This isn't going well...

Hi, everyone. Having trouble removing a 3d-printed master from the mold, as you'll see in the pictures. I've tried two different kinds of sand: one is Teton-Bond (190, ultra-fine), the other is an unknown, coarser sand that came with my flask. Both break cleanly when tested and bond reasonably well around the outside. The inner openings are consistently falling apart though. I've tried both sands several times each, rammed to different densities, with and without baby powder, primed and un-primed masters (the primed ones are sanded smooth on all sides), different draft angles, and a few other things. I haven't tried paste wax yet because I'm on a very tight budget. What can I do to preserve the details? Are there any cheap mold releases or methods that might work?

The grey one on the left is a failed print just there to show details. The black one has a much better surface finish.

 

Scale of this work? Looks like about 6mm holes in your flask?

 

It's about 1" long. I can tell you exactly when I get back home. It's been a few days since making the master and my memory isn't what it used to be.

 

It's 50.7mm overall, including the sprue. The top 3 holes are 5.35 x 3.7mm. I managed to get one good mold when I first started about 2 years ago, then life got in the way for a while. I'm going to try adding more draft angle, less depth in the grooves, and maybe a coat of paraffin if I can find a candle around here.

 

What is working for me with Petrobond is a painted 3D printed pattern with generous fillets and then a coat of #2 mold release wax wiped/buffed off then graphite then parting powder. There is enough paint on there to remove most of the 'washboard' effect from the filament layers.

https://forums.thehomefoundry.org/index.php?threads/pattern-rapping-fail-sailboat-part.2268/

 

Thanks. I thought fillets were just for structural purposes, but it's easy to see how they would extract better. I can always file the inside corners sharp later. The layer lines were all sanded smooth and filled with primer but those corners could be hiding some texture. I'll probably start another iteration today.

 

2 years? Is your sand that old as well?
My Petrobond is a lot older since last use. It doesn't age well. The sand I have would be useless for casting without some fresh R-OH and mix/mull.

 

The sand is in surprisingly good shape, especially since the Teton-Bond was really dry when I first got it. Almost sent it back until the supplier told me how to restore it. Both kinds have been sealed in bags the whole time. It's holding shape very well everywhere except the small holes. The coarse mystery stuff sticks to itself immediately after sifting and smells kind of like baking cookies. The ultra-fine breaks much more easily but has a smoother surface finish. I've heard the finer grains don't stick as well because of their smaller surface area. Feel free to correct any misconceptions though; that's why I'm here. Youtube isn't always accurate.

 

Understatement of the week!

Very little practical experience with bonded sand (not all that much with green) so I'm not going to be much help past this point.

 

Just looking at the pattern intricacy and thinking about its tiny scale, it's hard to imagine that it can be done at all in any kind of sand casting, as is. The sharp edges of the micro channels in places just seem destined to break off. I would have thought lost wax would be the best bet for something like this. In the few oil bonded sand jewelry casting videos I've seen, the pattern is more of a gentle bas relief, without penetrations and hard edges.

If it's do-able at all, then the problem on a micro scale is similar to what happens on a larger one when a greensand core breaks off. The adhesion to the sides of the recesses is greater than the adhesion to the base medium. Therefore the base breaks off. To have any chance at all, the pattern will have to be well drafted, very smooth, and inside corners radiused generously. It would benefit from being rapped somehow to reduce the side adhesion before removing the pattern. On something this small and delicate it's hard to imagine how that could be effectively done -- but maybe you have a way. Vibration might work better than conventional "rapping".

I don't know what commercial release agents or special parting compounds can be used with oil bonded sands, but baby powder may not be ideal as parting on this casting. In fact there are two kinds of baby powder, one talc based, and more recently, one cornstarch based. The talc based type is the ONLY type that would be suitable as parting in casting. Make sure your problem isn't exacerbated by the the wrong type. But also there may be better release and parting compounds commercially available for this problem casting.

Since you want the base of your casting recesses to have higher adhesion than the sides, it may be a benefit to apply parting to the surface you impress the pattern into, and the pattern itself before impressing. And it's important to get an even dust coating, not clumpy baby powder sprinkles. I hope you're dusting out of a bag, and the powder is fine and segregating.

I'm not saying any of the above will work, just that if there is any chance at all, they will be important to getting there.

 

If you really want to cast this piece it should be possible. I just cast a pattern with common letters 3/16" high in green sand and they came out clearly and cleanly. But, the pattern was carefully prepared.

First, put on magnifying glasses like Optivisors and make a few sandpaper files by gluing 120, 220, 600 paper to a suitable backing like a popsicle stick or similar and likely smaller. Be sure you can get into all the nooks and crannies of the pattern and then sand it, apply some lacquer, and sand it again. Use a light tangential to the surface you are working on at the time. Make sure all undercuts are eliminated and that all corners have draft and are radiused.. This may take several times going over the pattern and will require some patience. But once the pattern is actually well prepared it should draw.

I do think this could be done in petrobond and may be helped a bit by placing sewing pins in strategic areas of the sand to reinforce it. I'd insert them so the heads were just flush with the sand and likely place them in corners.

Prep the pattern carefully with ordinary paste wax. I like to use either Johnson's paste wax or Minwax. Many should work. Then use some graphite and brush it and work it with your fingers into the wax.

Then put it onto the table and put your flask over it and riddle on the sand and ram it up in a conventional manner after using talc, graphite, baby powder, or probably even ordnary flour as a parting compound. Petrobond is one of the most forgiving sands out there. Flip it over, place pins if you decide to do so. Now ram up the drag after applying a good dusting of parting. I like to apply talc so it is bordering opaque.

