Penguin: 3D Printing Applicability/Issues

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
Those look nice, but I'd be worried about the srength and for something outdoors the UV resistance. I have again from experience, little to no trust in either 3D printed or even stereo-lithographic plastic parts. The S-T from working with that process and the 3dP from the things my son brought home from his work. In real life those processes are mostly used to be a proof of concept hands on item not something that will last. But that's just my opinion and experience with it.
I feared this... I was going to 3D print parts for my penguin. Basically plates. I was going to print a 0.1” surface, backed by a 0.06” diamond grid for strength. I’m not sure of the type of plastic. I forget, but it’s in my online project file.

The plate’s outer diameter are subjected to the greatest forces. So I’m also considering ramping up the grid depth from a minimum at the outer edge, to a larger depth at the center (fulcrum) of the force.

I’m at a frustrating point of development. I have a prototype built from 5mm model aircraft plywood. But I need the “skin” to be completed before I can measure actual forces. My local business networking group is concerned because the skin development is being done by a creative whose actions are somewhat controlled by a trust...

My questions are:
  • Is 3D printing practical?
  • What other method of manufacturing should I consider? I like 3D printing, because I have almost complete control over the design using CAD software
  • Cost... I’ve budgeted manufacturing in my business plan based on 3D printed components. What increase/ decrease in manufacturing costs might I experience?
  • Are there any links/tutorials on 3D printing design? I’ve printed three different test models and still don’t have a working one. I’m having difficulty designing for material variations which results in my parts not fitting. For example, I’m making a sub-part which is a hinge. Model 1 fit perfectly, but the inner diameter for the hinge pin was too small. Model 2, the hinge pin fit, but the leaves didn’t fit. Model 3 - well - nothing fit!
  • At this point, I’m ready to start designing the entire plate, with the grid for strength, risers to support the hinges, external connectors, linkage mounting, and servo mounting. But if I can’t make a hinge - OMG what am I going to do. There are six design points and if I have to print each one three times, I’m going to go broke. I’ll consider professional design services, but can anyone help in selecting a resource? I went through this with another consultant, got an unsatisfactory product and the vendor ghosted me. Caused me a lot of grief and loss of reputation.
Five specific questions. Is there any AAC member with answers, even partial answers, to these questions?
 

xox

Joined Sep 8, 2017
565
I feared this... I was going to 3D print parts for my penguin. Basically plates. I was going to print a 0.1” surface, backed by a 0.06” diamond grid for strength. I’m not sure of the type of plastic. I forget, but it’s in my online project file.


The plate’s outer diameter are subjected to the greatest forces. So I’m also considering ramping up the grid depth from a minimum at the outer edge, to a larger depth at the center (fulcrum) of the force.


I’m at a frustrating point of development. I have a prototype built from 5mm model aircraft plywood. But I need the “skin” to be completed before I can measure actual forces. My local business networking group is concerned because the skin development is being done by a creative whose actions are somewhat controlled by a trust...


My questions are:

Is 3D printing practical?
What other method of manufacturing should I consider? I like 3D printing, because I have almost complete control over the design using CAD software
Cost... I’ve budgeted manufacturing in my business plan based on 3D printed components. What increase/ decrease in manufacturing costs might I experience?
Are there any links/tutorials on 3D printing design? I’ve printed three different test models and still don’t have a working one. I’m having difficulty designing for material variations which results in my parts not fitting. For example, I’m making a sub-part which is a hinge. Model 1 fit perfectly, but the inner diameter for the hinge pin was too small. Model 2, the hinge pin fit, but the leaves didn’t fit. Model 3 - well - nothing fit!
At this point, I’m ready to start designing the entire plate, with the grid for strength, risers to support the hinges, external connectors, linkage mounting, and servo mounting. But if I can’t make a hinge - OMG what am I going to do. There are six design points and if I have to print each one three times, I’m going to go broke. I’ll consider professional design services, but can anyone help in selecting a resource? I went through this with another consultant, got an unsatisfactory product and the vendor ghosted me. Caused me a lot of grief and loss of reputation.

Five specific questions. Is there any AAC member with answers, even partial answers, to these questions?

Yes, you can print very tough 3D parts so long as you use a good filament. A durable PLA works fine for most applications IMO. There are also techniques to improve the strength of your builds.


