3 D Printing 16mm models

[Trevor Thompson]

So back to the Kerr Stuart 4415.

Here is the inner chassis assembled and tested in three views. The sides were printed axle hole down on the printer bed. The 3mm holes are printed in site but only go as deep as the top of the Hexagonal part. The frame stretchers printed as separate items. The wheels as 6 wheels at once with the inside face on the printer bed.

The stainless axles were faced and centred on the lathe, partly so I could press the wheels on to the axles accurately using a home made fixed centre. (A bit of pointed rod) to engage into the centre formed on the axle. The wheels were made with 2.8mm diameter holes printed in them, and were just carefully mounted in the 3 jaw chuck and drilled with a 2.8mm drill to clean them up.

They have gone on OK - seem to run true, and a tight enough fit on the axle to hold the back to back measurement.

The gearbox was printed in 3 parts in white ABS. Why white - just what I could get during lockdown. The motor is held into the motor locating tube with a couple of dabs of Gorilla glue - the polyurethane one which foams.The holes for the gearbox axles are printed 2mm diameter and cleaned up with a 2 mm drill by hand. The three parts were held in position relative to each other by the 2mm axles - it seemed to work because the gears run easily. The gears were bought, I cant print such small detail (I have tried). Chain and sprockets bought - and a press fit onto 3mm axles.

Incidentally I tried printing the gearbox in one unit but couldn't get it to work properly. So three parts the motor tube and the bit which has a gear each side of it - one piece printed with he gear end of the motor on the bed. Second part the section to the right in the second photo, including the spacers, flat end down. Third bit the piece which the ends of the shafts go into on the left of the second photo. The three parts just bonded together using Acetone.

I am going to make a second gearbox with better quality gears, and 3mm intermediate and final drive shafts. Not that this doesn't work - but it seems a better way to do it.

Incidentally I like to print wheels and other bits which might have been black on the real thing in black filament. When the paint comes off the tread is still black (if I actually paint them).

So to the photos:

IMG_0683.jpg (322.18 KiB) Viewed 4988 times

IMG_0682.jpg (312.37 KiB) Viewed 4988 times

IMG_0684.jpg (243.32 KiB) Viewed 4988 times

The footplate, and chassis proper will be assembled later. Last frame side is on the printer now.

[ge_rik]

I saw that series of YT videos which suggested drawing something in 2D first and then extruding - it also stressed the importance of using a 3 button scroll wheel mouse and also having to unlearn click, hold and drag. That was probably the most useful SU video I watched. Unfortunately, despite Philip's best efforts and hours of reading and video watching, I've still not discovered how to avoid making errors which make my drawings unprintable - which is why I abandoned SU in frustration. Nothing worse than spending a few hours making a drawing only to discover it won't print and then spending many more hours trying to fix it - all to no avail.

I suppose, I need an Idiot's Guide to avoiding reversed faces and non manifold shapes.

Rik

[Trevor Thompson]

Yes I also saw that video on having a 3 button mouse. It was quite negative about the apple mouse. I use a iMac so it was a bit of a blow until I realised that what I had was a three button mouse!

I still use the apple mouse.

Reversed faces are something I have encountered and "inspector" usually sorts that out.

The other message I have never encountered.

[Trevor Thompson]

The chassis for 4415 is now complete. Printed and assembled:

IMG_0692.jpg (256.76 KiB) Viewed 4970 times

and the underside:

IMG_0693.jpg (376.62 KiB) Viewed 4970 times

and the spectacle plate is in the printer:

IMG_0696.jpg (312.21 KiB) Viewed 4970 times

Trevor

[philipy]

I've tried using the track pad on my laptop and its next to impossible, practically speaking. Apart from being very awkward to do, I get my head lifted off my shoulders by SWMBO because of the constant clicking of the trackpad buttons! She does have a bluetooth apple mouse which works fine and is silent, but is unfortunately only 2-button so that put me back to square one.However, a 3button mouse with a small USB dongle thingy solves both problems.

I can't say that I've ever seen Rik's 'non manifold shape' message either.

[Trevor Thompson]

I am now thinking about printing the bodywork, and the front and top of the engine casing in particular.

I find that I get the best results printing things flat on the bed with the "good" face uppermost. The rivets seem to print out cleanly.

This part is a bit more difficult let me show what I have been doing. Of course this relates to the slicing software that came with my printer, but I think that it still might be worth while. So:

My first thought was to try to print it out in one piece. This is an image of the slicer software with the component in the print area:

IMG_0698.jpg (316.16 KiB) Viewed 4272 times

and then generate the support structure. I prefer the matrix supports rather than a tree like support structure (which is my other option):

IMG_0699.jpg (324.85 KiB) Viewed 4272 times

The issue with this will be apparent in the next to photos. The print file is created and it displays what it intends to print including a raft to attach it to the printer bedNote that it is going to take about 8 hours to print:

IMG_0700.jpg (244.69 KiB) Viewed 4272 times

The time to print is the first issue. The second problem is that long overhanging slope. Look what happens half way through the print:

IMG_0701.jpg (283.41 KiB) Viewed 4272 times

That slope is likely to break off. Experience suggests that it will! I could manually add supports under the sloping section. However I am not convinced that the detail, rivets in particular, will print as well as they can.

