I have noticed that myself whilst browsing through some projects on thingiverse, in almost all cases I have ended up designing my own parts in openscad.

But I also noticed that the direction of your print always depends on which way you need the part to have the most strength, i.e. if you are going to print part of a "finger" it should be printed laying down so the strength of the layers will work in combination with each other, if you print it standing up, then that part will be weak if any pressure is applied to either end, it will just snap off at one of the layers.

and a quick dip in acetone bonds each layer together to form a much stronger bond.

I am not really sure on how much pressure/force you think these parts will be subjected to, if it is wear and tear you are worried about, adding rubber to the parts such as the finger tips, will reduce the amount of abrasion on them (use that rubber dip used for tool handles), for moving parts, I would reinforce each union with metal bearings, or bushings to reduce wear, if you are worried about longer/wider parts that are thin, these can be reinforced with steel or carbon fiber rods.

 

Getting a standardized material strength is so difficult because print conditions are so hard to duplicate between machines and even repeating on a single machine. We can make three consecutive parts and get three different results. One time the bridges are perfect and the next time we get sag or string balls. This is unusual but, in the real plastics world, you make 100 tensile bars and expect to get the same strength out of each.

Also, flow line directions are important in injection molding and some people take advantage of the anisotropic effects from the flow direction. But flow (print) direction is critical in 3d filament printing because strength along a printed filament can be 10s or 100s of times different than normal to the printed filament. Also, fill parameters like percentage, fill pattern, staggered full pattern repeat size of the fill pattern will all impact strength. Finally, whether or not the extruded material is "set" onto the surface as a cylinder, or "pressed" onto the surface as a ribbon. The set vs press is extremely important as fill ratios grow above 70%. It all really depends how close your print head is to the bed on the first layer and wether you overfill each layer by making ribbons when the software is not expecting it

Im aware of all or most of what you say; its almost a crap shoot, or it would seem so. It seems almost (in the 3d printing community in general) as if "results may vary," so lets not even bother quantifying our results and isolating what variables give us the best results. Almost all of the guidance i can find is related to print settings and provides no evidence of its merit. Getting past this "results may vary, so let's not pursue better ones" mentality is what I'm trying to get past with this test procedure.

 

I have noticed that myself whilst browsing through some projects on thingiverse, in almost all cases I have ended up designing my own parts in openscad.

But I also noticed that the direction of your print always depends on which way you need the part to have the most strength, i.e. if you are going to print part of a "finger" it should be printed laying down so the strength of the layers will work in combination with each other, if you print it standing up, then that part will be weak if any pressure is applied to either end, it will just snap off at one of the layers.

and a quick dip in acetone bonds each layer together to form a much stronger bond.

I am not really sure on how much pressure/force you think these parts will be subjected to, if it is wear and tear you are worried about, adding rubber to the parts such as the finger tips, will reduce the amount of abrasion on them (use that rubber dip used for tool handles), for moving parts, I would reinforce each union with metal bearings, or bushings to reduce wear, if you are worried about longer/wider parts that are thin, these can be reinforced with steel or carbon fiber rods.

The test that I designed tests the strength of objects printed in all 3 planes, to cover the bases you were talking about; some parts need to be printed in certain directions to maximize their strength in the plane that forces will be subjected on them. But you cannot design this way 100% of the time. Consider the simplest example, a cube with a tunnel in each plane. Say, its a corner bracket for a cube made of steel rods. It will be very strong in one dimension, less strong in the other, and utterly useless in the third. Some of the parts I need for this hand need strength in more than one plane. Ideally I need layer adhesion that borders on injection molded strength. Ill take the best I can get, but I need to know what actually IS the best I can get.

How strong do I need the parts? Strong enough to be pounded by a hammer, slammed in a car door, caught in a crevice and unknowingly yanked backward against their joints, etc.

 

You might be best off printing molds then filling with fiberglass mat then casting fiberglass (polyester or epoxy) resin into them.

I would prototype with printed parts and save the casting for the final part.

 

How strong do I need the parts? Strong enough to be pounded by a hammer, slammed in a car door, caught in a crevice and unknowingly yanked backward against their joints, etc.