Design the models in your editor of choice and then upload them to the slicer. Just remember that it's obviously better to do lots of "tiny" test prints rather than several bad (complete) prints in a row.

Cost is mostly going to depend on the quality of filament. If you really want to save some money, buy in bulk!
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
@djsfantasi Post #1 was taken out of context. Let's put this into its proper context.
What is the "Penguin"?
I have several threads about my penguin. It’s an animatronic. Controlled by a proprietary interpreted language implemented on an Arduino Mega.

The attached picture shows the plywood prototype. I’m looking to mass produce (low volume) it’s components.

I’m looking to contract for the 3D printing. Shapeways has printed my test parts thus far. I’m not looking to print them myself, as there are hundreds of parts required. And I don’t have a 3D printer.

E25D9F1B-21B0-47BA-A686-8F85A2595D48.jpeg

Here’s a link to a video of the animatronic in action. The Wobblies have been corrected since this prototype. The “risers” needed stiffening.
 
Last edited:

MrChips

Joined Oct 2, 2009
23,536
3D printing is useful for proof of concept in the development stage. It is cost effective when making one unit. It is not cost effective for production runs. Injection molding, laser cut or die cut wood would be more cost effective.
 

shortbus

Joined Sep 30, 2009
8,588
While it is well shown on Youtube I'd never build anything using one. Or nothing I really cared about or meant to use. One thing they are good for is to make a "model". And that is all I've ever seen them used for. Guys are now using them to make casting patterns, to reproduce broken or missing cast parts for stuff like old cars and tractors. Much cheaper than wood patterns for making a sand mold, especially for just one or two parts.
 

shortbus

Joined Sep 30, 2009
8,588
Can you weld at all? Or solder with a Bernzomatic torch? Rather than plywood you could bend a small diameter round rod for the outer hoop and then solder/weld the cross pieces in, made from the same rod. The rod could even be heavy wire like they sell at a farm store and the cross pieces cold have a right angle bend on them to give more comtact where they meet the outer ring, to make things stronger.
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
Good idea! But I don’t think it will work.

Each out rim supports a RC servo, which pushed up on the weight of everything above it. Besides the skeleton, it also has to push up on the weight of the skin.

The upper motors use plastic arms, but I had to use solid aluminum arms on the bottom levels; the plastic arms couldn’t handle the weight. And the plywood has 1/2” ribs to stiffen them. I think this would bend any rods unless they were greater than 1/2” in diameter.

Plus, I’d have to weld plates for the motor and linkage attachments. I think the total weight of steel rods and plates would be heavier than 3D printed plastic.
 

shortbus

Joined Sep 30, 2009
8,588
Just thinking out loud. Will think some more and post if anything comes to mind.

Quote, "Plus, I’d have to weld plates for the motor and linkage attachments. I think the total weight of steel rods and plates would be heavier than 3D printed plastic." I can't imagine that being any stronger but would be lighter.

How big in outer diameter are the rings? How big of a center diameter?
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
The rings are 10”, 12”, 10”, 8”, and 6”.

Also, the cross bars have to support risers. They have hinges at the top, connecting to the next ring. There is a lot of sideways force on the risers.

Have you looked at the picture that I’ve posted? After reading my descriptions, you might get a better idea.
 

shortbus

Joined Sep 30, 2009
8,588
I looked at the pictures and the Youtube. I'm betting that I can make a metal ring including the cross bars and center rings that will both be stiffer and lighter than your plywood or a 3D plastic one. It would be made from 24GA sheet metal, similar to the thickness of a furnace pipe/duct.

There is a simple to do process in sheet metal work that makes very stiff light things possible, called "hammer forming". You make a wooden shape that you want, then another one that is the same shape but slightly (1/8 to 3/16)narrower. You clamp your sheet metal cut out in between the two pieces of wood, and slowly tap around the edges of the metal, bending them down to make a 90 degree edge. It's the edge that makes it stiff and the thinness of the sheet metal that makes it light. Think of a "C" channel brace made of thing sheet metal.

Here is a description of a pretty involved part, a car dashboard, being made to show what I mean.
https://blog.garritys.org/2015/11/hammerform-time.html
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
Interesting...

