CAD 3D PRINTING MODEL MAKING

[timbologist]

Hello all.

For a while I have been thinking about writing something about what I have been doing in the area of Cad, 3D Printing and my Modelling.

For the last 9 months I have been working with a friend in the UK I'm in Australia on the revival of the Linka System. I have been Designing, and printing on a 3D printer new components and now new Tiles,
I do the printing at home using Fused Filament Fabrication which is the printers that use the spools of filament.
for those interested below is a link for the collection of items that have been printed, to give you an idea of what is possible.
http://s1328.photobucket.com/user/timbo ... t=3&page=1

It's not an easy road, there is so much to learn, My background is I'm a tool maker with Computer Aided Design and Computer Aided Machining experience, which is subtractive manufacture, whereas 3D printing is
additive manufacture. So you need to get a different mindset about it all. How you design items, and how they go together is different. because the process just can't do certain things.

With the right setup the results are very good not quite as good as Injection moulded but not bad.

The first few post I will go into describe the current project I am working on that is a water tower, the included picture shows its current status

The ridging is not printed but made from foil, the former that was used to make the ridging was printed, This is one handy thing about the 3D printer if you want something when the shops are shut you just make it.

hope this will be of interest to someone

cheers
Tony

water tower tank

P7150011-VGA.jpg (191.81 KiB) Viewed 6412 times

water tower tank

P7150012-VGA.jpg (234.03 KiB) Viewed 6412 times

water tower tank

P7150013-VGA.jpg (233.98 KiB) Viewed 6412 times

[GeraldH]

Very nice piece of work .

[timbologist]

Here we have the second part.

I appologise if some of the posts are a bit long, but don't want to leave something out.

Most of my buildings and structures are made with Linka tiles cast in plaster, the Linka tile is based on a unit of 3.2 mm cubed that is 3.2 mm wide 8 3.2 mm deep by 3.2 mm thick.
From some cardboard models I found of water tanks that had dimensions on,there length was about 100 mm on the brickwork.
The standard large tile is 10 units high by 11 units wide and all tiles are 1 unit thick therefore 32 mm high and 35.2 mm wide. So i try to use standard tiles where possible,
in this case 3 tiles wide gives me 35.2 + 35.2 + 35.2 = 105.6 mm because of 2 overlaps in the tile fingers of 3.2 mm each I end up with 105.6 - 6.4 = 99.2 mm long over the walls the footing bricks are extra.
For the tank I wanted overhang so used 1 unit 3.2 mm each end therefore the tank would be 99.2 + 6.4 =105.6 long, and I wanted 5 panels in the length.
Now I have to relate to my printer, my printer has a 0.2 mm nozzle that's about 0.008" in the old money. So any details need to be in multiples of this width, ( you can adjust the extrusion width to be different to that of the actual nozzle in the slicing software, I will talk about that later. ). Doing this makes it easier as you don't have to try to remember what you did, as they say KISS.
I picked 1 mm width a bit over scale but still looks good as indicated by the finished photo's.
Therefore the panels would be 105.6 - 6 verticals = 99.6/5 = 19.92 mm so the panels are 19.92 inside the border and 21.92 outside, I know what your thinking is but that's longer than the original dimension, but that is sorted out when
you use Boolean addition of the panels at the centre lines of the borders as they overlap. t is easier to draw in this manor considering each part as individual the assemble it at the end.

That will do for today. be back soon with the next part





The images below are pic 1 the drawing that I based my water tower on, found by Googling
pics 2 to 5 are screen captures of the model in the Cad software

pic 1 WATER TOWER AT ROBERTSBRIDGE STATION

THE-WATER-TOWER-AT-ROBERTSBRIDGE-STATION.JPG (56.65 KiB) Viewed 6355 times

pic 2 WATER TOWER CAD DRAWING

RS-01-WATER-TANK-COMPLETE-VGA-01.jpg (405.66 KiB) Viewed 6355 times

pic 3 WATER TOWER CAD DRAWING

RS-01-WATER-TANK-COMPLETE-VGA-02.jpg (114.96 KiB) Viewed 6355 times

pic 4 WATER TOWER CAD DRAWING

RS-01-WATER-TANK-COMPLETE-VGA-03.jpg (247.66 KiB) Viewed 6355 times

pic 5 WATER TOWER CAD DRAWING

RS-01-WATER-TANK-COMPLETE-VGA-04.jpg (183.39 KiB) Viewed 6355 times

[timbologist]

Hi all hope we had a good weekend

Continuing on the width of the tank I made as 3 panels witch makes the total width 64 mm which unfortunately does not work out to be standard Linka tile Widths so will have to cut some down.
as of interest I only need to cast 2 different tiles B1-1 and B6-3 need 20 of each one.

