Shoelast

Post here for help on using FreeCAD's graphical user interface (GUI).
Forum rules
and Helpful information
IMPORTANT: Please click here and read this first, before asking for help

Also, be nice to others! Read the FreeCAD code of conduct!
Leatherman
Posts: 155
Joined: Thu Feb 02, 2017 2:49 pm

Re: Shoelast

Post by Leatherman »

Hello,
and thanks to Microelly2. You're doing an amazing job for me. Now what you've done I can follow up. ;) Now when we come to the b-splines, we can do corrections and they are necessary for creating a new shoe last. I think, that it would be the easiest way to succeed if we can change the metric of the b-spline. Is it possible to change the girth of a b-spline like in the image. Let's say the girth is x and we want x + 5mm is that possible? For the first step I would be happy to create a basic last and could change the metrics according to the most important fittings (see image below), which could be retrieved as you've suggested. Shoe length (Size) could be adjusted by scaling, that's no problem. In case a parametric change for the b-spline is not possible than what? But I think it's possible, could somebody help out? How to measure the girth of the b-spline and how to change the volume? Great Forum here, really appreciate it.
thanks a lot,
Andre
Attachments
shoelast.jpg
shoelast.jpg (18.39 KiB) Viewed 2417 times
B-Spline.jpg
B-Spline.jpg (14.63 KiB) Viewed 2417 times
User avatar
microelly2
Veteran
Posts: 4688
Joined: Tue Nov 12, 2013 4:06 pm
Contact:

Re: Shoelast

Post by microelly2 »

Interesting non standard project.
I first will have a look at my shoes to understand the geometry behind,

but what you want will be possible.
I have created the FreeCAD worm some weeks ago
Image
EDIT: https://youtu.be/sWlaE4i2MAQ
And a shoe is a special form of a worm(spat) with a sole on one end.
Leatherman
Posts: 155
Joined: Thu Feb 02, 2017 2:49 pm

Re: Shoelast

Post by Leatherman »

Hello,
I need some time to understand your video, it's not so easy, but it shows that it should be possible. Let me give you a few small hints regarding a shoe last, because from the shoe you can't make out much. The shoe last has to present your foot. It's not identical, because it has to be longer, the heel has to be raised and it gets some final design, but certain measurements have to be identical with your foot otherwise the shoe can not fit or better saying the foot doesn't fit the shoe.
The foremost rule is always that the inside- and outside joints of your foot have to be identical to the insole shape j1+j2 of the below image. We are talking here about the plantar surface of your foot. Now on top of that, in the middle line (cortex of xy) is joint J (which is not the middle of j1-j2, outside is always wider than inside). This girth measurement of J1+j2+j, the position of the j's and the length (shoe size) and height (heel) are the basics of building any shoe last. With these positions fixed anything can be build, from a soccer boots to high-heels, it's parametric. For example find the Instep point from the previous picture for Shoe Size 6. Standard length (264 mm) / 4 = 66mm measure up from point J = I and for the girth of I take the girth of j’s, example 220 mm + 10 mm for the Instep = 230 mm. Instep completed.
That’s why I was asking for changing the girth of the b-spline. The position of any point I can locate with placement. Is it possible to determine the length (girth) of a b-spline and then change its length?
Hope that helps and thanks
Andre
Attachments
Foot5.png
Foot5.png (84.7 KiB) Viewed 2391 times
User avatar
microelly2
Veteran
Posts: 4688
Joined: Tue Nov 12, 2013 4:06 pm
Contact:

Re: Shoelast

Post by microelly2 »

Given some control points of a bspline the length can be calculated and displayed.
the other way from the lenght to the curve is ambiguous as you can drag the skin of the shoe.
So you can change the form of the bpsline until you get the right.
If there are more restrictions we can "scale" the inner control points until the wished length is reached. The outer controlpoints are defined and fix data (j1,j2) .
This should work like blowing a ballon.

Is the girth/joint perpendicular over J from picture 2 or is there an angle like in your 1st picture?
Leatherman
Posts: 155
Joined: Thu Feb 02, 2017 2:49 pm

Re: Shoelast

Post by Leatherman »

Hello,
microelly2 wrote: Given some control points of a bspline the length can be calculated and displayed.
This would be very helpful. Any idea how to do it?
microelly2 wrote:the other way from the lenght to the curve is ambiguous as you can drag the skin of the shoe.
So you can change the form of the bpsline until you get the right.

