If the table collapses, then there is a second problem.

The gas tank in iron is 7.2 lbs.
One flywheel is 21 lbs. in iron.
The cylinder/frame is 31 lbs. iron.
Water hopper is 18 lbs. iron.
Base may be 15-20 lbs. iron.

So that is almost 120 lbs. total.
That is starting to push what I can deadlift to tabletop height.
Total height on a base is 28".
This build is sort of pushing the concept of "model", and getting into the grand model scale.
I fully anticipated this weight, and previously decided to go for it.
I have looked at sizes of castings as far as what I can machine, and also as far as my crucible maximum size.
I will have to farm out the machining on the 14"flywheels, but I have someone I know who can do it easily.

It is not out of the question to cast some of the parts in 356 aluminum, expecially things like the base.
I much prefer gray iron though for engine parts, since it machines, drills and taps so easily, not to mention wears extremely well.

The bilge #30 I use is 100 lbs iron brimfull.

I have been considering building a mini folding manual forklift, and the footprint would be about 12" wide, and 16" long.
Hand-crank hoist.
I was a small electrical one, but a manual one would suffice for me.
I would either stand on the back, or use a water tank for ballast.
I have some full size small engines that sometimes I want to hoist up into the car, such as a Baker VJ.

Here is the 3rd print, which is the corebox with the window.
The outside is ok, but irrelevant because this is a corebox.
The interior is ok, but needs some filling work, which was expected.
The core will define the interior of the hopper, and the looks of the interior are not super critical, since it won't normally be seen anyway.
I think I will use a 1" diameter wood dowl through the center to create a vent hole.

 

[Jasonb]

By the time you have included the second flywheel, pulley, head, bearing caps, exhaust, valve block, etc and then all the machined parts like crankshaft, piston, conrod, gears etc. You will be closer to the 200lbs mark.

The other big item to machine will be the cylinder/frame at about 14" across the feet diagonally you are not going to bore that on your lathe

 

[GreenTwin]

Here is some info I found on the Little Pumpjack Baker Monitor, and it says 320 lbs.
That is with one flywheel at 18" diameter.



So I suspect you are right about it being in the 200 lb range.
Not really tabletop material.
I will probably build a small frame with wheels to go under the base.
I really like an engine sitting on a base, with flywheels clear of the ground.
The floor space that a cart takes up is too much for what I have in my shop.

As I recall, you (JasonB) spun your frame in the lathe chuck, I guess to machine the bottom of the feet.

Myfordboy bolted the flat side of the cylinder to the lathe carriage, and used a bearing in the bottom of the crankcase, so he could line-bore his crankcase/cylinder. In this configuration, I think my frame/cylinder would fit on my 12"x36" Grizzly lathe.
(Photo my Myfordboy)



Another Ball Hopper Lone Star build shows the cylinder/frame bolted to what looks like a horizontall mill table, with the mill bit cantilevered out to bore the cylinder. I don't have a horizontal mill.

Here is a similar boring bar arrangement from the Southbend manual.
I think this is the way to do it.
For the bottom of the feet, I think I am going to remove the draft angle from the pattern halves, and cast the bottom of the feet flat.
Then I can just sand off the bottom of the feet, and use some bluing dye to level the feet exactly.




The crankshaft bearing will be poured with babbitt, and so the temporary shaft will be squared with the bore prior to pouring the bearings.




I am not sure about truing the top of the cylinder.
I will have to think about that one.

.

 

[GreenTwin]

The bottom of the frame at the feet is 6.3 inches by 11.7 inches, but requires a circle diameter of 13.14".
I am not sure if my Grizzly has an exact 12 inch swing, or maybe a little more ?

Edit:
I have a wood backplate that I attached to my Grizzly backplate, and it is as large in diameter as can possibly fit on my lathe and still rotate.
I will go measure that.

Edit 02:
Looks like I can rotate a disk that is about 12.45" diameter, but that is the maximum I can turn on the lathe.
.

 

[GreenTwin]

4th print, which is the last corebox half is complete.
.

 

[GreenTwin]

Some assembly photos.
One end of the pattern halves fit into one end of the coreboxes.

Need to do some smoothing and filling, and then we can think about making flasks and molds.
.

