3D Printers... do you have one? Do you recommend on? What do you use it for?

[RedRamage]

I've been sorta thinking about getting a 3D printer for a while now.  I know just enough about them to be dangerous though... I clearly don't know enough to be an expert on it so looking for some recommendations if anyone has one or has extensive experience I'd be interested to hear your thoughts.

Mostly I'd be getting it to make various odds and ends... like phone mounts for me car or headphone hook or what-have-you. I don't have a lot of requirements other than I'd something relatively easy in terms of setup so I'm not going massive calibration as I expect I'll be printing semi-infrequently.

Ideally something that isn't extremely loud (maybe an enclosed machine?) as it'll be in my home office.

I think that filament would be better than resin as different filament types would provide more flexibility, but I'm not necessarily opposed to resin if that makes the most sense.

So... any input?  Good experiences? Bad experiences?  Recommendations for or against any brand/type/whatever?

EDIT TO ADD: Price isn't a huge deal... I mean, I don't want to pay $5K for something.  I'm not opposed to paying a bit more for a better machine but probably looking to spend less than $1K if possible.

 

[Screwball]

I just saw this thread.

I used a 3D printer but it was about 4-5 years ago, so things are probably much different now.  We bought one for making prototype parts.  I was working for an appliance manufacturer.

We bought it off Amazon for about 2 grand. It was a little larger than a toaster, so the size of the parts were limited. It had different spools of plastic line in different colors.  The line looked similar to line for a weed whacker and came on spools.

You can make fairly complicated parts, but many times they would have to be sanded, filed, etc. to get the desired finished product, depending on the function.  That may have gotten better now that the technology is better.

But let's back up a little.  Whatever you want to make needs to start with a 3D model.  That's the toughest part IMO.  To create a 3D model you need some sort of 3D software.  That wasn't a problem for us as we had a high end 3D CAD software available at all times - this was our primary design software as it was.

Once you have a 3D model, it is exported as a .stl file to the 3D printer.  An .stl file stands for stereolithography type file.  Some may also use a SLA file.  This file gives the printer information so it can start at some point and lay down layers of plastic.  Once you send the file to the 3D printer, the software optimizes the exported model for the printer.  Our printer came with this software.  Depending on the shape, sometimes the software would add supports, gussets, or whatever to help it create the final shape.  This is also why you may need to final finish the part.

It is not hard to learn the process IMO, and trail and error will also help.  I don't know what is available today, and I'm sure they have gotten better for the price.

Depending on what you want to make, the 3D software might be key IMO.  There are all kinds of free and highly available 3D modeling packages, but they are not all the same.  Some are very cumbersome and hard to learn.  What is available for free I have no idea, but any commercial CAD package is better than the free stuff IMO, but you can't just go buy one of those - well you could if you wanted to spend the money - but not practical unless you are using it all the time.

My advice would be to look 3D modeling packages over and find one that you like, and will make the kinds of models you want.  Or, if you know someone with access to commercial 3D software, they could make you a model and export the .stl file.  One place you might look is a local school or college that teaches CAD.  They could maybe make you a model and export the file.

The 3D modeling is the hard part IMO.  Once created and exported, the printer and software will take care of the rest.

Hope this help, and if you have any questions, feel free to ask.

[gehringer_2]

2 hours ago, Screwball said:

The 3D modeling is the hard part IMO

We have our students try to learn a little 'Solidsworks' so they can do 3D printing for their projects right off the bat, and I'm always surprised at how good they are at figuring out how to configure their shapes so that they can be printed with FDM. I think those of us who grew up understanding reductive machining actually have a harder time with it!

[Screwball]

Designing shapes to fit 3D printing is a bad approach.

You design a part for form, fit, and function. ASME y14.5-19XX (year) would be a good place to start if they are teaching that stuff.

[gehringer_2]

4 minutes ago, Screwball said:

Designing shapes to fit 3D printing is a bad approach.

You design a part for form, fit, and function. ASME y14.5-19XX (year) would be a good place to start if they are teaching that stuff.

But if you have to prototype with an FDM....

The paradigm in my particular area is to get them doing something that generates some hardware with as little bootstrapping as possible. Only a fraction of them are even headed for ME. It's primarily about the experience of dealing with any/all kinds of constraints faced in getting from Idea to working model. The simplest possible electrical, computer and mechanicals. Learning how to do it right comes later. We want them find out at the beginning why it's worthwhile to learn the hard stuff needed to do it better.

[Screwball]

1 hour ago, gehringer_2 said:

But if you have to prototype with an FDM....

The paradigm in my particular area is to get them doing something that generates some hardware with as little bootstrapping as possible. Only a fraction of them are even headed for ME. It's primarily about the experience of dealing with any/all kinds of constraints faced in getting from Idea to working model. The simplest possible electrical, computer and mechanicals. Learning how to do it right comes later. We want them find out at the beginning why it's worthwhile to learn the hard stuff needed to do it better.

Quote

But if you have to prototype with an FDM....

True, but you don't design to it.  You 3D model a part for what will be needed if it goes far enough to be put in production.  The 3D printer, using it's own software will do the layering and optimization of the part to be printed (the prototype).

No different if it was a non prototype (non-production) part that you wanted for your lawn mower. You don't design to the layers even though it's all about X,Y,Z.  It's starts at 0,0, lays a substance of some sort along the X,Y plane, then indexes upward (Z) by whatever the software tells it (thousands of an inch), and repeats. This builds the part. Their software will figure that out.

This technology helped us manufactures learn if a design is viable in a shorter period of time.  R&D money is hard to get so time is as usual - priority one.  We bought one because time on the corporate machines were backlogged and hard on the projects budget.