I am familiar with hammer forming. Isn't it how they make custom car body shells?

How would the thin risers between plates be formed?

The edges would have to be rolled, in order to attach to the cloth skin and protect workers during moving and dressing.

And how long would it take to make? Rough estimates would be more than 200 individual parts. 20 each of at least 10 unique parts.
 

shortbus

Joined Sep 30, 2009
8,588
There are to ways of hammer forming but they don't really differentiate them too much. The body shells are done using shot bags and stumps to make the contours of a body. Where the other that I'm talking about the metal is just pretty much flat and the edges are formed over to make things stiffer.

Any thing that can be made flat can usually be made from thinner metal by forming an edge on it. Look at most structural parts, like channels angle iron or I beams they are all stiffer and able to be made from thinner metal by the shape of their form.

Are you talking about the edges of the rings, when you say attaching the cloth skin? Those rings can also be hammer formed with a formed over edge.

What would take the most time is two things, cutting the sheet metal to shape and making the hammer forms them selves.

The sheet metal, many places now days have either plasma cutter tables or water jet tables. They would need your CAD files and cut the parts, really isn't all that expensive or so I'm told.


For the hammer forms for light metal like this and a small number of parts MDF would most likely work. Many people use it.

I'm betting after you get the things rounded up you could do the whole bunch of them in the same time you would 3D print a few of one size. Like most things the first couple would take longer untill you find what is the easiest way to do them.
 

shortbus

Joined Sep 30, 2009
8,588
More thinking out loud.

If hammer forming isn't to your liking, how about making the rings by vacuum forming? Doing it like that the whole ring and cross/spokes could be made in one piece and the area between the spokes could be left filled it too, to add more rigidity. Using ABS sheet would allow the skin to be glued on also. and where the motors and things need more strength a small backup plate could be glued on. https://en.wikipedia.org/wiki/Vacuum_forming

Another idea. Instead of hinges or risers between the rings, why not simulate a back bone? By using a aluminum flex tubing like they use for a dryer vent. And offsetting it toward the back of the animal a slight amount, like a real skeleton it might give it a more life like movement.
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
Now we’re getting into redesigning the whole body! The risers provide several advantages. They’re basically a spine with four bones. I get a large range of motion of the torso all the while minimizing the forces required for motion.

Bottom line is that 3D printing is not advantageous for mass production, it all depends on what mass production is!

I’ve priced out the costs incurred for 20 items versus the break even cost for injection molding. 20 items are cheaper to 3D print. I guess my requirement is below my break-even point.

If I were to order 100 or 1000 units, the economics would be different.

So back to my original question. While in terms of quantity, 20 units are economically justified, will 3D printed PLA pieces be strong enough to last for 6 months? Or more?
 

shortbus

Joined Sep 30, 2009
8,588
So back to my original question. While in terms of quantity, 20 units are economically justified, will 3D printed PLA pieces be strong enough to last for 6 months? Or more?
I don't know. A lot, from what my son tells me, about the strength of 3D printed stuff is the layering(I know that isn't the real phrase but can't think of it ). If you use a thick pass or layer there is less bonding to the layer beneath it but it prints fast. If you use a thinner layer it bonds more but takes much more time.

Was there a reason to dismiss the idea of vacuum forming? The molds/shapes would be easy to make, your already doing it with the plywood just change it to 1/4 MDF. The vacuum formed stuff is plenty tough can even be made from acrylic/plexiglass. Look at the RC car bodies that's how they are made.
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
I don't know. A lot, from what my son tells me, about the strength of 3D printed stuff is the layering(I know that isn't the real phrase but can't think of it ). If you use a thick pass or layer there is less bonding to the layer beneath it but it prints fast. If you use a thinner layer it bonds more but takes much more time.

Was there a reason to dismiss the idea of vacuum forming? The molds/shapes would be easy to make, your already doing it with the plywood just change it to 1/4 MDF. The vacuum formed stuff is plenty tough can even be made from acrylic/plexiglass. Look at the RC car bodies that's how they are made.
  1. Unfamiliarity with the process is why I dismissed VF. VF means thin cross sections unable to support 200 oz/in of force to me.
  2. Since both top and bottom contains mounts or protrusions, I didn’t understand how to make that with VF.
  3. The head contains multiple levels and would have to be created with many parts
  4. I don’t get how thin spacers could be created unless they use multiple parts. And the risers would have to be separate, a 4x6” rectangle that has to remain perpendicular while force is applied to it.
  5. How do you create a hinge with VF? There’s a 4-6” long hinge at each level
  6. How are Internal cutouts made?



