Ball Hopper Monitor - Casting Project

[GreenTwin]

Comparing old catalog cuts, photos, and finalizing the dimensions of all the parts is a round-robin affair for sure.
The more photos you look at, the more features you spot; seems like an endless process to get it accurate.
I have spotted several features that I don't see in any casting kit.

There must be a 1/2 a bazillion measurements in that engine; perhaps a full bazillion.

Round and round we go in a 2D section, until it all fits correctly.
Everything has to be rounded to a somewhat normal decimal number (no metric on this engine; sorry metric folks.......long live Imperial).

Every change affects the adjacent part, and often other parts too.
Tedious, but necessary to get to the point where accurate 3D models can be started.

.

 

[Jasonb]

I'm not sure why everything has to be rounded to a "Normal" decimal number. With the scale factor you are using I would only adjust things to suit stock materials and cutting tools.

For instance, your ratio was about 0.42:1 so a 1" shaft on the original I would adjust probably to 0.438" (7/16") same with a matching hole or slot. But if it were a cast feature of 1" then I would leave that as 0.420". The cast feature does not need maching and the printer does not care what it is working to so may as well keep it to what the item scales.

As for working out sizes I've just been sent 34 images many with rule or callipers showing sizes of a single part to draw up for a pattern and I could do with a lot more as I don't really want to guess at some. This is a valve block for a full size engine, long lost or damaged. This also really wants working back into imperial as that is what the engine would have been built to but the owner is in mainland Europe so took metric sizes.

 

[GreenTwin]

Yes, for shaft sizes, I go with a standard ground rod dimension.

Cast features are to just round off from four decimal places to two or three sometimes.

That looks like a tricky piece to model and cast.

.

 

[GreenTwin]

I am still thrashing out dimensions in 2D for the Ball Hopper Monitor.

There are a surprising number of details that have to be included in just one piece, such as the gas tank.

The more you look at the photos, the more details and features you notice.

It all has to be worked out before any serious effort begins in 3D modeling, else things will not fit, and the engine will not run.

Standard fuel line fitting sizes have to be selected at some reduced size.

Making progress, but it is slow going for sure to create an accurate scale model.

Pull directions (for molding) have to be determined for every part, in order to get the draft angle correct.

The fuel tank pulls from the sides, with a core that must be supported from either end.

Water hopper pulls from the top/bottom.
The window on the side of the hopper may require a retract, or it may pull cleanly downward (I think it will pull downward with the correct draft angle on the window).
Large core on the hopper supported from the top and bottom.

Frame/cylinder pulls from the sides, facing the ends of the crankshaft, with some tricky cores for the water passages.

Muffler is two-piece, and so not a problem to mold those pieces.

I will cast the piston in gray iron, and will use the 3D printed piston halves as a core box.

The water hopper cover could pull from the top/bottom, but then there is that tab on the top with a hole in it that really needs a side pull, so perhaps a 3-piece mold.

The flywheel will be easy enough, with side pulls.

I am contemplating casting a base below the engine, in lieu of wood runners, to give the engine a more rigid base to sit on.
The base would be similar to what a typical Frisco Standard engine sits on.

.

 

[Jasonb]

The the 3D modelling was not too much trouble, spent more time scaling sizes off the photos that actually drawing it. Just need to add the machining allowances to the two main flanges and the valve guides but wanted to show the guy what the final item will look like first. Will be split horizontally to form the two pattern halves which is why the rectangular flange looks bent a sit has draft on the sides and face.

 

[GreenTwin]

That is a great looking 3D model/pattern.

I got the Ball Hopper Monitor head and hopper worked out in 2D CAD.
I am not at my 3D computer, so can't put it in a 3D model yet.

The hopper is very tricky with the sloping window with draft, and head inside of the hopper.

I think I am finished with the fuel tank design too (2D CAD).

Piston, rings, and wrist pin are designed in 2D.

I have started working on the top of the frame/cylinder/water passages, and figured out what the brackets for the 4hp gas tank support look like, and where they are located.

The crankshaft is designed in 2D.

When I finish the frame/cylinder design, I think the worst will be over, and I can focus on the fidgity stuff.

.

 

[Jasonb]

Boolean subtract should take care of the cut out in the hopper side. make the part you subtract a extruded profile of the cut out with the end sliced off at your 2 or 3 degree draft angle and radius the corners.