Pic 1 shows the panel front section, I won't go into the method of drawing unless someone asks how, as there are many ways to get the same result.
Seeing I used 1 mm for the outer beading for want of a better word, I made the rest of the design the same width from being 5 times my nozzle width, the total height is 1mm.
So that's 0.5 mm of straight at the back and a 0.5 radius on the front.

Now the interesting bit what happens at the printer, Pic 2 shows roughly what you would get if I set the printer with extrusion width of 0.2 mm ( nozzle dia ) and layer heights of left 0.2 mm, middle 0.1 mm and right first 5 layers at 0.1mm and the last 5 at 0.05 mm. The blue section is the infill which can be set at various angle but is the same height as the outer layer. you can also change the extrusion width to be wider or narrower than the nozzle dia but there are limits on going narrower as its like stretching chewing gum gets to a point and you have no control as to what happens.

that shall do for now hope you are finding it interesting

PIC 1 Views of panel

RS-01-WATER-TANK-PANEL-01.jpg (138.66 KiB) Viewed 6301 times

pic 2 different layer heights

extrusions.jpg (104.46 KiB) Viewed 6301 times

[timbologist]

Hello again

Going back to part 3 as you can see the surface finish on the curved surface is not that smooth,but is easily cleaned up with fine emery.

But if you put things into perspective these irregularities in the surface finish in most cases don't mater, why you might think, well look around at the finishes on most old buildings,
not new and pristine like our injection moulded kits, so what do we do we get them and make them look old and beat-up.
If you adjust the printer by over-extrusion or under-extrusion you can get pre-distressed models straight of the printer.

One thing that I do on roof panels is I actually tilt the bed of the printer so that the top surface of the model is at right angles to the z axis giving me smoother flat surfaces. I have gone to 10 degrees so far with fantastic results.
I can probably go to 20 degrees with no problems. because my nozzle is only about 0.7 mm outside dia where most standard nozzles are 2 to 3mm dia at the tip on the od so I can get greater angles before I start digging into the previous layer.

Which brings me to a bit more detail on the printing process. From the diagrams in part 3 the layers do not just sit on top of each other as in my diagrams. But are combined into a single solid mass. as the nozzle actually squeezes the new noodle of plastic into the previous layer as the plastic is molten as it comes out the nozzle and has enough heat to melt into the layer below before it starts to cool.
And of course different plastics perform differently so printing parameters change from plastic to plastic.
that's it for now

[timbologist]

Hello all another weekend is upon us.
For today I have changed the name to reflect what happened yesterday in my world of 3D printing.

Pic 1 shows the completed job which is the pattern to make moulds for the new Linka compatible B7 mould, if you examine the photo closely you can see the lines on the base section, this is
due to under extrusion that is not enough plastic being applied. Now my printer is set up that all this is correct so what happened.

Up till now I have been using PLA on this printer with the 0.2 mm nozzle, generally PLA the hot end is set at about 190 C, using 0.4 nozzle, but I have found using 0.2 mm nozzle I have had to increase the temp to
210 C for reliable operation.

This particular job I wanted to use ABS which with a 0.4 nozzle generally 230 C is the temp used. But with the smaller nozzle I set the temp at 240 C as this was what the manufacture had run ABS at.

But alas things went pear shaped as pics 2 3 4 and 5 will show, I noticed this nasty at about 2 hours into a 5 hour print job.
So I decided to keep going and as can be seen from pic 1 the part is still usable, I will need to do a bit of repair work on the back plate surface and this is done with a mix of ABS and Acetone.
The ABS dissolves in the Acetone to make a slurry depending on the ratio you can make any type from watery to treacle or just under solid like a putty, I'll use something just over watery.
with a light sand afterwards and all will be good as new.

so in this case all was not lost. the model weighed 26 grams and the growth weighed 14 grams so the under extrusion was pretty dramatic till the growth sealed up the leak.