 

[GreenTwin]

For a while, I was studying pour basins, sprues, sprue basins, runners, gates, risers, etc., and was working on a spreadsheet to calculate it all to the 3rd digit.
After watching all sorts of configurations work prefectly, I have reached the conclusion that for hobby work, you really don't need to calculate these things.

For these permanent pattern aluminum castings, I will just approximate what I think will work.
The rules are:
1. Keep the metal velocity as low as possible.
2. Use a low sprue with no pouring basin.
3. Use the gates to control flow of metal into the mold cavity.
4. Use bottom feed to avoid the waterfall effect.
5. Find a place for some sizeable risers.

Since this is a thin part, I thickened the wall up slightly.
Since I am trying to keep the velocity low, but also I am trying to fill the mold cavity as fast as possible, then I will use two large knife gates into the bottom of the bottom flange.

I think this is what I will try.
Hopefully it will for for both iron an aluminum.
Aluminum seems to need a bit more runner/gate space than iron, but many of the iron runners/gates I have seen are set up for iron that comes out of a cupola, and that iron may be significatantly hotter than the iron I can melt, which is probably 2,500 F maximum.

Sprue is in green; I pour directly down the sprue.
Runners are V-shaped horizontal, also in green, shaped like a horseshoe of sorts.
The gates shown in blue are knife gates, and will be a little longer than one side of the flange that is at the bottom of the casting.
The gates enter the bottom of the flange, so that they can be machined away completely when the bottom of the flange is machined.
The mold will be bottom-filled.

Before the metal can go through the gate, it must travel down into the spin trap, which is in orange.
There will be a spin trap at the end of each runner.
The spin trap is a place for the entrained air/sand/slag to go, sort of sweeping out the runner system.
Once the spin trap has filled to the top of the runner, the metal will begin to flow into the mold cavity, while also continuing to fill the spin trap.
Another purpose for the spin trap is to slow the velocity of metal, and let any splashing of the metal occur in the spin trap, and not in the mold cavity.

The mod cavity will fill from two sides simultaneously.
The gate located at the top of the runner helps scrape off any floating slag.
The riser (or maybe two risers) are in red, and they will be tied in near the top of the mold sides.
I was going to terminate these pieces from the riser onto the interior of the casting, but that could make trimming them off difficult.
I guess I will have to tie the riser or risers to the outside of the casting, and then grind that section off flush.
I will stay off the top bead, since it would be more difficult shape to grind the riser piece off and maintain a good bead shape.

This configuration will make the mold split vertical.
I can seal the mold halves using mold cement, and clamp the two halves together.
I will probably make this a 4-piece mold, with a disc at the bottom for the sprue and runners, and a disc at the top for the riser(s).
Some of this is "make it up as you go", but will be based on using modifications of previously successful casting work, so not entirely new ideas.
I guess I could make a 2-piece mold if I use retracts for the runners and riser pieces.
I may have to mock this up somehow, perhaps in 3D.

I could place the mold split horizontal, and that would shorten the spure quite a bit, but would violate one of the big 10 rules of "avoid waterfalling". I don't really like such a tall sprue, since that alone will require quite a bit of metal, but the spin traps will remove the initial surge of churned metal, and the gates will control the veolocity after that.
The problem with waterfalling is that the sides of the waterfall can solidify as the metal runs downwards, and then a cold joint is formed at the molten metal rises back upwards, where the upwards metal meets the downwards metal.
You can get a cold joint that is weak and can crack.
What you want is a smooth laminar flow that progresses from one end of the mold cavity to the other, without turbulence.

Edit:
I guess some would put the sprue down the center of this mold, but that could cause all sorts of problems.
I want to cast this successfully the 1st time, so better to stick with what has worked well in the past.

 

[GreenTwin]

Worked on flask layout.
The logistics of having a vertical parting line are very difficult.
I may have to pour this casting with the parting line horizontal, with gates into the sides of the exterior.
It would be worth a try in the configuration, since it could be tested quickly and easily.
It would take some careful grinding on the round exterior, but I have done that before, so it could be blended in.
.

 

[Jasonb]

Pat why are you worrying about making flasks for the complete hopper, at the moment you only need cast the two halves to get your aluminium masters.

You have the whole inside of the pattern that does not need a decent finish as the core will take care of that so can run any gates etc to that without affecting the external surface.