Order would be something like

Revolve profile to create the hollow "ball"
Boolean addition to add the inner wall of the cutout
Boolean subtract to form the external cut out
Add any bosses around the bottom

 

[GreenTwin]

The Ball Hopper Monitor is a very tricky engine to model, mainly because there are so few photos of disassembled engines.

If it were not for Joe Prindle, I would guess there would be no photos of disassembled BHM engines on the net (thanks Joe).

I can count on one hand the number of partially disassembled BHM engines I have seen.

The flange on the bottom of the water hopper is not symmetrical.

Little details keep showing up as I study the photos.

I would like to get this design as accurate as possible.

I have found someone who has a 4hp BHM, and he is willing to take measurements and send photos, and so I think he and Joe Prindle will save the day as far as getting an accurate design.

I am modeling the 4hp BHM, and there seems to be fewer photos of that model available.
There are quite a few 2hp BHM photos, but still almost no photos of disassembled engines.

There are quite a few photos of various casting kits for BHM's, but those are accurate to a degree, and then they become an approximation of the original designs.
I don't want an approximate design; I would like an exact design, or as exact as I can get it.
With today's modern design tools, I don't see any reason to approximate anything on a BHM model engine.

.

 

[GreenTwin]

First attempt at the hopper was a fail.

I tried a simple rotate section, then a window cut, and then adding the sloped back of the window.

No dice.

2nd attempt worked better, but no draft angle on the window, or no draft angle on the window that I liked.
The draft angle was linear, and I really need non-linear.

It is hard to get the back of the window to slope, and then make anything else work, but the back of the window slope is cricitical, and so that is how I did it.

I will fill the sides of the window, inside and out, to get a non-linear draft angle.
And a bit of filling on the bosses inside to allow a pull.

I think this model will work ok at a pattern.

.

 

[Jasonb]

Did you do it using Booleans as I suggested?

 

[GreenTwin]

No sorry, you lost me at "boolean".

I have no clue about how to do that.

.

 

[Jasonb]

With today's modern design tools, I don't see any reason to approximate anything on a BHM model engine.

I have no clue about how to do that.

Well the tech is OK so long as you know how to use it otherwise you may have to approximate.

Actually there was one additional stage to what I suggested , a need to remove the excess Boolean addition beyond the profile of the hopper.

Here it is with draft so it can be pulled from the sand, fillets internal and external and a constant wall thickness. I just used a simple shape in the example

 

[GreenTwin]

Well the tech is OK so long as you know how to use it otherwise you may have to approximate.

LOL, Jason, you have left me in the dirt on the 3D modeling thing.

I got no clue how to do that.

Some impressive sounding works, but I have no idea where to start.

Can you put it in "moron" language ? so I can follow it ?

.

 

[GreenTwin]

Gas tank is done.

.

 

[Jasonb]

Could do with some fillets on the internal corners where the various bosses meet th emain body, just enough to stop the sand crumbling at the corners. Old boys would have used fillet wax but it can be added at the printing stage.

Also do you need some draft on the longer lower fuel outlet

And the lugs to mount it to the engine.

 

[GreenTwin]

I forgot the mounting bosses.
Its tricky finding photos of those things.

Have not added fillets yet.

The molds pull front-back, so the fuel outlet on the bottom should pull, since it is directly inline with the pulls.
Probably needs some fillet work around it.

.

 

[GreenTwin]

The addition/substraction is a powerful feature.

I recall learning how to do that with mold making, such as molds and cores for the Galloway, but this was more for looking at what the molds would look like, not for modifying a part.

I always thought of the add/subtract in terms of molds and cores, but not actual parts.

The add/subtract feature would have been handy to use many times in the past, but have always worked around that.

.

 

[GreenTwin]

Molds and cores for the Galloway, such as this head.

Or this piston.

.

 

[Jasonb]

Good luck with pulling the parallel sided fuel outlet.

Even when casting simple shapes like a round disc or square box it is usual to have some draft so it pulls cleanly rather than vertical faces.

 

[GreenTwin]

The top filler boss does not need vertical taper, since it pulls sideways, but it does need a little fillet.

The exit out the bottom of the tank boss could use a little taper in the pull direction.

I think everything else will pull ok.

.