Don't take it the wrong way but, it seems to be asking a lot from a 3D printed part, it sounds as if you are looking for the strength of titanium in 3D printed plastic. Unfortunately, everything will have its limits, even a machined aluminum part could get damaged when slammed in a car door, or smacked with a hammer. In my opinion, I would get the form and function taken cared of first and worry about longevity afterwards, once you get a working prototype then you can worry about improving it to make it more durable, but all these things you are talking about overcoming, are not things that would happen to people everyday, accidents do happen I will give you that, but no matter what you come up with for a material that is lightweight enough to use, will not overcome all these limitations, you will not find something that can be printed on a 3D printer that will be able to withstand heavy impacts, twists, bends, and be heat proof too (some may be able to overcome some of these limitations, but not all).... just my 2 cents.

With all these tests you performed, were any of the ABS parts treated with acetone first? The parts become much stronger than a part that is not treated, the acetone helps bond the layers better....
As seen in the photo below, the part on the right was dipped in acetone for about 5 seconds then air dried.... the part on the left is "untreated"

 

Understood, and agreed. But I want the best possible. I anticiapte the hand to be more likely slammed in doors, etc than a real hand, because it is not part of the body, and has no feeling. In addition, the hand has been gone so long that having something there again will not be intuitive; total awareness of it's presence (like, instinctual awareness, things that kick in when your coffee cup slides off the counter) will take a while to achieve.

In regards to getting the function ironed out first and worrying about the longevity later, sure I can see the logic of that. But I can see the logic of starting in any number of aspects. As I said before, there's no real good place to start. I chose to start here because this is where my interest and ambition took me first. I see no reason to fight it.

 

Have you guys tried t-glase? I just ordered a spool.

Here's what people are saying about it:

It's stronger than ABS with great layer bonding - more similar to nylon. But it's stiffer than nylon - closer to ABS in that respect.

Sounds good. I just ordered a roll, and I will subject it to my test and post the results; I will compare it to Nylon, ABS, and ABS dipped in acetone. I got the clear filament. Would be cool to have a clear(ish) hand.

 

Im aware of all or most of what you say; its almost a crap shoot, or it would seem so. It seems almost (in the 3d printing community in general) as if "results may vary," ...
...

Now you can see why I built a CNC router/mill machine for material removal, not a 3D printer for material deposition.

I can cut plastics or some metals from a solid block, very strong, and be able to use engineering plastics including; high strength types, high temperature types, glass filled types, lubricant filled types etc.

I hope it doesn't look like a gloat, but for actual engineering tasks a CNC router/mill is generally much better than a homemade 3D printer which (in my opinion) is still largely a hobby toy to tinker with.

 

Now you can see why I built a CNC router/mill machine for material removal, not a 3D printer for material deposition.

I can cut plastics or some metals from a solid block, very strong, and be able to use engineering plastics including; high strength types, high temperature types, glass filled types, lubricant filled types etc.

I hope it doesn't look like a gloat, but for actual engineering tasks a CNC router/mill is generally much better than a homemade 3D printer which (in my opinion) is still largely a hobby toy to tinker with.

You might be right on the money. I've wanted a CNC machine for a while. I might just be forced to go that route, but before chalking this 3D printer up to a $500 plaything, I'm going to exhaust every promising polymer out there and every processing trick I can find documented or think up.

 

RB, what type of plastic would you make this hand out of in your mill? UHMWPE or what?

 

RB, what type of plastic would you make this hand out of in your mill? UHMWPE or what?

I have several machines, a 3 axis cnc plasma cutter, a 3 axis mill, an d now a 3 axis 3D printer I find each machine is usefull in its owwn way, but not one is better than the other.

I would use Delrin if I was to machine these parts out of something light, Delrin has good impact resistance and good machining material, or for prototyping, I would use HDPE.

 

Delrin (also known as: poly acetal, or POM), is very chemical resistant, but can be very slippery.

I would actually of the other route and use cast silicone or polyurethane foam. Letting it crush and rebound is much easier than preventing it from crushing and cracking. Kind of the same strategy that Mother Nature used. A small steel, aluminum endo skeleton would make it perfect.

