Resize a STL File For Me

[MarkF]

I'll make it real simple:  NEVER USE CTRL-A in FreeCAD!

This is version 0.16?

Either select 'Block' or 'Fillet007' to export.
Assuming 'Block' is a 0.21 body. 
If 0.16, you probably can't select 'Block' and your only choice is 'Fillet007'.


Select ONLY ONE item in the tree to be exported.

[MarkF]

The FreeCAD tree is a recipe for Chocolate Chip Cookies.

When you want a cookie:

  YOU DO NOT
    - Add flour, sugar, eggs, chips, etc. to a bowl
    - Mix the dough
    - Drop on a cooking sheet
    - Bake for 10 minutes
    - Let cool
    - Eat

  YOU JUST
    - Eat the cookie


If you make everything you selected for export visible, you will see what you have in the slicer.

Also, you can fillet more than one edge at a time.  For edges with the same fillet size, you can add additional edges by CTRL-Click more edges before doing the fillet.  You can also add/remove edges from the fillet tab.

[bostonman]

  YOU DO NOT
    - Add flour, sugar, eggs, chips, etc. to a bowl
    - Mix the dough
    - Drop on a cooking sheet
    - Bake for 10 minutes
    - Let cool
    - Eat

The analogy is not only a good explanation, but hilarious. I get the tree structure better now, but, to be a bit sarcastic. If I add a sketch (sketch1) to a face, and then 'pad' it or make it a 'pocket', the tree becomes <pocket> and underneath it <sketch1>. i.e. 'sketch1' is the last in the tree.

This is similar to adding "clean the plate" which now becomes the last item on the list under "eat", so (normally) I would have picked "clean the plate" thinking "clean the plate" wouldn't be included if I selected "eat the cookie". Hence, I'd lose "sketch1" because I selected the "pocket" above it.

Yes, I'm being a bit sarcastic, but I'm also pointing out that it's not as simple as picking the last in the list.

Now that I get the tree structure better, I should have less issues exporting.

[MarkF]

I must point out that a sketch is not a 3D-object.

It's just a two dimensional drawing.
It has no height.
It has no interior.
It's not solid.
It's not something you can print.

No. It's not as simple as picking the last item.
You have the option of picking anything.
To over simplify, my point is you ONLY pick one item.

[bostonman]

I watched the video posted in Reply #216 and thought it was excellent (I need to watch it again), but helped explain adding shapes to non-planer shapes - this was the video regarding adding shapes to non-planer surfaces and involved making a bracket (?).

Regarding that video and a discussion from prior. I asked whether "touching" a shape to another shape would result in a weak bond when printing and believe the answer was overlap is fine. My question revolved around a box and a cylindrical shape. I asked whether "touching" the cylindrical shape to one of the sides of the box would result in a weak bond.

In that video, he created half the bracket (?) and then mirrored it to make it a full bracket. Does that create two "halves" (i.e. a microscopic gap) that will result in a weak bond when printed or does the software (FreeCAD in this case) know enough to make it one uniformed shape?

Another question I had is piggy backing off my last one. In the attached shape (I tried to come up with a comical name since some of you enjoy my file naming), let's call it a purse since it resembles that, how do I edit the non-planer part to say curve the strap a little to avoid the edges cutting into the hand carrying it?

Not curve the entire strap into a U because I think that may be simpler since it's a uniform shape, but say I wanted to curl the edges slightly and leave the rest of the strap flat without cheating by making a fillet on the faces.

[MarkF]

In that video, he created half the bracket (?) and then mirrored it to make it a full bracket. Does that create two "halves" (i.e. a microscopic gap) that will result in a weak bond when printed or does the software (FreeCAD in this case) know enough to make it one uniformed shape?

Since the bracket has symmetry around the center, he saved time by only drawing half and mirroring it.  The results would be the same if he had drawn the whole bracket without the mirror.

