3D printing an engine frame?

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
The first "word of wisdom" is that learning 3D CAD is difficult, regardless of the program. I started with Fusion 360 but figured out fairly quickly that Autodesk was not likely to leave this free to hobbiests with all the features. FreeCAD may not be as easy to use and may not have all the features but the features are being added, not subtracted.
 
The first "word of wisdom" is that learning 3D CAD is difficult, regardless of the program. I started with Fusion 360 but figured out fairly quickly that Autodesk was not likely to leave this free to hobbiests with all the features. FreeCAD may not be as easy to use and may not have all the features but the features are being added, not subtracted.
Agreed! About 12 years ago I had a student license for SolidWorks and I was doing real well with it, but that was when I had a Windows computer at work to use. Powerful program, and as a veteran, I am still entitled to student rates with that program. However, I’m a dyed in the wool Mac user (since 1986), retired, and no longer have access to a Windows machine with enough RAM, etc. Therefore, it’s time to re-educate myself and dive headfirst into FreeCAD, which as you’ve said, continues to grow in capability.

We sure do have a great hobby, and I feel very lucky and privileged to be surrounded by such a talented and helpful group of fellow enthusiasts. Thanks for all the guidance folks!

John W
 
The first "word of wisdom" is that learning 3D CAD is difficult, regardless of the program. I started with Fusion 360 but figured out fairly quickly that Autodesk was not likely to leave this free to hobbiests with all the features. FreeCAD may not be as easy to use and may not have all the features but the features are being added, not subtracted.
I disagree with you on 3D CAD being difficult. HOwever, it IS quite helpful to have a good tutorial for whatever program you use.
 
After starting to learn 3D CAD using FreeCAD this spring, I’ve become fairly comfortable with the basics. I’ve designed and printed a number of useful projects and slso some not so useful ones. Although I’m slowing down on this basement workshop stuff for the summer I plan on going full out again come fall/early winter.
Getting back to my original post, I’m still anxious to explore how 3D printing can be used in building a model IC engine, in my case the Tiny. The frame is an obvious candidate but less obvious are the gears and perhaps even the camshaft. I’ve become comfortable using PLA, PETg and TPU but understand I’ll have to become proficient at using other filament types to continue.
If there is any interest, I’d be up doing a group project on say a small air engine using FreeCAD and a 3D printer. Could be fun!
 
I use Solidworks, and have gotten heavily into 3D printing patterns.
Example below.
Learning 3D was difficult for me because there were no tutorials that showed how to design engines using 3D modeling, and because the concepts were so different than 2D drafting.

rImg_3082.jpg
rImg_3087.jpg


rImg_9015.jpg


rImg_9020.jpg
 
John, 3d printing is indeed fun - and despite what one sometimes reads, useful.

Generally, the articles that say that 3d printing failed to live up to its promise note that people thought they had a Star Trek replicator that would automagically print whatever they needed ... and then they discovered that 3d design requires work. :) More about this below ...

I am happy to share my thoughts on 3d printing, with the understanding that this represents only my own experience and needs; mileage may vary. First, thoughts on filaments:

1) PLA is actually stronger than you may think. It is easy to dismiss it as a "beginner" filament, but really, it is quite versatile and reasonably strong. Its one great weakness is that it can deform at relatively low heat, so don't plan to use it for a part that lives inside a hot car, for example.

2) PETG is a nice all-around filament. It is softer and more flexible than PLA, so one might think in terms of PLA being strong but brittle while PETG is more tough and flexible, with better performance in hot environments (though only relatively better than PLA - it might survive the hot interior of a car, maybe, but nothing hotter than that).

3) ABS is not as hard to print as you may think, IF you have a heated bed that can go to 100°C AND you use "ABS glue" (a thin solution of some ABS dissolved in acetone) to wash the plate. But it has poor UV resistance, is smelly when printing (potentially dangerous fumes), and in general I have not found a reason to use it over PETG or PLA.