My earlier suggestion of having the parting line where the bead is would have allowed you to cast the final iron part as per your post #392. Probably a 3 layer flask so the gates can be cut in the bottom layer and then just push a tube in to cut the sprue in the other two flasks, dig down into the top one for the red risers

 

[GreenTwin]

I have this printed sign that hangs on my wall, and it says "Plan Ahead...".
And the text is inside a box, but the box is too small, so the text rather awkwardly protrudes up at an angle at the end, up and outside one side of the box.

Another motto I have is "Don't do anything twice that can be done once".

There is no way I am going to part at the bead line.
I really think that would ruin the contour there, not to mention that is not where the Baker factory parted their pattern.
And the patterns and coreboxes are printed, so the die is set in concrete, so to speak.

I think the hopper could be cast with a vertical parting line, but that gets very complicated very fast.
This is where "ideal" configurations meets "reality" configurations.
I have laid out a new flask a few minutes ago, with the parting line horizontal, and it will work with either the pattern half, or with the complete casting.

For casting the pattern halves, I will use a two-piece mold, horizontally split, with a thin drag that contains the runners and the gates only, and a thicker cope that contains the pattern half.

For casting the entire piece, the runner and gates will still be in the drag, but the drag will be the same height as the cope, and both pattern halves will be used.

The runner in all cases is shaped like a horseshoe, with four gates.
The gates are on the outside of the casting, and will have to be ground off of the full casting.
For the half-pattern, the gate will be in the drag, and so only the mating face will have to have the gate ground off.

I am finalizing my revised flask layout as we speak.
It will work for either one pattern half, or two.

I guess I will try this with the hopper first, and if this method works well, I will use the same method for the gas tank, which is a very similar item to cast.
For the gas tank, I may try two long knife gates on the bottom of the gas tank only, so that I only have to grind off two gates, and that grinding will be on the bottom of the tank.

I considered using only two gates on the hopper, but after seeing how much trouble Barney had casting his full sized hopper, I don't want to risk a partial fill, and so will stick with four gates on the hopper.

Flask footprint will be about 13" square, and the cope will be 4.6" tall.
The drag for the pattern half will be 1" tall, and for the full casting the drag will be 4.6" tall.

This water hopper is a bit of an odd shape, and somewhat on the thin side, so I am erring on the side of caution.
My molds are not reusable, so I have to get it right the first time.

I think this will work out ok, and this is a simple arrangement to mold.
Hopefully there is enough gate area to allow the mold cavity to fill quickly, and eliminate any waterfall problems.
The metal will waterfall into the drag on this layout.
.

 

[GreenTwin]

The flask/spure/runner/gate/spin trap/riser arrangement is finalized.
Now I need to make some flasks.
I usually make custom flasks for pieces, unless perhaps it will be a one-off casting.
The tighter the flask is around the pattern, the less sand that is used.

Back when I started learning foundry work, the big thing was casting your own flasks in aluminum.
I did a few designs for some cast flasks, but the problem with cast flasks is that they are not flexible at all.
A fixed flask size is good for greensand, where the quantity of sand in the flask is somewhat irrelevant, since all the sand will be reused.
For bound sand that may not be reclaimed, the square or rectangular fixed flask size is extremely wasteful.

Here are some cast flask parts I was considering casting in 2012.
I don't use greensand anymore and so I never actually cast these pieces.
.

 

[GreenTwin]

I started making flask hardware, and folks spend a lot of time making flasks and flask hardware; often more time making this stuff than time spent on the actual pattern or casting.

I did not have good luck with flask hardware, due to the time required to make and install it, problems with keeping it aligned, and the fact that the flask that has pins protruding from it cannot be laid down flat.

My flask hardware, along with many other traditional methods used for foundry work, went in the dumpster.
I decided there must be a better way than to make and use flask hardware.

If you are into making flask hardware, you can create a long taper on a pin by chucking it into a drill, and holding it at an angle up against a sanding disk, which is how I made this pin.
.

 

[GreenTwin]

I made a bunch of randomly sized flasks, in anticipation of using greensand, but I have never used any of these for foundry work.
I have used some of them as table supports, and other things, not wanting to waste the wood.
In the beginning I was just blindly copying what everyone else was doing, which was a blunder.
I was in blunder mode for at least a year, perhaps more depending on the magnitude of the blunder.
.