After the mirror, there won't be any microscopic gap.  The bracket will be one complete solid.  The strength will depend on the cross-section where the mirror occurred.  The same with the two shapes in the example you created.  Once you have them so they form one object, the strength comes from the cross-section where they meet.  As you build your shape, it becomes a solid and the strength comes from the cross-section at any point and the characteristics of the material you print with.

Another question I had is piggy backing off my last one. In the attached shape (I tried to come up with a comical name since some of you enjoy my file naming), let's call it a purse since it resembles that, how do I edit the non-planer part to say curve the strap a little to avoid the edges cutting into the hand carrying it?

Not curve the entire strap into a U because I think that may be simpler since it's a uniform shape, but say I wanted to curl the edges slightly and leave the rest of the strap flat without cheating by making a fillet on the faces.

Here you need to think outside the box a little.  No one says the handle has to be a single circle.  In the sketch, build the upper half as a semicircle.  Then extend the ends with an arc to connect to the purse.  The extensions can be of a different diameter than the semicircle to get the shape you want.  If you want to go overboard, you can use splines or pipes to create a custom shape.

"So, how long have you been carrying a purse?"

[bostonman]

The strength will depend on the cross-section where the mirror occurred.

I must have misunderstood initially.

So if I create two cubes (let's say a pair of dice) and "touch" them, it will print as one solid (the same as if I created one large square/rectangle) rather than a microscope gap between the two; and it would be a strong bond since the cross sectional area is quite large (assuming a standard size pair of dice and the entire sides are touching).

"So, how long have you been carrying a purse?"

As soon as I send this 3D design to the printer and it prints.

[bostonman]

I created two blocks and "touched" them. I created the first block, added a second to the face, created geometry lines, and used the geometry to start the side of the second block.

Looking at the slicing software, it appears it's one "complete" solid part, and believe I answered my own question.

I assume if I put a microscopic space between the two blocks of something so small such as 0.001", then it would technically be two separate blocks that could split apart; assuming a perfect world and perfect printer.

[MarkF]

I can't figure out where you're going with all of this.

You seem to be banging your head why you get the 'multiple solids not currently supported' in FreeCAD.  Most of the time this is due to limitation in the precision of floating point numbers.  Sometimes a 5 in a floating point number is 4.999999999999.  You will notice in the last tutorial he pointed out where there was a small gap due to the angles involved and the solution is to overlap the pad a small amount.  It doesn't hurt to embed the pad into the other solid to fill in any gaps.  This shouldn't be a problem if you are able to map the sketch onto a face.

In your dice example, look at how the layers are printed.  If the nozzle moves in two loops for each layer, it will not be as strong as if the nozzle moved in one big rectangle.

[bostonman]

The direction I was going was just understanding whether (using the video as an example) mirroring half the bracket results in a weak point in the middle because it would print them separately or whether it ends up being one complete solid uniform piece.

To experiment, I tried a similar approach by just touching two cubes together (using geometry) and the slicing software showed that it was truly printing it as one complete uniform piece.

My statement that if I added a microscopic gap by placing the second cube a fraction of an inch away, then technically it would be printed as two separate pieces.

This is all I was stating. I wasn't doubting the feedback/advice from anyone, and sometimes my messages are rhetorical or stating that I took the time to "experiment" in order to get an understanding of what was discussed.

[Old Printer]

Bostonman, you can put this to bed by doing a couple test prints. Put two blocks together as you have described then watch them print. My experience is limited to a Fusion, Bambu P1S and Bambu Lab slicer, and I had this same concern. My tests printed as one. The nozzle went past the seam as if it was not there and the finish in the join area showed nothing.

[bostonman]

Yes, you're correct, but it was answered by others already and seeing the slicing software showing it as a whole helped.

Hopefully nobody took my questions as doubting the answers or any other negative fashion. Sometimes I ask obscure questions that are interpreted as a waste or explain how I proved the answer to visually see the results, but it confuses people because they already provided an answer.