4) Nylon seems like such a great idea - it should be super tough, great for wear parts - right? Well ... the first problem is, a stock Ender 3 or similar printer is not going to print it successfully. Nylon WARPS like mad; even with a 100°C heated bed, it is nearly impossible to print anything other than a very small part without corners lifting and the part warping - unless you have a heated enclosure, which I do not. Add to that the fact that Nylon loves to absorb water, changing dimensions in the process, and I decided that this filament was not so great after all.

5) POM aka acetal - the common trade name is Delrin - again, seems like such a great idea. I LOVE to use Delrin in machined parts for strength, stability, low speed bearing surfaces, and more. Printing it is ... challenging. Even worse than Nylon, POM is near-impossible to print without a heated enclosure. I have had some success with small parts, such as a linear bearing printed in "vase mode" or a small bushing for a wheel.

6) TPU - flexible filament - I have had success in printing both 85 and 95 TPU (the number referring to how flexible it is). The conventional wisdom is that it requires a direct drive extruder ... but I have been successful with a Bowden extruder. The key is a custom-designed extruder that completely constrains the filament path so that it cannot "squirt out" around the extruder gear. Also key - to either Bowden or direct drive - is greatly slowing down the print speed.

Note that some (most?) of these filaments need an all-metal hotend; depending on the nature of the bed, some of these benefit from some sort of stick/release layer (e.g., glue stick); in particular, without glue stick or similar, TPU can stick so tightly to the bed that it will not release without ripping up chunks of the bed.

Note also that there are tons of variations on many of these filaments, and some of those variations can make a significant difference in strength and/or stability - a particular example being carbon fiber added to the filament. But note also that many fillers, such as carbon fiber, are abrasive and will require the use of a hardened nozzle for best results.

This is getting way long, so I will continue in a second post.
 
Continuing on thoughts / experience with 3d printing, with focus on the usefulness and design issues:

I got into 3d printing thinking that it might turn out to be one of those things that is interesting at first, but ultimately goes onto the shelf to catch dust. Nope, definitely not. I don't print every day, not even every week; sometimes a month or more might go by without turning it on. But over and over again, I find myself needing something for which 3d printing is perfect, and the printer goes into action - often with several design iterations until I get just what I want and need.

Some of the highly useful parts that I have 3d printed:

- Boxes to hold gear cutters, drill bits, reamers, taps, etc.
- Boxes to hold electronics and other projects such as an ignition system for model engines
- A teleprompter (works incredibly well)
- All sorts of knobs
- Car parts, such as the gear selector button and a replacement key fob case for my daughter's Honda Accord
- Household items, such as curtain rod holders, handles, funnels, lids, soap dishes
- Jigs and templates for wood and metal working - e.g., a jig to hold items securely at a given angle or shape, or a layout template, or so on
- prototypes - such as an ignition cam for a model engine, or a mounting plate for the band saw, or endless others
- Miscellaneous parts for machines and projects - spacers, shrouds, faceplates, grills, mounts
- And on, and on, and on

Of course, all of these had to be designed. I find myself using two tools for designing 3d printed parts. One is FreeCAD; the other is OpenSCAD. They are very, very different in approach. FreeCAD uses the familiar WYSIWYG sketch-extrude-loft and boolean combine/cut 3d modelling that is similar to Fusion360 and others. Each step of the modelling process builds on the previous step, and you see the effect each step has. OpenSCAD, by contrast, uses a programmatic approach, where one builds up the model by instructing it to manipulate and combine/cut basic primitives. Any time you want to see the results, you have to render the model - essentially rebuilding it from start to finish each time.

It might seem strange, but for many projects I find OpenSCAD to be faster and easier to use. For other projects, FreeCAD is the clear winner.

While FreeCAD is "parametric" - meaning you can adjust various dimensions and features without having to rebuild the whole model - there are limits on what will work. For example, it is generally trivial to change the dimension of a hole or the length of a side. But changing a knob from 5 lobes to 6 is probably going to involve a redesign process. In OpenSCAD, by contrast, everything is parametric. If you build your program correctly, you can set up an OpenSCAD project that lets you enter the desired features of that same knob - OD, height, number of lobes, bolt size, etc. - and it will build the model accordingly. I have built up a collection of such projects, so now when it comes time to make a new box to hold a new set of reamers, or a new knob, or a new button or spacer or bushing, I just plug in the numbers and go.