Next insert whatever you plan to use as drawspikes (I recently saw a video posted by Meteor where the guy just tapped shapened needles into tiny holes in the also tiny pattern. He was making type more intricate than your pattern by a long shot.) into the holes made for the purpose. Then lightly rap it mostly up and down and very gently side to side. Should work.

Bound sand using something very cheap and available like linseed oil would provide a mold with more strength. I like to use sodium silicate and CO2, but I understand it is crucial for you to use cheap and readily availble materials. A well-made bound mold will reproduce any fine detail you can pattern provided the pattern has draft and the sides are smooth.

If you find an area or two of the pattern to be breaking out sand, you can be virtually certain there is a pock mark or undercut or some sort of roughness that needs to be taken care of. Fix it and try again.

Recently Tops had a pattern that drew poorly. He sat down with some Bondo, sandpaper, and lacquer and, lo and behold, it drew nicely and he made a great part.

Good luck.

Denis

 

Thanks everyone, this is exactly the kind of help I was looking for! I'm currently redesigning the piece with more draft angle, corner radius, and maybe slightly less surface area. Also recalibrating the printer for a better start. Some sources have said to use a slightly matte finish to hold parting powder, others say a slick surface works better. I wonder if using a solid wax master would make a difference? I plan to mostly focus on small detailed jewelry pieces, so the extra steps might be worthwhile if it would preserve detail and reduce finishing time.

@SRHacksaw: I'll eventually get into lost wax casting, just working with what I have on hand for now. Always have to do it the hard way first... As for powder, I'll have to make a bag. Should have some muslin around here somewhere. Shaking the piece in a ziploc bag is probably not the best way to get an even coat.

@Melterskelter: I really do want to cast this piece because it has sentimental value. The corners are probably the biggest issue here. Drawspikes are something I never thought about. Have to look into it.

 

Finish to hold the powder: I'd put in a strong vote for slick. Most of the times I have had trouble with poor draws is because of rough surface, undercuts, pock marks. You do not need to "hold" powder. A matt surface will definitely "help" hold the sand onto the pattern.

Bag: No need to go to much trouble right now. I used a sock for a long time and it worked just fine. A muslin bag as suggested to me by HT1 was an improvement. But you've got enough to deal with.

Drawspikes: Just put a couple 1/16" holes in it and grind down a couple 3" box nails at a shallow taper to insert in the holes. The heads will give enough to tap up on using a `1/8" by 8" bar for both rapping and up-tapping for extraction. Up-tapping definitely helps as the sand's inertia tends to cause it to fall off with a jarring tap, but it might stick if gently lifted. Remember, go easy on the tapping.

I said above I like to use powder heavily at the parting line. But, I usually brush/blow it off the pattern. Heavy parting will cause an ugly finish.

Denis

 

um m you need to rub that down with some graphite on 000 steel wool.

Denis is all over my advice, but get some graphite powder and make hours of love to that part with 000 steel wool

I like to use auto filler primer, steel wool and graphite to prep patterns, and yes I occasionally work with printed parts, as small as that is it may almost be a NOGO, but just from the OP picture you used no parting, on this you NEED Graphite, it makes a part slippery and will overcome NO draft situations, not all the time, but enough for a one off part

spray with auto body filler primer, let dry, sand with steel wool courser to finest 000, when it appears smooth and properly drafted final coat, steel wool with graphite powder, sprinkle with graphite powder, THERE can be no lumps, they will transfer to the casting, so a good partin bag is helpful, you can get around this, by sprinkling from very high and letting the parting compound (graphite) settle , so kill your fans , and work indoors , no breezes

Good luck

V/r HT1

 

I thought it might be a good idea to talk about the problems lateral rapping will cause with your pattern. It is common to see people rapping patterns quite vigorously in all directions prior to drawing them from the mold. And, for simple patterns that are block-like or have very large shallow windows, that can be ok. But, your pattern has narrow deep "windows" in it and the sand contained in them cannot compress practically at all with lateral rapping. It will just be ripped from its vbase and move WITH the pattern as it is displaced side-to-side. If, on the other hand, you rap vertically and somewhat delicately, the sand will tend to shear away from the vertical walls and will release its grip on those walls. I do not think I would rap horizontally at all.

It took me a long time to figure out why certain patterns with letters on them were drawing with letters recognizable but "dirty." It was because of lateral rapping. Once I just rapped vertically, they pulled cleanly.

Here are some much smaller letters I did today.

You can see that even the windows in the "A" and "R" are preserved. This is 100 mesh green sand, BTW.

And later while drawing blind risers out of the mold on this pattern, I was struck by the fact that the grain in the wood of the riser was rendered in the sand. This is not due to any special skill on my part. It just shows that sand can register fine detail relatively easily. That riser is just over 1" in diameter and about the same in height.



Denis

 

R= generic hydrocarbon......chem shorthand

OH = alcohol.

 

Tom Cobett has provided a lot of information on petrobond and shop-formulated oil-bound sand—-K-bond. I have followed his recommendation to use Glycerol (glycerin) as the rejuvenating and activating alcohol in oil-bound sand. He does say any alcohol can be used. But, being a large molecule compared to, say, ethanol, means it will evaporate more slowly than most commonly available alcohols. It is cheap and widely available.

Denis

 

The re-design phases is taking longer than expected due to learning new modeling skills. Not sure how to move some elements properly in Blender so I might have to go back to an earlier model and change the order of some steps. Could be ready to print sometime tomorrow if all goes well. Meanwhile, can someone define 'rapping' in this context? I don't know much of the vocabulary here.

 

That's tapping on the pattern to free it from the sand.
My favorite is wood. It has a bounce to it that steel doesn't have.