Anvil Fever

[GreenTwin]

I attended a local craft festival a few days ago, and ran across one of my friends from the local Ornamental Metal Museum.
We had a great chat, and got caught up with what he was doing in the art-iron world.

I noticed that they had a number of paperweights on their flyers, and one of the paperweights was a small anvil.

I was able to make a donation, and am now the proud owner of a Metal Museum mini-anvil.

So now I have "anvil-fever", and I really want to cast some of these in gray iron, and so I will either model one in a 3D program, or perhaps just freehand a pattern in wood.

I really like the look and feel of these tiny anvils, and it makes for a great conversation piece.

There are some odd shapes, and draft angles in an anvil, and so I will have to figure out where to start the 3D model.

These would make a good calling card/advertisement too, if you could figure out how to make those tiny letters.

This anvil was cast in 356 alloy aluminum, most likely using Petrobond (oil based) foundry sand, which gives a superb finish if poured at about 1,350 F.

 

[Jasonb]

Would ductile iron not be a better choice just in case someone tried to use one?

Letters should just be a simple CAD item and then printed. comes down to the quality of your printer as to how crisp they will come out

Looks like the one in your photo had a simple central split line and the draft has been filed filed flat

 

[GreenTwin]

After a few hours working on this tricky little item, I have come up with a pattern in 3D modeling.

I may have to work on some draft on the sides, since there is none right now.

About 5.6" long overall.

There are a lot of odd curves in an anvil.

I sectioned it in order to print a pattern half.

The letters have 30 degrees of draft on them, and luckily they drafted ok (sometimes letters and logos won't draft).

Ductile iron would be an excellent choice for an anvil, and if I had some nickel-mag, I would cast it in ductile iron.
I have looked all over the US, and cannot find any nickel mag, or can't find anyone who has it that is willing to share a bit of it.

This could be quenched in water right after it solidified, and it would basically be the hardness of tool steel.
That would be the best I could do as far as hardness.

 

[GreenTwin]

Here is the whole model, without fillets.

I took out the fillet on the top.

 

[GreenTwin]

I finally wiggled some draft angle into some of the surfaces.

This was causing convulsions (for me and the program), and it had to rethink it several times.

This is what I came up with for draft angle.

The curves surfaces on either end do not respond to normal draft angles, at least not with the approach I used for this model.

I ended up adding a flange in the center of the model, and will manually fill up to the flange will bondo or something similar, after I 3D print the pattern.

I will have to 3D print two pattern haves, which will be mirror images.

I think there are some 3D features where you can pull/stretch surfaces, but I am not sure how to use those.

The draft angle surfaces are not merged into the model, and they show up as light blue.
The draft angle material can be toggled off in the view, to see the anvil in the "as-machined" state.

 

[GreenTwin]

I would probably mirror this layout in the flask, and I guess use six on either side, with a runner down the center, with the sprue in the center of the runner.

This would be a good application for using a matchplate.

 

[william_b_noble]

why not use the one you are holding in your hand for a pattern, use a silicone mold agent and make a mold of it, then cast a plastic copy whcih you can modify with a different logo if you want, then repeat making a master mold of the modified copy. you can make several, put the spru in place, and then make the sand mold for pouring the metal.

 

[GreenTwin]

why not use the one you are holding in your hand for a pattern, use a silicone mold agent and make a mold of it, then cast a plastic copy whcih you can modify with a different logo if you want, then repeat making a master mold of the modified copy. you can make several, put the spru in place, and then make the sand mold for pouring the metal.

One of the ulterior motives I have in making 3D models for things is that with every model made (they are all a bit different), I run into situations that I have not seen before, and so I am forced to come up with new solutions to get the model done.

With every 3D model made, I add a few more tools to the toolbox, and I often find that some of those tools become very useful later one with other engine designs.

This anvil was no exception as far as challenges.

I started with a random approach, since I have never modeled an anvil, and that worked pretty well until I started trying to add draft angle to some of the pieces.
The draft angle did not go well at all, but I finally was able to get most of the surfaces angled, and can add a bit of filler for the sides to get those angled.

An object with so many curves does not act nice whey you start trying to add draft angle to things after the model is done.

It would be really nice to have the option to only add draft angle to only one side of a shape, but Solidworks does not have that feature.