 

[GreenTwin]

I started drilling through the wood cope flask sides and into the drag flask, and using 1/4" pins, in lieu of using flask alignment hardware.
I was using 3/4" thick pine wood for the flasks, and that was a bit thin when trying to drill through it vertically.

I was told that the drilled alignment holes had to be very precisely drilled in a drill press, and I bought into that at first, but then started just hand drilling the alignment holes with a 12" drill bit, and the alignment accuracy was the same.
It is funny how of the "absolute truths" that so many folks have told me have turned out to be so completely false.
All I can guess is that perhaps people have seen foundry work, but not actually done it, or if they have done foundry work, they just followed along with the way things were always done.

I did not know anything at all about foundry work in 2011, and that turned out to be a good thing, and allowed me to experiment to find the best solutions to a given problem.
I learned very quickly to ignore how it has always been done, since there are far better and more modern methods that work well.

I discovered sodium silicate used to make cores in 2012.
I started out using Petrobond (tm) in 2012, but oil-bound sand tends to dry out and lose its greenstrength fairly quickly, and so needs mulling.

I made a few molds using sodium silicate bound sand, but it adheres to patterns in a remarkable way, sort of like superglue.
And I was using CO2 to set the sodium silicate, and that was prone to either over-gassing, or under-gassing, both of which can produce weak spots in the mold.

At some point in 2012 I discovered resin-bound sand, and I use it exclusively now.
Resin-bound sand sets with a hardener, and can be cut/drilled etc. after it sets.

In 2012 I was pouring the oil-bound sand with the wood flask around the mold, and like everyone else, I would inevitably spill molten metal onto the wood flask and burn it.
I discovered snap-flasks in 2012, and so I started using those, and removing the flask from the sand prior to pouring.
I have not damaged a flask since I made this change.

My original snap-flasks were just 1x4's screwed together, and the flask was released by removing screws.
I wanted to find a better way to hinge the flask open, and so when I started using 2x4's, I looked at finger joints.
I stumbled across the finger joint method that I use, where I cut the finger joint on the table saw, and the cut is circular into the end of the 2x4, but when the joint comes together on the inside of the flask, the joint is closed.
This was an accidental discovery, but saves a lot of time, and makes the finger joint cutting much easier.

Two of the finger joints on the flask are glued, and two are not glued.
Pins are inserted at the two unglued joints.
The cope and drag flasks can be easily screwed together, and you can stack any number of flask pieces for a taller flask, in increments of either 1/2", 1", or taller, so a flexible modular way of making custom flasks.

I also ripped and planed a number of standard height pieces of some nice 2x4's (not pine), so I can start with an accurate shape.
Luckily my daddo blade arrangment is still set up on the table saw, and so I can go out and start making flasks without doing a lot of setup work.
And I still have a large amound of cut and planed flask stock, so that material is also ready for use.

.

.

 

[GreenTwin]

You can see in this photo that I make notes on the flask with a sharpie, often to include sand weight, resin/catalyst/hardener weights, top/bottom designation, and different marks to distinguish pin arrangement, since I started using all 1/4" diameter pins.
Looks like I also have flask inside dimensions marked, as well as the flask height.
I also have started drawing a band down the side of all the flask parts, to provide immediate verification of correct 180 degree alignment when the flask pieces are assembled.

Since the flask is removed prior to pouring, I drill 1/4" holes in the sand and insert wood dowels prior to removing the flask.
The two mold halves are then cemented together, and the wood dowels are removed.

The steel ring on top the mold is a catch bowl to prevent molten metal from running across the top of the mold once the sprue is completely filled.

This is the flask arrangement that I use, and it works very well.
It is easy to make a flask of any size, easy to screw the flasks together, and easy to modify a flask if necessary.
These flasks remain reusable indefinitely since they never get exposed to molten metal.

A flask can be reused for another smaller pattern by temporarily boxing out the extra space.
So one flask can be used for multiple sized patterns.
If the flask is not deep enough, another section can be stacked on top.

This is a flexible and versatile flask system, and very quick to make since there is no flask hardware.
And with no flask alignment hardware, and no side handles, you can set any flask on any edge, which is a real plus when you are handling flasks/molds.
.