 

One thing to remember is that this project doesn't end with my dad's hand. The other half of it; the half that naturally gets lost in the scuffle, is that it is meant to be an open source project that is supposed to hopefully be able to be replicated by anybody with the will to do so. And without investing thousands of dollars in costly machines and becoming a 3D drafter, Thermoplastics Engineer, Mechanical Engineer, and an Electrical Engineer. I am willing to try on those hats to lay the groundwork, but the end solution should be as to-the-point, and simple as possible. That's part of the reason I chose the Prusa I3 kit; it is cheap, easily attainable, and capable. I envisioned this being a matter of "scale the hand to your size and click 'print'." Of course that's an over simplification, but that's the idea at heart. I would like to not stray too far away from that. But if it can't be helped, it can't be helped...

 

If that is the case, I would also view the hand as being disposable/rapid-rebuild. If they have access once, then they will have access to a 3D printer again another day when needed.

Also, stores like staples and Fedex/kinks have made announce,eats that they are considering to add (or will add) 3D capabilities to some shops. It is becoming more and more available. Replacement seems to be a more viable option than a printable polymer with the strength and elongation of cast titanium 64.

 

RB, what type of plastic would you make this hand out of in your mill? UHMWPE or what?

I generally consider PE to be garbage, don't be fooled by the "ultra" tag or the "high density" tags etc. It's soft, has a low melting point, warps and fouls tools when machining and is often warped when you buy sheet stock.

My two favorites are clear acrylic, the clear feature has a lot more value for prototyping than most people imagine. It's hard, machines well and it very strong provided you don't snap it from excessive force as it is a bit brittle.

Delrin is fantastic. Awesome to machine and a really strong, practically unbreakable material. But you can't glue it, it's not clear, and it's expensive.

For machining a hand I would use acrylic, it's cheap and the clear feature means that things like tapping threads are a dream as you can see the thread form. And it glues really well which gives a lot of options for adding something to an already finished part.

Once you have some finished and tested parts, I would echo GopherT's suggestion of making silicone molds and casting the parts. You can cast in many easy polyurethane materials including clear, and/or specific strengths and mild flexibiliy features etc.

 

I was working at the caterpillar plant recently and saw them lining the inside of dump trucks with UHMWPE, I just assumed that's what you want, if you want to be able to toss boulders onto a piece of plastic and have it survive. That's the extent of what I know about plastics. Thanks for the tips. I'll look into acrylic.

 

That stuff is good for wear. It's soft but is slippery/greasy type of plastic. And being soft the surface re-forms as the next impact smooshes the previous damage back together again.

I have tried a couple of times to make stuff out of HDPE and UHMWPE, by cutting shapes and bolting together etc. And always been disappointed with the result. It's too soft and weak to take a thread, so the last time I cut cavities and pressed steel nuts into them. But even then, when tightening the nuts it deforms etc.

I just don't like to make things from the stuff, but I guess it's almost a good material for a kitchen cutting board or a dump truck liner.

 

That stuff is good for wear. It's soft but is slippery/greasy type of plastic. And being soft the surface re-forms as the next impact smooshes the previous damage back together again.

I have tried a couple of times to make stuff out of HDPE and UHMWPE, by cutting shapes and bolting together etc. And always been disappointed with the result. It's too soft and weak to take a thread, so the last time I cut cavities and pressed steel nuts into them. But even then, when tightening the nuts it deforms etc.

I just don't like to make things from the stuff, but I guess it's almost a good material for a kitchen cutting board or a dump truck liner.

There are some really cool applications of UHMWPE from military to medical devices. It is cheap but can do things that no other material can duplicate - except lead.

 

This kit looks good for hobbyists although it probably is a little over priced if I'm honest.

It's made by Velleman: http://m.rapidonline.com/ProductSearch/ProductDetails?ProductCode=25-0000&ModuleCode=525459

That is a UK link although they do ship to the USA and all over the world at cost.

I'd imagine it is available from a USA distributor too.

Hope this helps

 

This kit looks good for hobbyists although it probably is a little over priced if I'm honest.

It's made by Velleman: http://m.rapidonline.com/ProductSearch/ProductDetails?ProductCode=25-0000&ModuleCode=525459

That is a UK link although they do ship to the USA and all over the world at cost.

I'd imagine it is available from a USA distributor too.

Hope this helps

That actually looks not too bad. Looks rigid, and clean with all the aluminum. I've already got a printer, but if I didn't, I might consider it.