There is a "deform" feature, where you can stretch things a bit, but I can't control it well enough to use it for draft angle.

So its like learning how to drive a race car.
One can't become a good driver by looking at the car as it sits in the driveway; you have to actually get out and drive, and of course sometimes crashs into things (in Solidworks only).

.

 

[william_b_noble]

great answer - the key is the objective, which is only incedentially to produce little anvils and is actually to learn a skill. The same reason I'm trying to program an FPGA to do something - I have no need for an FPGA to do anything importaint (now) but I want to know why it's so hard.... (and I'm finding that out)

 

[GreenTwin]

One side printed.

I had a bit of bed lifting on one side, which caused some rough texture on one side of the front.
I have not resolved all of my bed lifting issues, but am working on that.

First half looks ok all things considered.

 

[GreenTwin]

I should have included a couple of 1/4" holes on the inside surfaces for alignment pins.

I can add some pins by drilling through both pattern halves when they are taped together, using a 1/16" drill bit.

Hindsight is always 20/20.

I have considered filling these patterns, which will be necessary I think on the sides (and thus the flange I used), or perhaps casting them as is for all but the side surfaces, making two pattern halves in aluminum, and then buffing them out with a ceramic sponge.

Sometimes the crevasses are pretty deep on a 3D print.

.

 

[100model]

I look forward to you casting the anvil in your finest gray iron.

 

[GreenTwin]

I look forward to you casting the anvil in your finest gray iron.

Thanks ironman.
This is one of those projects that is not too involved, and so I can work it in during the off hours.

This would also be an ideal project to try out the mini-iron furnace.
I would like to be able to do small iron melts with the mini-furnace so I don't have to drag out the big one every time I want to do a small casting project.

https://www.homemodelenginemachinist.com/threads/building-a-mini-iron-melting-furnace.34301/
I will probably cast some permanent aluminum pattern halves first, and then cast a few anvils in iron.

Here is the second 3D print pattern half.
I turned off the air conditioning, and made sure there was no air flow around the 3D printer, and this print did not experience any lifting problems.
I am beginning to think that 3D printers are very sensitive to air flow/air drafts around them, which causes the print to shrink and lift a corner off the bed.

 

[GreenTwin]

Here are the two pattern halves together.

I think this turned out ok.

 

[Jasonb]

What happened to the writing? You might be better off having the raised panel that the writing is on as a recess that way you can print out different loose pieces to fit into the recess that can have whatever you like printed on them so they don't need to be the same or a whole new pattern produced if you want different text.

Should you have included core prints for the small core to form the square hole?

Regarding the draft angle, if you just started out with a "cube" and drew the profile on the mid plane can solid works not do cut-extrude with a draft angle from the mid plane? Also as yoy say you will machine the top flat would a bit of machining allowance be worth adding as well as the draft angle.

 

[GreenTwin]

I decided to cast a generic "put your logo here" anvil.
Bascially any logo/lettering could be 3D printed, and adhered to the side of the pattern.

I considered a recessed panel, but the top edge of the anvil may chip off in use, if this were a real anvil.
Since this is not a real anvil, I guess a recessed panel may look ok, but would be more difficult to machine off flat.
With lettering, I guess you would not machine the panel off anyway.
Maybe a recessed panel may look pretty good.

If stick-on letters were used, you could machine or sand the protruding faces smooth, to get rid of the 3D print lines, before the letters were adhered.
For recessed letters, you may have to pour epoxy or paint around the letters to fill the 3D lines.

The square hole is still in the pattern, but covered up by the draft angle I added.
I could extend that out a bit and make a core print top and bottom.
For the round hole, it is not on the centerline, so that could be trickier to cast in place.
A core for the round hole would have to extend out the side of the mold, and the sand would have to be carefully packed under and around it, and the graphite rod withdrawn out the side of the mold, the pattern removed, and then the rod reinserted into the mold (this could be done with bound sand).

For the square hole, I am going to file a piece of graphite rod square, to use as a core, and it can be a short core since it is on the pattern centerline.

I don't have a broach that small, else I could drill a round hole and broach it.

I think the draft that was added on the top is enough to also be used for a machining allowance.

I did forget to add draft angle to the bottom, and so I may have to reprint these pattern halves.