But this is all part of the experience in pushing past the edge.

so hope you all have a nice weekend, as I am down to the workshop now to turn up some new parts from brass and aluminum. to fix the printer actually 2 of them as my second printer has the same setup.

pic 1 the new Linka compatible B7 pattern

P7250005.JPG (271.2 KiB) Viewed 6199 times

pic 2 the growth

2014-07-25-06-VGA.jpg (248.43 KiB) Viewed 6199 times

pic 3 the crooked hot end

2014-07-25-07-VGA.jpg (244.86 KiB) Viewed 6199 times

pic 4 more uglyness

2014-07-25-10-VGA.jpg (253.52 KiB) Viewed 6199 times

pic 5 still more uglyness

2014-07-25-08-VGA.jpg (254.61 KiB) Viewed 6199 times

[timbologist]

Hi all
Well after a bit of work the printer is back up and running, had a bit of fun getting my newly designed hot end working properly, made a few errors of judgement in a few areas, but it's all a learning exercise.

I am embarking on a new project at the moment, and that is of designing and building a locomotive from scratch using available drawings and photo's to produce it.
But as a starter or lead in to this project I have several Loco bodies that I have collected, that I shall make new chassis for the first is a Triang R59 standard class 3 2-6-2T.

There are many good resources for drawings of Locomotives and rolling stock available the above I have the wonderful book Drawn and Described By Ian Beattie which is a collection from Railway Modeller.
And the Ian Allen book Historic Locomotive Drawings in 4 mm scale by F.J.Roche. And I am fortunate to have over 500 issues of Railway Modeller going back to 1954 to present.
And of course on the internet there are many good photo's of most locomotives. I just purchased a collection of about 10,000 photo's so I have available all the reference material I need.

To start this type of project I will scan the drawings I require and import them into my Cad program as a background image.

I can then scale the image to be 1:1 for 00 scale 4mm to the foot. I will use this to build a 2d wire frame in the 3 or 4 views depending on the drawings.

Even though this first attempt I only need a Chassis I will construct the whole Locomotive to get practice for the major project.

So I will be jumping from different aspects of the Cad and 3D printing aspects and Will also introduce another aspect and that is of the 3d Router Engraver which will be used to make the motion gear and maybe the wheels.

that's it for today

cheers
Tony

[timbologist]

Hi all

I have made a start on the modelling of this Loco Image 1 shows the imported bitmap of the drawing Sorry if it's hard to see but gives you an idea of what is going on. The scan was done at 1200 DPI and the file ended up about 27 megs
the Image 2 is a closer look at the bitmap.
Now Image 3 shows a different way of doing things, I have converted the original bitmap scan to a Vector drawing in Autocad dwg format and brought it in as a drawing, then the 4 views have placed together in the correct positions in the 3 views. It forms a 3D co-ordinate system to create the model, when you take orthogonal lines from the corresponding 3 views of a 2d point at the intersection of the 3 lines is the 3d position for that point. Of course there is a caveat on this method that the conversion from Bitmap to Vector can be a bit inaccurate in places depending on the quality of the original image Image 4 shows the background bitmap with the converted image over the top. But it may also be that I need to tweak the conversion settings a bit. You can also see the pixels of the bitmap so some interpretation may be required.

I will be using the bitmap image and the vector to create this model. and on the desk I have an A3 copy of the drawing for reference, in those moments of confusion.
And Image 5 shows the start of this long process.

The methods that I have outlined here are similar to those that the companies that produce the of the shelf kits and RR models use in these present times

hope you are finding it of interest

IMAGE 1 BACKGROUND BITMAP

LOCO-IN-CAD-PROGRAM-1.jpg (352.75 KiB) Viewed 6079 times

IMAGE 2 BACKGROUND BITMAP ENLARGED

LOCO-IN-CAD-PROGRAM-2.jpg (272.26 KiB) Viewed 6079 times

IMAGE 3 SECOND METHOD

LOCO-IN-CAD-PROGRAM-4.jpg (253.36 KiB) Viewed 6079 times

IMAGE 4 converted image

LOCO-IN-CAD-PROGRAM-3.jpg (367.25 KiB) Viewed 6079 times

IMAGE 5 BEGINING OF ACTUAL MODEL

LOCO-IN-CAD-PROGRAM-5.jpg (257.81 KiB) Viewed 6079 times

[Old Man Phil]

Just as a matter of interest, taper boilers were only tapered on the top and sides. The bottom was not tapered. It's easier to explain thus: Take a cone, lay it down on its side and then recut the ends vertically at 90° to the flat side.