It's because I'm the type that if I'm driving somewhere, come to a fork in the road, knowing I need to bare right, I may bare left just to see what is there.

[bostonman]

Does a way exist to pad at an angle, or does fancy geometry need to be done?

I’d like to attempt making some missing feet for a table saw stand - see attached.

I can create most of it except the angle.

[MarkF]

I don't understand what you want to pad?

Can you describe it more or draw over your photo?

[MarkF]

Oh...  The photo is the actual foot you want to duplicate.

Just draw the triangle on the bottom and pad it.
Then draw the right angle and pad it.
Both sketches on the XY plane.

You can pocket out the interior or just use the slicer fill pattern to do it.

[bostonman]

I made the mistake of thinking everyone could read my mind.

I started by making the triangled flat part that sits in the floor and padded it, maybe 3/16”.

After I created another sketch on top making the V shape, padded it (maybe 3/4”), and then added a groove for the slot.

The issue is trying to make that V with the groove an angle.

Maybe the steps I’m taking are wrong, but that’s the route I took (although it was a rough drawing without correct dimensions).

[MarkF]

So the V or right angle tall part is not vertical but sits at an angle to the floor?

Does it look like this?

 

[bostonman]

Not sure if this picture helps.

Basically what you posted seems to be the same as the feet I’m looking to duplicate.

[MarkF]

I still can't tell.

Basically the table saw legs are a piece of angle iron?
They go inside of the plastic foot in the groove shown?
Are the legs angled from the floor or are they perpendicular to the floor?

[MarkF]

I can draw it with a 'pipe' but that may still be beyond you yet.
Take a look.  It's NOT fully constrained and you would need to adjust ALL the dimensions.

[bostonman]

I couldn’t get a good angle.

Yes, the four legs are angled both on the X and Y plane. So they are (maybe) 25 degrees in both directions; and it’s angle iron.

The table saw was being discarded years ago on CL. Now the legs with missing plastic feet tend to make 90 degree rust marks on the concrete and thought printing feet would be a nice project.

[MarkF]

The angle in my model is currently at 10 deg.
You can change it in the 'Sketch-profile' under the 'AdditivePipe'.

I measured my Delta Contractors table saw and it was 5 deg.


Actually there are 3 sketches that need to be modified to change the skew angle to the floor:
  the 'Sketch-profile', 'Pocket', and 'Pocket001'.

[bostonman]

You're spoiling me by doing this stuff for me.

That foot looks identical to the original and looks great, however, I haven't' applied any dimensions yet to tweak it. Your honesty about a "pipe" being beyond me is appreciated. It's obvious I don't try to lie about my lack of FreeCAD skills but I'm trying to learn. Unfortunately I lack a bit of creativity, so I get stuck in a 2D world with simple padding to make it 3D.

For reference, I took measurements of my table. It's 23" tall. The front (can we call it a trapezoid?) is 18.25" across on the top and 26" across on the bottom - this would be looking at it from the front. The "trapezoid" on the side (now I'm walking to the right side of it and looking at the side view) is 20.5" on the top and 28.25" on the bottom - obviously still 23" tall.

If I did my trig correctly, that's (rounding off) 20 degrees which isn't far from my estimate of 25.

[MarkF]

I get something different.

Front:
  26 - 18.25 = 7.75  wider at bottom than top
  7.75 / 2 = 3.875  wider on each leg
  tan^-1(3.875 / 23) = 9.5 deg,  angle = arctan( opposite / adjacent)

Side:
  28.25 - 20.5 = 7.75
  7.75 / 2 = 3.875
  tan^-1(3.875 / 23) = 9.5 deg

An angle of 10 deg would seem correct.

[bostonman]

Oh….. I didn’t divide by 2. I took the inv tan of the height and delta of the width.

Thanks for the correction.

My angle was correct; if I wanted one leg vertical and the other angled.