The problem with this pattern is that wonky horn sticking out the front.
The horn does not play nice whey you are trying to get it to any sort of correct shape while including draft.

I guess could trim off everything but the horn, and then try an extrusion with draft from a shape on the centerline.

Sometimes the only way to find out how to model an object is to first find out how not to model an object.
The best modeling solution is seldom obvious when modeling a new complex shape.

.

 

[GreenTwin]

I tried JasonB's suggestion, and traced an outline at the center plane.
I first chopped off everything except the horn.
Then I extruded outwards with draft.
This worked pretty well, but........if you use the original size of the object, as you extrude it outwards, you lose size.

You can always draw the original shape somewhat larger, but this is somewhat of a hit-and-miss affair.

If you created a plane out from the anvil, at the outside face of the foot, then you could extrude the originally sized shape inwards, with draft angle, and the shape would become larger, which would be desirable.

The horn still has to be dealt with, but Jason's idea is a good trick to remember, so into the toolbox it goes.

With everything I model and cast, often the lessons learned become far more valuable than the object/casting itself.

The only way to learn 3D modeling or metal casting is to jump in there and try things.

Edit:
It is worth mentioning I guess that when modeling engines, I model the engine in an "as-machined" state, and then later add machining allowances and draft angle.
Sometimes I include draft angle in the original model if it does not affect the as-machined size.
The reason for this approach is that when you create 2D drawings for an engine, you want to toggle off all of the features, such as what was added for machining allowances and sometimes what was added for draft angle, so that your 2D drawings reflect an as-machined state of the model.

If you toggle the machining allowance and draft back on, then you can 3D print a pattern.

So if you use Jason's method suggested for the anvil, there would be no way to derive 2D drawings of the as-machined state, at least no way I can think of. I do think Jason's approach is a good idea for an anvil, since the machining on its surfaces is not critical as far as dimensions.

.

 

[Jasonb]

Pat in the same way you toggle on/off your draft angles that were added at the end you simply sketch the material to be machined off the casting and extrude cut that and then you can toggle the cuts on/off so the one 3D modeled part will show the pattern or the finish machined part simply by where you leave the bar down the left hand side, 2D drawing will be produced from where that bar is down the left side so I can easily produce a drawing of teh pattern and a drawing of the finished part

Here you can see I have sketched three rectangles on the central plane that represent the machining of the top, bottom and bearing caps and cut them through all material

And in practice the pattern that was produced from a Step file exported from the point that the first image above was taken along with the corebox

This is the casting being machined to a drawing taken at the point down the list as shown in the second image above

I'll explain the recess/loose piece method a bit later but you don't end up with a recess on the cast anvil.

 

[Jasonb]

Round hole can simply be drilled and just core the square one.

If you intend to make a match plate with 10 or 12 pattens on it then having to remove existing lettering, make good and then add new lettering would be time consuming. Even worse if someone comes back and askes for a second batch as you would have to redo it all again. Same would apply to just casting one at a time, it will be easier to just change the insert plate and you can keep them if more of the same text are ever needed

So if you print a recess say 1/16" deep into the side of the pattern you can then simply place a flat plate with the new lettering on into the recess. If that plate were say 1/8" think it would have the raised surround and the letters standing out from that. You could either print it complete with letters or simply use 1/8" thick plastic or aluminium and glue separate letters in place. Letters could be 3D printed, pattern makers letters or smaller "Slater's letters"

Here is a quick anvil with exaggerated draft

The pattern would end up looking like this so no risk of a narrow edge chipping out as there is not actual recess once cast.

An alternative would be to have the recess say 3/16" deep so that if you dropped in the same loose insert it would have a recessed surround, can't see a problem if the recess is far enough down from the top edge. It would also give the option to fill the recess with paint on the finished item to make the writing stand out.

 

[Shopgeezer]

I had a bit of bed lifting on one side, which caused some rough texture on one side of the front.
I have not resolved all of my bed lifting issues, but am working on that.

I had significant bed lifting problems when I used a flexible magnetic bed. I had a banana shaped base on all my prints. Going to a solid base helped but I suspected that my drafty shope was not helping either. I purchased a commercial enclosure for my printer and no longer have base lifting problems. With the addition of a strip of led lights the enclosure makes a nice printing station.