So another approach. Im going to cut it into three parts and print each flat on the bed. This slicing software allows me to position a cutting plane where I want to cut it and then cut:

IMG_0702.jpg (220.35 KiB) Viewed 4272 times

I have cut it twice and laid all three parts flat on the bed. Added the supports. This time the supports are much shorter. Less time. Less filament. Less electricity:

IMG_0706.jpg (262.99 KiB) Viewed 4272 times

and finally create the printer instructions:

IMG_0707.jpg (269.52 KiB) Viewed 4272 times

This time it will take about 4 hours to print, and the detail should come out better. I will share the result when its printed.

Trevor

[Trevor Thompson]

The last photo on the last but one post showed the spectacle plate printing.

Let me share the results with you:

Firstly the print was mostly successful. The top surface is really good:

IMG_0708.jpg (310.75 KiB) Viewed 4254 times

The only minor issue is that 3 rivets are not there. You might just be able to see them on the right under the window as you look at it. That was caused by the three rivets sitting in space a short distance away from the plate. I went back into Sketchup and lowered them onto the surface. However I am not satisfied with the reverse face:

IMG_0720.jpg (305.54 KiB) Viewed 4254 times

You should be able to see the lines of filament passing diagonally between the spectacle plates. many of them are loose - not attached to the rest of the model. This spectacle plate is also a bit thin. Too flimsy. It is probably the result of tracing off the drawing, which encourages me to make things to scale thicknesses. I had the same issue with the footplate and that was certainly caused by tracing it.

Let me take this further. Before I got around to printing the top of the engine box I had the right hand side of that box printing. I had also changed from black filament to white filament thinking it would be easier to photograph! So look at this:

IMG_0709.jpg (235.59 KiB) Viewed 4254 times

I took it off the bed - and then realised I should have a photo of it. So I have just placed it as it was. The jar is just holding it in the correct place for the photo. You can see how the whole of the edge has come free. I stopped the print.

Back to black filament. Print again:

IMG_0712.jpg (375.26 KiB) Viewed 4254 times

Perhaps you can see the diagonal lines of filament some of which are floating free of the surface. More importantly the way the light falls should show that that surface is not flat. In fact the short edge is actually thinner in the middle than at the ends. The middle came free from the raft while printing. So the point of all that is that there is a problem of lack of adhesion between the raft and the model.

The usual recommendation I have found on uTube to solve adhesion problems is to increase the extruder temperature and/or to lower the speed that the extruder moves. So I raised the temperature to the maximum for this printer, 235 C, and used High definition rather than standard. That lowers the speed to 40mm min rather than 60mm min. It also makes each layer of print thinner. IE more passes for a mm of height. The result was much improved. No loose filament on the reverse side. Flat and even thickness all over:

IMG_0718.jpg (297.09 KiB) Viewed 4254 times

So back to the white filament:

IMG_0716.jpg (243.42 KiB) Viewed 4254 times

I left it printing overnight I found the printer door open, a thin layer of printed engine casing on the floor and this pile of filament.
The raft is there but the engine case must have come totally free, caught on the extruder and been shot through the door.

I was tempted to just put the black filament back in. The issue was adhesion between raft and print - so try removing the raft. Tried with both balck and white filament and nothing stuck so abandoned that.

So in conclusion:

The smaller items, wheels (even 6 at a time), frames and buffer beams all print well on standard settings. 230 C and 60mm min. Good clean edges on top and sides, just the diagonal filaments on the bottom But they are firmly attached.

The large thin items don't print so well on standard settings. Hotter and slower solves the problem.

The white filament? I can't print it reliably. The clue is probably on the side of the box. The black filament says print between 220 and 250 C. The white says print between 230 and 250 C.

Back to printing the engine casing top (in black!).

[philipy]

The white filament? I can't print it reliably. The clue is probably on the side of the box. The black filament says print between 220 and 250 C. The white says print between 230 and 250 C.

I mostly use PLA rather than ABS, but with adhesion problems I tried all the usual remedies, Prit Stik, blue tape, etc, but in the end I solved it by increasing the bed temperature for the first layer to 60deg and then dropping it to 40 for the rest of the print. Don't know if that principle would help with your problem ( obviously starting with higher bed temp for ABS anyway).