[timbologist]

Just as a matter of interest, taper boilers were only tapered on the top and sides. The bottom was not tapered. It's easier to explain thus: Take a cone, lay it down on its side and then recut the ends vertically at 90° to the flat side.

Well thank you for that information I did not know that. You could say that puts a different slant on things

I think that will take a bit to get my head round that, I think I can extrude the end of the straight section so that it tapers out and up but flat on the bottom using a 2 rail extrusion. well next time you look I may have it corrected.

[timbologist]

Here is the repaired model with what I hope is the tapered section now in the right orientation.

I have to admit it fits within boundaries of its location better now. It was actually too wide before so thank you again for picking me up on that.
all I had to do was set 2 guide rails one for the top taper and the other along the line of the bottom of the straight section. Then extrude the circle at the end of the straight section along the 2 guide rails.
As they say easy fixed. Better to get it right now than trying to repair it later.

IMAGE 1 FIXED BOILER TAPER SECTION

LOCO-IN-CAD-PROGRAM-6.jpg (268.31 KiB) Viewed 6062 times

[Old Man Phil]

There you go, you've got it. Actually there is a mathematical formula that I use when rolling up a boiler in nickel silver or brass which I have used many times over the past 30 years. I've been tinkering with this Castle for more than two years and I doubt if I will ever get it finished due to my deteriorating health.. If anybody is interested, I have a complete white metal parts kit and tons of other bits. O gauge, of course.

Unfinished Castle.jpg (120.18 KiB) Viewed 6044 times

[timbologist]

That's a nice model Phil O would be a nice size easier to see and handle but alas I am in 00.

I have done a little bit more on the Loco But today I have something else. I am going through my scrap yard box, of all the broken models I have bought.
This ones a pair of Hornby R113 drop-side wagons that had no drop sides, but I had a third one that was complete. so removed the drop-side and measured with the verniers to produce the drawing in Image 1
Image 2 is the original drop-side and Image 3 is the printed one, with a coat of grey under coat.
You will notice that the finish is rather rough this was done by over extruding, which gives that used look, which will aid the weathering process. Saves getting the pliers and other tools to make it distressed looking.
This texture looks like it is badly corroded, I made an error with the bolts though made them 0.6mm dia should have tried 0.4mm or 0.2mm so the look to big, but when things like this rust the scale makes them grow

The cad program I use is called a surface modeller. In that you draw your model as surfaces and join those together to create you model these surfaces form a closed polysurface which is your solid object for printing. there is other software where you create your model directly as solids.

The drop-side was printed with a 0.2 mm nozzle with a layer height of 0.1 mm

when I have finished painting I shall post photo's of the finished job

Image 1 cad model of drop-side for Hornby R113

Clipboard02.jpg (155.27 KiB) Viewed 5976 times

Image 2 original Hornby R113

P8120008-800X600.jpg (280.28 KiB) Viewed 5976 times

Image 3 printed drop-side

P8120007-800X600.jpg (335.57 KiB) Viewed 5976 times

[timbologist]

Have been a bit quiet on the cad/ printing side of things, been actually working on the layout laid another 6 metres of track and added another 8 square feet of building space.

So now have 3 loops or I can run the inside as a figure 8 but one loop is inside the other, using a double slip.
I now have to climb under the layout to get to my desk so have a loop around me so I can run them any time now. Don't have to block the door ways.
My layout is located hearhttp://www.newrailwaymodellers.co.uk/Fo ... 22&t=43713

But back onto topic the last post I spoke of a wagon repair, That's now complete and I have included photo's of the finished wagons.
Must admit got a bit heavy handed on the weathering. But as you will see the rough printing adds to the distressed appearance.

hopefully get back onto the loco soon plus write some more about printing

the 2 wagons

P8180011-800.jpg (366 KiB) Viewed 5903 times

wagon 1

P8180008-800.jpg (458.55 KiB) Viewed 5903 times

wagon 1

P8180007-800.jpg (371.77 KiB) Viewed 5903 times

wagon 2

P8180010-800.jpg (439.16 KiB) Viewed 5903 times

wagon 2

P8180009-800.jpg (374.78 KiB) Viewed 5903 times

[jelly_p]

Those trucks look properly rusty and cruddy! Absolutely fantastic weathering.