Yes true, in a real model we build up or reduce and measure with a measurement tape, similar it has to done here as well. It would not be such a problem for just a dozen points, which should be sufficient for a good start.
microelly2 wrote:If there are more restrictions we can "scale" the inner control points until the wished length is reached. The outer controlpoints are defined and fix data (j1,j2)

Yes and don't forget the many points of "C" which following the side-profile pattern of the last and the C1+C2 respectively according to the insole pattern. Their positions can be located as well. As I have seen so far, the loft tool is following a very nice flow, which should do great for shaping the shoe last. Usually for real work on a model I divide the "C" sections equally in 11 +1 parts over the model, where is 0 at x=0 at heel and 11 the toe of the foot - not the last. I think the loft will not have a problem with this. Will try and post.
microelly2 wrote: This should work like blowing a balloon.
Exactly. Look every type of footwear requires certain type of last categories by its shape and form. You can't compare a sandal last with a boot last, even though certain points like "J" are identical in both of them, so the b-spline may be of the same length, its form has to differ according last type and requirement. Finally there have to be certain groups.
microelly2 wrote:Is the girth/joint perpendicular over J from picture 2 or is there an angle like in your 1st picture?
No, it's perpendicular - you need an angle to locate the point J.
Attachments
B-Spline.jpg
B-Spline.jpg (42.52 KiB) Viewed 2378 times
Leatherman
Posts: 155
Joined: Thu Feb 02, 2017 2:49 pm

Re: Shoelast

Post by Leatherman »

Hello,
after working with b-splines as per advice I've progressed a little bit. My model is not solid, so I have problems with the back part of the last, so I left it for today. Otherwise the loft works as a copy, I didn't change the spline yet.
Andre
Attachments
lastloft1.png
lastloft1.png (24.22 KiB) Viewed 2364 times
lastloft.png
lastloft.png (73.19 KiB) Viewed 2364 times
User avatar
microelly2
Veteran
Posts: 4688
Joined: Tue Nov 12, 2013 4:06 pm
Contact:

Re: Shoelast

Post by microelly2 »

Getting a solid should not be a big problem.
As demonstrated on my violin example you can cut the top of the loft with a simple cube.
I have played a bit with the curve length function.
It is possible to calculate for a given BSpline curve and its length a visualization of the area where a point of the curve must be placed to get exactly the wished length
bp_363.png
bp_363.png (9.19 KiB) Viewed 2355 times
in the pic there are 3 polygons for different length. If you put one of the base points of the bspline to a vertex of these polygons then the Bspline interpolation has exactly the right length.
this can become as intuitive as pressing the thumb onto the leather to slip the form away.
It s in interesting task like bending a cooked spaghetti or a rope: Length preserving design.
At the moment I work with polygon approximation but smooth curves will be possible too.
Leatherman
Posts: 155
Joined: Thu Feb 02, 2017 2:49 pm

Re: Shoelast

Post by Leatherman »

Hello,
I'm not sure if I get you 100%, but it brings me more to the direction to work with patterns. Let's say the frame is a pattern for a certain cortex position or if you like the J1-J-J2 position on the last and the spline is the actual shape. The frame and all points are defined in a Spreadsheet.
In my image below the new points could be an predefined option (calculated with length and volume) in a Spreadsheet. What do you think?
regards
Andre
Attachments
B-Splinepattern.png
B-Splinepattern.png (8.34 KiB) Viewed 2351 times
User avatar
microelly2
Veteran
Posts: 4688
Joined: Tue Nov 12, 2013 4:06 pm
Contact:

Re: Shoelast

Post by microelly2 »

Leatherman wrote:Hello,
I'm not sure if I get you 100%, but it brings me more to the direction to work with patterns. Let's say the frame is a pattern for a certain cortex position or if you like the J1-J-J2 position on the last and the spline is the actual shape. The frame and all points are defined in a Spreadsheet.
In my image below the new points could be an predefined option (calculated with length and volume) in a Spreadsheet. What do you think?
regards
Andre
Yes of course, there must be more comfort
There are two questions
1. what I can calculate to support the spline design
2. what is a helpful presentation of the data
Today I only wanted to get an answer to the first questions and at the moment I can calculate all possible positions for a given length with fixed endpoints (j1, j2).
So I have learned my lesson for today :D
Next step can be done next week.
here the first result with some explanations what I can do: https://youtu.be/526SwK8JZnw
User avatar
microelly2
Veteran
Posts: 4688
Joined: Tue Nov 12, 2013 4:06 pm
Contact:

Re: Shoelast

Post by microelly2 »

How many points do you need to define a single frame/rib
2 for each endpoint j1, j2 (one for position, one for tangent direction)
2 for the top j
and one more on each side j1-j and j -j2
these are 8 points

How many ribs you need for the shoe design? in your picture there are about 10.
Lets add 5 more for the back.
Then there are 8 * 15 = 120 points.
Post Reply