 

andrewmm

Joined Feb 25, 2011
1,474
Mechanical engineering is a serious design effort. which I am not..

I've just bene around a bit and locked up a few bits.

For instance,

Take a bridge,
One made from wood or stone, or steel, or iron would have a different shape to do the job,
You design to the properties of the material, minimise the cost, and meet the design requirement,

For instance , I'd be asking does the unit have to be made out of all the same material ?

Take 3D printing,
You are paying per ml of plastic.

So if you make a thick part, its normally to make it honeycomb, so you save plastic,
disadvantage is you tend not to get such a smooth surface finish,
that might be OK, or you might want to look at over printing, to get a better finish.

3D printing you can also have multiple colours, so you could make a different colour top / bottom in one piece,

The number of variables in how to make something are to me almost infinite,
and as with the bridge, how to make in one material might not be the way to make in another.
 

shortbus

Joined Sep 30, 2009
8,588
Unfamiliarity with the process is why I dismissed VF. VF means thin cross sections unable to support 200 oz/in of force to me.
Why do you think it only works with thin cross sections? 1/16" is done easily and your spokes will make it as strong as the plywood is. If 3D printing with all of it's layers that may or may not bond will work, solid plastic sure should. With VF and a thin form like your plywood looks like(1/4"?) 1/8" would even work. Added strength comes from not having to make a triangular cut outs between the spokes they would be the plastic that is pulled down to the bottom level, this is why I said 1/16" material.

Since both top and bottom contains mounts or protrusions, I didn’t understand how to make that with VF.
How did you make the protrusions with the plywood one? If they were glued or screw to ply the same process will work with VF.

I don’t get how thin spacers could be created unless they use multiple parts. And the risers would have to be separate, a 4x6” rectangle that has to remain perpendicular while force is applied to it.
Thin backup or mounts would be cut and glued to make the area that needs the extra strength, It would be like making the same area thicker with 3D.

How do you create a hinge with VF? There’s a 4-6” long hinge at each level
What did you use with the plywood? Same hinge for VF ring, with small screws and nuts.


How are Internal cutouts made?
Again same way you did with plywood. Your just substituting plastic for wood.


Unfamiliarity with the process is why I dismissed VF.
I can't imagine that. Don't you make Halloween stuff and things for theater? There is or was a lot of Youtube people doing this, VF, to make things for that and cosplay.
 

Thread Starter

djsfantasi

Joined Apr 11, 2010
7,693
Yes, I did. But the cross-section of my props was less than 0.04”. Some were like 0.01”. And the VF piece wasn’t under strong forces.

You say I just do several things the same way I did plywood. But that was only a prototype. I don’t want all those extra steps in the assembly process. The hinge was printed as a 1/4” by 3/8” solid piece with mating leaves and a centered 1/8” hole for the pin. Not possible with VF. And for internal strength, external hinges aren’t desirable.

The risers are a 1/16” solid plane with side supports of 1/16” struts. If I were to VF that, I’d have to make two individual sides, glued together and glued (or otherwise attached) to the plates. And I don’t see how to include the hinge.

Strength is added by thin ribs stiffened by a triangular honeycomb.

Mountings for the RC servos and linkages are just printed as part of the base. No dealing with separate parts to be assembled

With 3D printing (by a commercial company) cutouts are just not printed. Plus prints are guaranteed. I have expectations of using a commercial company. That includes no separation of layers and free reprinting if that happens.

I imagine final assembly to be:
  1. Mount a pre-registered servo to a servo mount with two self-tapping screws into printed holes
  2. Slide in a hinge pin to pre-printed hinge
  3. Use ACC glue on one end of the hinge pin
  4. Connect servo arm and linkage ball joint to RC servo
  5. Connect linkage arm to ball joint in pre-printed linkage mount
And repeat for the remaining plates. No time spent attaching servo mount, linkage mount, riser, stiffening ribs or hinges
 
Top