I am sorry for these lengthy posts, and I hope they do NOT come across as any sort of "look at me"; my goal is rather to help anyone think about all of the ways that a 3d printer can be useful!
 

Attachments

  • IMG_8094.JPG
    IMG_8094.JPG
    112.1 KB · Views: 53
  • frame1.png
    frame1.png
    1.5 MB · Views: 61
  • IMG_7977.JPG
    IMG_7977.JPG
    182.4 KB · Views: 50
  • IMG_7983.JPG
    IMG_7983.JPG
    189.7 KB · Views: 53
  • IMG_7996.JPG
    IMG_7996.JPG
    163.7 KB · Views: 52
  • AC Control Knob.png
    AC Control Knob.png
    26.1 KB · Views: 54
  • Honda Shifter.png
    Honda Shifter.png
    32.8 KB · Views: 52
  • Key Fob.png
    Key Fob.png
    57.5 KB · Views: 48
  • curtain rod holder.png
    curtain rod holder.png
    11 KB · Views: 55
  • holding jig.png
    holding jig.png
    25.6 KB · Views: 49
A very comprehensive and interesting report from Andy.

I was very sceptical about 3D printers having seen some work and parts that were truly junk. Then I saw some guy printing carbon fibre filled PLA on Youtube using a Dremel printer, I thought the results looked good and bought a second hand Dremel 3D20 for £200.

This turned out to be money that was very well spent as I have had some very useful parts off of the machine, As with most things it is how you use the machine and apply yourself to the task that determines the result.

The Dremel will print right out of the box with just the print bed needing levelling. After levelling the Dremel can just just start printing the trinket files that come with the machine.

I have only ever printed two free download files, everything else has been original design. I sue the Simplify 3D programme to slice STL models that are produced in a number of 3D CAD packages.

Below are some examples of real work that can be done with good FDM printers feed with good code and using decent filament.

The two orange models are from the moulds produced from the yellow pattern. The upswept wing tips on the models are actually direct 3D printed parts and black arms form part of the aileron control linkage on the same model.

The fuel tank holds methanol and nitromethane fuel and is printed with a good quality PLA, we just drain those tanks down carefully at the end of the day.

Any day now I will be printing some wax for doing some investment castings, that will be interesting.

B.
DSC_1272.JPGmoulds.jpgC32 pylon racers with 3d printed wing tips.pngDSC_1612.JPGDSC_1736.JPG
 
After starting to learn 3D CAD using FreeCAD this spring, I’ve become fairly comfortable with the basics. I’ve designed and printed a number of useful projects and slso some not so useful ones. Although I’m slowing down on this basement workshop stuff for the summer I plan on going full out again come fall/early winter.
Getting back to my original post, I’m still anxious to explore how 3D printing can be used in building a model IC engine, in my case the Tiny. The frame is an obvious candidate but less obvious are the gears and perhaps even the camshaft. I’ve become comfortable using PLA, PETg and TPU but understand I’ll have to become proficient at using other filament types to continue.
If there is any interest, I’d be up doing a group project on say a small air engine using FreeCAD and a 3D printer. Could be fun!
you should be able to find gear models on various web sites you just need to know if metric or imperial some you will get the exact size others you may have to specify . I’m not really up to speed on the new 3D plastic machines . My last experience was in the early days of rapid prototyping. It wasn’t too hard but you had to be mindful of wall thickness . Some times a sample part was made to check actual sizes then a minor adjustment allowed very close fits.
 
Continuing on thoughts / experience with 3d printing, with focus on the usefulness and design issues:

I got into 3d printing thinking that it might turn out to be one of those things that is interesting at first, but ultimately goes onto the shelf to catch dust. Nope, definitely not. I don't print every day, not even every week; sometimes a month or more might go by without turning it on. But over and over again, I find myself needing something for which 3d printing is perfect, and the printer goes into action - often with several design iterations until I get just what I want and need.