I also used a raft in my early printing days, but apart from adding ages to the print time and wasting huge amounts of filament I always had problems getting the print to separate from the raft. I now use a brim instead and have no probs with adhesion, plus the base always has the flat surface finish of the bed at the bottom.

[Trevor Thompson]

Interesting.

Unfortunately 100 C is as high as the bed temperature goes.

I will try a brim instead.

I tried it without a brim or a raft and nothing stuck - black or white!

I don't have problems separating the raft - rather the opposite at the moment!

I suspect that my printer is intended to use a raft - as instead of a levelling system.

Trevor

[Trevor Thompson]

Tried printing with a Brim and the filament didn't stick to the bed. So I will continue to print with a raft for now.

I have good quality prints for both sides of the engine cover:

IMG_0721.jpg (352.87 KiB) Viewed 4226 times

and a finished print of the engine cover lid/front. While most of that is good I am not satisfied with the sloping face. lots of grooves across it caused by the edges of the different layers of filament being exposed. Its OK but could be better. As for fit onto the sides it seems to have a different shape. The sloping section is too long and the horizontal and vertical faces too short. I shouldn't be surprised really - making it from 3 parts is not ideal:

IMG_0722.jpg (349.45 KiB) Viewed 4226 times

So I will try that again but on its side this time. I have removed the exhausts to print separately:

IMG_0723.jpg (226.34 KiB) Viewed 4226 times

Trevor

[Andrew]

Hello!

I see I'm a little late to this thread, but it's fascinating stuff, thank you! And those Festiniog wagons and carriages ae looking very good indeed.

Coincidentally, I'm also planning to have a go at producing some CAD drawings for a model of a KS diesel (the smaller 30HP version), although I'll be aiming for a 2D kit of parts to cut from MDF - or maybe acrylic sheet. 3D stuff may follow, so I'll be re-reading this thread more closely - thanks for sharing your experience so far...

Cheers,

Andrew.

[Peter Butler]

I have said elsewhere on the Forum that 3D printing is wonderful equipment for model makers and I have built 3D printed kits myself. I notice that professional kit makers have altered the contents of their kits to now include HIP sheet where plain, flat surfaces are required, cab sheets, footplate etc. as they require no finishing to make them presentable on the model. In the process shown in this thread many of the parts could be similarly designed but with only small additional printed detail added. Items like rivet detail are impossible to work around when cleaning up a 3D printed surface, so seem to me to be time and material wasted.
I remain unconvinced, so far, that the entire model should be printed, but wait to see the finished model.

[Trevor Thompson]

Peter

You are of course entitled to your opinion.

However the whole point of this is that most of the time I am not doing ANY cleaning up of surfaces. The models I have shown have just been printed and painted.

Trevor

[Trevor Thompson]

Andrew

Than you for your encouragement!

I will be interested to see how you get on reported on this Forum. The only comment I would make is that MDF seems a strange material to make models from ( know you can buy kets made from it). Acrylic I understand. The problem to me is that MDF is Hydroscopic. It will almost inevitably absorb water from the Atmosphere.

Trevor

[Andrew]

Hi Trevor,

Yes, I'll almost certainly write the project up here - I generally do!

My thinking (such as it was...) behind potentially opting for MDF was that a) I've seen it done elsewhere (http://www.alspcs.com/Kerr_Stuart.html), b) I've recently had experience of working with MDF and have been pleasantly surprised, whereas acrylic is an unknown (is it brittle, can you glue it like plasticard??) , and c) there are "tame" 16mm kit suppliers who understand our needs and will do small jobs on demand, but I don't know if they'll work with acrylic...

But... I agree about the potential water absorption (I'd like the finished loco to be able to go out in the rain!) and don't entirely like the idea of wood representing metal, so I may well change my mind. Happily, I'm sure it will be many months before I progress far enough for a decision to be required!

Cheers,

Andrew.

[Trevor Thompson]

I have put together a guide to how I go about making the models that I print in Sketchup.

It is intended to follow on throw the beginners guides in the reference section.

This is the link to download it. https://www.dropbox.com/s/9etz7iqswzqzh ... e.pdf?dl=0

If you are interested in this topic have a look at it. Let me know if you find it helpful or not. If it is helpful I can take it further. Perhaps how I made the gearbox for 4415, how I draw wheels with tapered flanges, how I made a wagon turntable, how I made the spectacle plate for 4415. However it represents significant effort to do it - and I will only do so if I am asked to do it!

