3D Printing items - Help please

[sml1983]

So I made a topic a few weeks back (http://www.rmweb.co.uk/community/index.php?/topic/77386-prototype-ready-for-manufacture/)

 and decided last night that the title was irrelevant so have begun a new one. 

 

I am completely new to the process of 3D printing, in fact all processes of manufacture are beyond my grasp, and so haven't a clue how to make items cost effective or what the next steps in processes are.

 

I intend to eventually get prototypical carriages that I can use on a p4 gauge layout.  These may eventually possibly become available for others to buy if costs do not make this unfeasible.

[sml1983]

So now i'm to the stage where I have the single bolster version of the BR1 bogie finished with all normals facing the correct way:

 

What I would like to know is how do I avoid costs of having tiny parts made as the axleboxes, tie guards and central brake rod are separate parts to allow fitting of wheels and sprung suspension.

 

Any help greatly appreciated.

[Revolution Mike]

Provided all your parts are butted up against each other then Shapeways will probably print all as one anyway (better is often to make things slightly overlap).  Just upload the model to Shapeways and see what the uploaded file looks like (it will be obvious if pieces are missing!).  It will then need to pass Shapeways print checks - I assume you followed the design guidelines for minimum thicknesses etc for the material you want to use.

 

BTW you didn't need to check each normal individually - Blender allows you to select all and re-calculate the normals on the outside in bulk.

[Marbelup]

Very nice 3D model, but my first impression is that many of the parts are just too fine for 3D printing, assuming you are thinking of using a "consumer level" 3D printing bureau.  For example, the brake and suspension rods, horn blocks, flanges on the channels, etc. look very fine.

 

As an indication, I have been using the "Prime Grey" material from i.Materialise and the minimum thickness for small detail items, to ensure reliable printing, is 0.5 mm (since they changed the print specs in June 2013, 0.3 mm previously) and "structural" parts should be at least 1 mm thick.  So, some compromises are necessary to produce a practical 3D printed item.  Whether the necessary compromises destroy the prototype fidelity is for you to decide.  

 

I haven't attempted to make working parts such as horn blocks and axleboxes, but I have done locating devices such as pins and sockets and it is difficult to get these to fit with great precision.  It would probably be easier with your own 3D printer, as you could expect consistent results but the commercial outfits have many printers, some with varying specs for the same material, so the finished product can vary somewhat from one order to the next.  

 

Also, the process of "slicing" the desired shape to form the layers for 3D printing can vary as there is no absolute reference level, so surfaces which are horizontal (when printed) can vary in relative height by an amount similar to the layer thickness of the printer.  For example, to represent an overlay (such as a door) I get most reliable results from making the overlay at least the thickness of two print layers (which is 0.1 - 0.125 mm per layer for Prime Gray). An overlay just one layer thick may disappear completely.

 

Based on my experience (1 year of 3D printing), I suggest it would be quite difficult to design a bogie with working axleboxes in 4 mm scale, so maybe you should explore other methods of springing if that is what you want.  It is also hard to design a bogie which is sufficiently flexible to allow the wheelsets to be inserted, which doesn't tend to "spread" outwards from the weight of the vehicle on the pinpoint bearings.  I have have found that the Prime Gray material is prone to "spreading" over a period of weeks when subject to a force.  The other option is to make the bogie as rigid as possible, to avoid spreading, but in two pieces to allow the wheels to be inserted.  I have made some successful freight bogies using this principle.  A third option, would be to slot the bearing holes to that the wheels (with bearings??) could be inserted from below into the rigid bogie, with some other mechanism to secure the wheels in place.  However, this would most likely compromise the appearance, as the bar which secures the axlebox in real life is in the way of inserting the axles from below.

 

Here is a pic of my 2-piece freight bogie.  I deliberately made the sideframes quite chunky and added extra thickness near the bolster to minimise spreading, and it is working Ok so far (about 6 weeks).  The bearing holes take 2 mm brass bearings, and the two halves (which are identical) are held together with M1.4 screws through the holes in the bolster.

 

 

I did make a 2-piece 4-wheeled wagon with a working compensated chassis in 16.5 mm gauge using a "living hinge" in Prime Gray, but the "spreading" issue was a problem.  Mainly for that reason, I have done back to using conventional brass W-irons in my latest version.

 

So, getting back to your specific question, the cost of printing the parts required is an issue, but the practicality of making the thing work is probably going to be the first priority.   

[sml1983]

Provided all your parts are butted up against each other then Shapeways will probably print all as one anyway (better is often to make things slightly overlap).  Just upload the model to Shapeways and see what the uploaded file looks like (it will be obvious if pieces are missing!).  It will then need to pass Shapeways print checks - I assume you followed the design guidelines for minimum thicknesses etc for the material you want to use.

 

BTW you didn't need to check each normal individually - Blender allows you to select all and re-calculate the normals on the outside in bulk.

 

Sorry didn't make it very clear.  I knew about the recalculate button but until I made the model full size and not 1:76.2 the normals were just so close to each other that blender couldn't automatically work them out.  It was quite interesting asking it to recalculate and seeing how many different ways it made the normals point lol.

Very nice 3D model, but my first impression is that many of the parts are just too fine for 3D printing, assuming you are thinking of using a "consumer level" 3D printing bureau.  For example, the brake and suspension rods, horn blocks, flanges on the channels, etc. look very fine.