Learning how to do things right should start from the very get-go.  Nobody said it would be easy. Prepare accordingly.  But to be honest, after watching corporate America over the last 40 years, I wondered many times how the hell they made or developed anything.

[gehringer_2]

23 minutes ago, Screwball said:

True, but you don't design to it.  You 3D model a part for what will be needed if it goes far enough to be put in production.  The 3D printer, using it's own software will do the layering and optimization of the part to be printed (the prototype).

No different if it was a non prototype (non-production) part that you wanted for your lawn mower. You don't design to the layers even though it's all about X,Y,Z.  It's starts at 0,0, lays a substance of some sort along the X,Y plane, then indexes upward (Z) by whatever the software tells it (thousands of an inch), and repeats. This builds the part. Their software will figure that out.

This technology helped us manufactures learn if a design is viable in a shorter period of time.  R&D money is hard to get so time is as usual - priority one.  We bought one because time on the corporate machines were backlogged and hard on the projects budget.

Learning how to do things right should start from the very get-go.  Nobody said it would be easy. Prepare accordingly.  But to be honest, after watching corporate America over the last 40 years, I wondered many times how the hell they made or developed anything.

One of the stranger exercises is to CAD something like up a simple bracket, specify the loads and then let a stress analysis program optimise it for 3D printing. I haven't done it myself, but we have people around who do and they get these weird looking things that look like close-ups of bone structure, or a heavy webbish sort of structure...  or ...something. Odd perforations all over the place and no flat surfaces except the mating faces.

Edited June 20, 2022 by gehringer_2

[Screwball]

That is strange because I don't know anyone who does that.  Let's walk through this one more time.

This is a simple part 3D modeled in Solidworks (since you used that as an example).  This is what you do first, no matter what 3D solid modeling package you have access to.

This part is then "exported" to the 3D printer using a .stl file which the printer can import and optimize.  By optimize (for lack of a better word), it will create the layering used for the printing process.  Think old dot matrix printer than can move up as well (z).  It starts at a point (home) and moves right and left (better yet x and y) on a flat plane laying down small droplets of plastic (for lack of a better word) to build the first layer of the part.  Once the first layer is put down, it indexes a few thousands upward, and prints another layer on the shape.  This is how the shape (part) is built.

The software will also do the stress analysis (if that's what you want to call it) for you.  For example; let's say you wanted to create a small airplane.  The body of the plane is fairly rigid, but the wings are not.  The software may choose to add a support/gusset/column at the ends of the plane wing because their own weight (as small as it is) may cause the wing to bend slightly therefore messing up the finished part.  Again, this should be left up to the 3D printing software, with exceptions of course.

A stress analysis program should be run on the real part, and is a very common part of the entire design process.  FEA (finite element analysis) is a great tool, but like 3D printers there are good ones and bad ones.

And yes, after the model is sent to the printer and optimized by the software, the finished model that is printed might have quite a bit of extra weird looking stuff attached.  This is why I explained above how you might have to sand, file, or massage the final part to get what you want.  Some of the supports/gussets/etc can be broken off if you are careful.

That is the most common way they do things the last time I was allowed anywhere near corporate America.

 

 

Edited June 20, 2022 by Screwball
clarity

[gehringer_2]

21 minutes ago, Screwball said:

The software will also do the stress analysis (if that's what you want to call it) for you.  For example; let's say you wanted to create a small airplane.  The body of the plane is fairly rigid, but the wings are not.  The software may choose to add a support/gusset/column at the ends of the plane wing because their own weight (as small as it is) may cause the wing to bend slightly therefore messing up the finished part.  Again, this should be left up to the 3D printing software, with exceptions of course.

Yes, this is sort of the normal routine. What I was referring to were SW packages  that run between the CAD and the slicer - primarily to optimize material use (and print time which follows). It's not something that is done conventionally but some thing the academic guys here play with. It's gaining use in upper level ME and CE use now. The correct terminology I was trying to come up with is 'Topology Optimization'. This clip is about 'Abaqus' which is what I was thinking about. It is an FEM based package that can generate 3d printable optimized topologies.

https://www.youtube.com/watch?v=i8rKJ5RlxF8

For instance this is the kind of thing it comes up with for a motorcycle frame. Kind of a kick ain't it? Looks more like it 'grew' than that it was fabricated.

Edited June 20, 2022 by gehringer_2

[Screwball]

Yes, that is very cool, but not what we are talking about here.  His work is about using the 3D printer AS the main manufacturing devise.  In our conversation here that Red is asking about is an entirely different animal as we are only making simple parts, even though the 3D printer in this case, is, the primary source of manufactured part.

The video is an example of where this technology has come over the last X amount of years, and the things it can make when the parts are too complex to be built in other ways (materials and the ability of different printers to use these different materials which has opened the door to these fancy type parts).

In our "normal" manufacturing processes to make our lives better via cheaper more affordable products (look around your house - almost everything is 3D modeled) prototyping is one of the first stages of the design process and the small 3D printers have been a huge help in time to market for our products.

And in this case, the weekend warrior who wants to make simple parts on a desktop printer is the audience.

[gehringer_2]

I have an FDM at home but almost wished I had gotten an SLA. Your material choices are much more limited but the design space is so much wider for the objects you can make.  One thing I would recommend now is to get a dual head FDM (which I didn't). Using soluble resin in a second head to create extractable support for geometries you couldn't otherwise do easily is something I wish I had. With a single head, your slicer will add support webs as needed, but you may not be able to get it out easily if that is what needed and the finish is usually poor where support is removed.

Edited June 20, 2022 by gehringer_2

[Screwball]

Follow up, been a long time.  Did you ever do anything Red?