Some of the highly useful parts that I have 3d printed:

- Boxes to hold gear cutters, drill bits, reamers, taps, etc.
- Boxes to hold electronics and other projects such as an ignition system for model engines
- A teleprompter (works incredibly well)
- All sorts of knobs
- Car parts, such as the gear selector button and a replacement key fob case for my daughter's Honda Accord
- Household items, such as curtain rod holders, handles, funnels, lids, soap dishes
- Jigs and templates for wood and metal working - e.g., a jig to hold items securely at a given angle or shape, or a layout template, or so on
- prototypes - such as an ignition cam for a model engine, or a mounting plate for the band saw, or endless others
- Miscellaneous parts for machines and projects - spacers, shrouds, faceplates, grills, mounts
- And on, and on, and on

Of course, all of these had to be designed. I find myself using two tools for designing 3d printed parts. One is FreeCAD; the other is OpenSCAD. They are very, very different in approach. FreeCAD uses the familiar WYSIWYG sketch-extrude-loft and boolean combine/cut 3d modelling that is similar to Fusion360 and others. Each step of the modelling process builds on the previous step, and you see the effect each step has. OpenSCAD, by contrast, uses a programmatic approach, where one builds up the model by instructing it to manipulate and combine/cut basic primitives. Any time you want to see the results, you have to render the model - essentially rebuilding it from start to finish each time.

It might seem strange, but for many projects I find OpenSCAD to be faster and easier to use. For other projects, FreeCAD is the clear winner.

While FreeCAD is "parametric" - meaning you can adjust various dimensions and features without having to rebuild the whole model - there are limits on what will work. For example, it is generally trivial to change the dimension of a hole or the length of a side. But changing a knob from 5 lobes to 6 is probably going to involve a redesign process. In OpenSCAD, by contrast, everything is parametric. If you build your program correctly, you can set up an OpenSCAD project that lets you enter the desired features of that same knob - OD, height, number of lobes, bolt size, etc. - and it will build the model accordingly. I have built up a collection of such projects, so now when it comes time to make a new box to hold a new set of reamers, or a new knob, or a new button or spacer or bushing, I just plug in the numbers and go.

I am sorry for these lengthy posts, and I hope they do NOT come across as any sort of "look at me"; my goal is rather to help anyone think about all of the ways that a 3d printer can be useful!
Thanks Andy, excellent advice as usual! I’m trying to draw Elmer’s Geared Engine (I believe it’s number 5), the one with the inner ring gear. I’ve gotten the ring gear and spur gear modeled already, and will begin the frame valve box, and cylinder. I’m planning on using a glass dashpot as a cylinder liner, with a graphite piston. Perhaps printed in ABS to give me a little more temp. resistance (I live in Arizona, and it gets mighty hot!).

Someone mentioned an interest in a group project, maybe this would be a good one to try out.

Thanks again for your usual excellent advice,
John W
 
Thanks Andy, excellent advice as usual! I’m trying to draw Elmer’s Geared Engine (I believe it’s number 5), the one with the inner ring gear. I’ve gotten the ring gear and spur gear modeled already, and will begin the frame valve box, and cylinder. I’m planning on using a glass dashpot as a cylinder liner, with a graphite piston. Perhaps printed in ABS to give me a little more temp. resistance (I live in Arizona, and it gets mighty hot!).

Someone mentioned an interest in a group project, maybe this would be a good one to try out.

Thanks again for your usual excellent advice,
John W
one thing or engine I’d like to see is ageless 18 cyl radial as a clear or see through plastic model . I have never seen a model r 4360 27 cyl . RadialBut 3 d printing could make a desktop model possible
 
I would be up for a group FreeCAD project. That would be an excellent way for anyone interested to get a good handle on how to use it.

I know the Elmer engines only by name ... are there any public domain plans available? That would make it considerably easier to do a public group project; if the plans are copyrighted, I would think each participant would have to purchase plans, and the group FreeCAD work would need to be kept private to respect the copyright.