 

As an indication, I have been using the "Prime Grey" material from i.Materialise and the minimum thickness for small detail items, to ensure reliable printing, is 0.5 mm (since they changed the print specs in June 2013, 0.3 mm previously) and "structural" parts should be at least 1 mm thick.  So, some compromises are necessary to produce a practical 3D printed item.  Whether the necessary compromises destroy the prototype fidelity is for you to decide.  

 

I haven't attempted to make working parts such as horn blocks and axleboxes, but I have done locating devices such as pins and sockets and it is difficult to get these to fit with great precision.  It would probably be easier with your own 3D printer, as you could expect consistent results but the commercial outfits have many printers, some with varying specs for the same material, so the finished product can vary somewhat from one order to the next.  

 

Also, the process of "slicing" the desired shape to form the layers for 3D printing can vary as there is no absolute reference level, so surfaces which are horizontal (when printed) can vary in relative height by an amount similar to the layer thickness of the printer.  For example, to represent an overlay (such as a door) I get most reliable results from making the overlay at least the thickness of two print layers (which is 0.1 - 0.125 mm per layer for Prime Gray). An overlay just one layer thick may disappear completely.

 

Based on my experience (1 year of 3D printing), I suggest it would be quite difficult to design a bogie with working axleboxes in 4 mm scale, so maybe you should explore other methods of springing if that is what you want.  It is also hard to design a bogie which is sufficiently flexible to allow the wheelsets to be inserted, which doesn't tend to "spread" outwards from the weight of the vehicle on the pinpoint bearings.  I have have found that the Prime Gray material is prone to "spreading" over a period of weeks when subject to a force.  The other option is to make the bogie as rigid as possible, to avoid spreading, but in two pieces to allow the wheels to be inserted.  I have made some successful freight bogies using this principle.  A third option, would be to slot the bearing holes to that the wheels (with bearings??) could be inserted from below into the rigid bogie, with some other mechanism to secure the wheels in place.  However, this would most likely compromise the appearance, as the bar which secures the axlebox in real life is in the way of inserting the axles from below.

 

Here is a pic of my 2-piece freight bogie.  I deliberately made the sideframes quite chunky and added extra thickness near the bolster to minimise spreading, and it is working Ok so far (about 6 weeks).  The bearing holes take 2 mm brass bearings, and the two halves (which are identical) are held together with M1.4 screws through the holes in the bolster.

 

 

I did make a 2-piece 4-wheeled wagon with a working compensated chassis in 16.5 mm gauge using a "living hinge" in Prime Gray, but the "spreading" issue was a problem.  Mainly for that reason, I have done back to using conventional brass W-irons in my latest version.

 

So, getting back to your specific question, the cost of printing the parts required is an issue, but the practicality of making the thing work is probably going to be the first priority.   

The way i'd planned on getting the wheels in place is as follows:

The Axle tie is a separate piece allowing the wheels to be inserted into two axleboxes that then slide into position.  The Axle tie is then replaced stopping the axleboxes from coming away.  Guitar wire is then inserted into each axlebox to allow for springing.

 

I wasn't sure what the tolerances for 3D printing were. Just got this though:

Really happy with the quote.  Does this mean that the parts are large enough then?

 

Thank you for all your help.  Like the idea of a 2 part model.  What material do you use, or is 'Prime Gray' the material?

[brack]

The initial printability check means precisely nothing I'm afraid - shapeways will still reject it if the bloke operating the machine that day feels like it (or if it doesn't meet their criteria ) I suspect that you'll be very disappointe dby the finish/detail you gte from white strong/flexible, as it gives a very rough surface which is hard to smooth, and will not reproduce all the fine detail you put in. I won't use anything less than FUD for body parts, with FD for chassis (I find WSF doesn't give the required accuracy for chassis construction). I suspect your model will be rejected, but if you contact shapeways customer services they can eb very helpful if you explain what you're trying to do (eg. ask for advice on how to make it 'pass'). With FUD I find they're often rejecting 0.4mm thick bits these days, even though their material specs say 0.3 is possible. A year or two ago they'd print almost anything (0.17mm thick handwheels for example), and do a good job usually. These days the prints tend to be somewhat crisper, but they won't try and do anything near the outside of their comfort zone. A smaller or more expensive print shop might give better results if its accuracy and detail you're after, but at a price (nobody can really touch shapeways for price/quality at the moment, though i.materialise could come close but have silly ideas about sprues...)

[Marbelup]

Yes, "Prime Gray" is a material from i.Materialise.  Shapeways do not have an equivalent.

 

I agree with Brack's comments that you would be dissappointed with the appearance of WSF from Shapeways or i.Materialise equivalent (Polyamide).  The tolerances required for moving parts would certainly not be achievable with WSF/Polyamide.

 

I have attached a photo of my freight bogie in Prime Gray (left) and Polyamide (right).  You can see the stepping due to the layers in Prime Gray but the surface quality is much better than the rough finish of Polyamide.  I have requested a different print orientation for my next order (in Prime Gray), which should minimise the visible stepping on the sideframe.  Shapeways do not allow the client to specify print orientation, but it can make a huge difference.  i.Materialise allow the client to request a particular print orientation, but don't guarantee to print in the requested orientation, although they generally comply.  I generate the STL in the "normal" orientation, then rotate it to the desired orientation using the free edition of NetFabb, then I submit the rotated STL for printing.

[Zen]

sml1983 I am looking for cheap O gauge BR1 bogies for some Lima coaches. What price in O gauge Please???