Trevor

[Trevor Thompson]

The engine box lid and top has now printed. Here is the result straight out of the printer:

IMG_0759.jpg (344.37 KiB) Viewed 4524 times

This has worked as well as I had hoped it would. The surface is very smooth - and ready to just paint. However when I print things on their side like this (roofs are normally printed this way around) there is a tendency for the layers of the model to part. It isn't visible in this photo but it started to show as I handled it. The solution is to bruch acetone over both sides of the component. It makes the surface unevenly shiny - but doesn't otherwise effect the surface finish like it would do if you did the same to an Aifix model with solvent glue. the square lumps on the model are the supports that th eprinter uses to support overhangs while it prints them> They just fall off when you rub a finger over them. They are not attached at the side just at top and bottom.

So the same part with the supports removed, and fitted into position on the chassis:

IMG_0766.jpg (336.7 KiB) Viewed 4524 times

I have yet to work out how I will fix the engine cover to the chassis. Perhaps 2 frames glued inside the engine box and screws holding it down from underneath. I could just glue it on after fitting the batteries. Decisions regarding batteries and radio control have to be taken first:

IMG_0767.jpg (303.66 KiB) Viewed 4524 times

It is beginning to look like a loco (but it is a pretty ungainly looking thing!)

IMG_0769.jpg (363.82 KiB) Viewed 4524 times

Have to print the redesigned gearbox next.

Trevor

[Trevor Thompson]

I thought you might like to see what the engine casing looks like when ti has some paint on it. So a thin coat of Valejo acrylic polyurethane airbrush primer, followed by a single coat of thinned Humbrol matt Eggshell blue which I found in the box of paints:

IMG_0771.jpg (340.83 KiB) Viewed 4506 times

And showing more of the front:

IMG_0772.jpg (325.04 KiB) Viewed 4506 times

The coach in the background has had a very thin coat of the same primer. The locomotive seems to have been painted in a light green, and had the makers name on it in large white letters. So there are a few more coats of paint to go yet - and that can wait until the model is nearly complete!

[Trevor Thompson]

Now for something slightly different.

I placed this thread under the general heading because I want to focus on how 3 D printing can affect all parts of garden railways, rather than on a particular model. As some of you will be aware I have been extending my railway during lockdown, and the extension features a station area modelled (loosely?) on the track layout at Tan y Bwlch (That Festiniog Railway again) during the 1860's. ( Perhaps more of that in another thread later). That layout features 3 wagon turntables - which seem to have been prominent features of all British railways until comparatively recently.

So I have printed three wagon turntables. Firstly one I made earlier which has been in situ for a couple of months:

IMG_0757.jpg (353.21 KiB) Viewed 4474 times

It seems to have acquired some cement dust during ballast laying - and Im waiting for the rain to stop so I can clear the trimmings off the track!

The turntable consists of three components:

The top deck which has the grooves for the wheels in it:

IMG_0774.jpg (294.59 KiB) Viewed 4474 times

The bottom part of the top deck which has a rectangular block sticking up 0.5mm. This engages with slots in the lower deck to make it locate in four places to align the tracks. The top deck is in two parts so that the best quality surface is available for locating things accurately. This is the lower part of the top deck upside down:

IMG_0776.jpg (303.28 KiB) Viewed 4474 times

Finally the lower deck which has the slots in it:

IMG_0775.jpg (305.95 KiB) Viewed 4474 times

Now back to assembling the gearbox for 4415.

Trevor

[Trevor Thompson]

The gearbox.

It consists of three parts as shown in this view:

IMG_0777.jpg (347.47 KiB) Viewed 4448 times

They were printed with the side which is on the right of each component flat on th ebed. Printing small parts like this at 5 degrees higher than default makes the face nearest the bed solid and clean. It does make it harder to separate from the raft. The holes for the shafts were opened up with a 3mm drill and then reamed 3mm. Sounds wrong but the plastic cuts undersize. To assemble 3mm shafts were temporarily inserted to align the holes and the end cover bonded to the motor housing:

IMG_0778.jpg (296.18 KiB) Viewed 4448 times

When that was solid the third component was added. It did need a bit of filing on its mating face because it was an interference fit! It was gently held in the vice and bonded into place:

IMG_0779.jpg (351.2 KiB) Viewed 4448 times

Note that this third part is thicker because it locates the chain wheel and gear. It is also glued on so as the whole assembly just fits between the frames. That means it can be glued in with my usual acetone. The gears have now been assembled onto their shafts:

IMG_0782.jpg (338.39 KiB) Viewed 4448 times

The red gears and sprocket are an interference fit on the 3mm shaft, and the grey one a free fit. That meant that the upper shaft was just pressed into the tight hole. However the lower shaft required its holes drilling out to 3.2mm. I said it cuts undersize! The motor has been fitted and lightly bonded into place:

IMG_0783.jpg (356.35 KiB) Viewed 4448 times

Finally the assembly was fitted into the frames on 4415. It needed the chain shortening by 2 links, but that means there is less slack in the chain:

IMG_0784.jpg (294.09 KiB) Viewed 4448 times

Trevor