An 18 or 27 cylinder radial might be a bit too big a bite for a group project involving folks new to and learning 3d CAD ... but I agree that I'd love to see a clear model of one!
 
Hi Folks,

It appears that a legal version of Elmers Engines is available at:

https://www.myheap.com/metalworking/resources-downloads.html
I seem to recall that the family choose to make it available for free, but might be thinking of the Engine Boys books.

In any event, it's good to have these books available, far too many valuable plans or collections of a lifetimes work simply disappear when a magazine or author dies or fails. Think of all the great plans that were in Modeltec and a few other fallen mags.

Please note that this version of Elmers Engines is not placed in the public domain, just made available by the copyright holder for free.

Best to all,
Stan
 
Thanks, Stan. Hmm ... "not in the public domain, but made freely available" - sounds something like one of the "Creative Commons" licenses, which retains the copyright but makes it freely available with attribution.

What say ye all - shall we tackle #5 as a group FreeCAD project? And how should we go about it? Here's an initial thought - not at all sure this is the best way to do it - just to get the ideas flowing:

1) Agree on the order of developing the parts (e.g., first the cylinder, then the piston, then ...)
2) Agree on a schedule, e.g., one week per part
3) Each person who wants to participate can work on modelling the part for the first half of the week; then in the second half of the week, everyone shares their approach / ideas.

Again, I am not at all sure that the above is a good approach or reasonable time frame or anything else; hopefully, though, it moves us closer to figuring out how to proceed.
 
I too love to use my Ender each time I have to make a little gizmo...mainly because it is so easy to reprint a refined version !

there is a pair of beautiful yellow helical gear in a picture above (# 25, Greentwin)
I wonder if it is possible to model it in 3D with freecad too ?
I'm just starting with freecad...
 
I too love to use my Ender each time I have to make a little gizmo...mainly because it is so easy to reprint a refined version !

there is a pair of beautiful yellow helical gear in a picture above (# 25, Greentwin)
I wonder if it is possible to model it in 3D with freecad too ?
I'm just starting with freecad...
I've not had occasion to try it out, but there is an add-on workbench that can do all sorts of gears, including helical: FCGear Workbench - FreeCAD Documentation
 
you should be able to find gear models on various web sites you just need to know if metric or imperial some you will get the exact size others you may have to specify . I’m not really up to speed on the new 3D plastic machines . My last experience was in the early days of rapid prototyping. It wasn’t too hard but you had to be mindful of wall thickness . Some times a sample part was made to check actual sizes then a minor adjustment allowed very close fits.

This guys instruction videos on gears for fusion 360 users I found very helpful.

He is not missing out some of the important details that make the difference between a 3d printed gear pair, and a 3d printed gear that "cooperates" with existing standard gears.
 
Thanks, Stan. Hmm ... "not in the public domain, but made freely available" - sounds something like one of the "Creative Commons" licenses, which retains the copyright but makes it freely available with attribution.

What say ye all - shall we tackle #5 as a group FreeCAD project? And how should we go about it? Here's an initial thought - not at all sure this is the best way to do it - just to get the ideas flowing:

1) Agree on the order of developing the parts (e.g., first the cylinder, then the piston, then ...)
2) Agree on a schedule, e.g., one week per part
3) Each person who wants to participate can work on modelling the part for the first half of the week; then in the second half of the week, everyone shares their approach / ideas.

Again, I am not at all sure that the above is a good approach or reasonable time frame or anything else; hopefully, though, it moves us closer to figuring out how to proceed.

Is it this one? You are refering to? Nasty internal gear and some smaller parts?



Greetings Timo
 
Is it this one? You are refering to? Nasty internal gear and some smaller parts?



Greetings Timo

I believe that is the one that was suggested above. Personally, I'm open to whatever people want to work on. This one would be more challenging to build due to the internal gear, but it would not be any harder to model in FreeCAD.
 

Latest posts

Back
Top