I spoke too soon - More Mef Wheels

[krow113]

I agree with the above with the only change being casting the hub and rims in metal.

Metal can be polished , chromed , or painted. It also affords a weight to the model, something bike kits need as they are almost always top heavy , and would benefit car kits as well. Easier to work with than resin, as it can go back in the melter if it doesn't work out. Tires and rim combos can be done easily enough as well.

Kit tires are ok in most cases and the kit rims and hubs can be used as masters.

Spokes have to be metal to withstand lacing tension. It took approximatly 7.5 minutes to make a jig for bending the 90  and cutting all the spokes for 1 wheel.  Nipples can be of metal tube as well

This is how I do it on my models.

 

A jig would be ok , I used toothpicks and the plastic top of a rattle-can.

Overcomplicating things , in order to put a certain tool to use , in this case a 3d printer , is disingenuous and not needed. Those things should be for making masters not production work of this size. In the case of wheels the striations rendered during the print process would be horrendous to smooth out and unless the materiel printed has some structural integrity , it will most likely not withstand the lacing procedure or the tension created.

Nice work , but a clearer direction has been pointed out now.

Edited October 25, 2016 by krow113

[Jnkm13]

I agree with Roy's suggestion. It's better to cast them in resin. I think doing 4 sets in 3D print would cost as much or more than the kit. Resin might only be slightly cheaper though.

 

One method i accidentally discovered to soften the 3D prints by Shapeways is soaking in IPA. When i wanted to remove paint from the shock absorbers of the Bentley, i soaked them in IPA for more than 24 hours. The shocks became soft and bendy. It was then quite easy to sand to remove the striations. Leaving it overnight, it soon regained its rigidity. Not sure what sort of magic was at work.

 

Regards,

Jeremy 

[NickD]

@Roy vd M. clear and concise as always

@krow113 Those wheels look fantastic. Think it might take me much more than 71/2 mins to do that

@Jnkm13 Agree on casting hub and rim as a route. though I would like to know jsut how much they would cost for the complete wheel so I will probably run that to completion though it is not a sensible option. A couple of questions. How successful was sanding he nylon material? What route are you taking on the Bentley?

 

So to summarise, the problem we started with was that the wheels on the Italeri Mef have the wrong tyres, wrong hubs and wrong spokes. Apart from that they are perfect.

 

A range of zealotry has been on display: Bottom end - me with OOB wheels and tyres

 

Olivier de St Ralph - with cast rims 72 spokes based on the kit wheels

 

Little Andi - wire spokes - kit rims and hubs (I think)

 

To finish this thread I plan to show how to convert meshes to a printable object and then see how much Shapeways would charge to print. I might even print one if it is not too expensive. As stated several times above it is likely to be over scale in the spoke department, the striations are likely to be an issue too.

 

I'll also see how hubs and rims with holes might work.

 

Thanks again

 

Nick

 

[Roy vd M.]

22 minutes ago, NickD said:

the problem we started with was that the wheels on the Italeri Mef have the wrong tyres, wrong hubs and wrong spokes. Apart from that they are perfect.

 

Positive thinking award granted!  

[NickD]

I haven't actually contributed any new content for a while. The next bit is one of my favourites. My last post had a model that comprised 8 separate meshes. To print it needs to be one. Doing this manually would be a real pain. Blender has yet another modifier called a Boolean and just as it suggests it generates a single mesh that is the union (i.e. adds both meshes together and removes the hidden bits), the difference (removes the bits of the mesh that intersect with the second object), the intersection (i.e. removes all the mesh apart from the piece that intersects with other object).

 

For this model, I manually joined the rim and tyre together as a single mesh and imported the hub into the same mesh. I then did added a boolean modifier set to Union that used each ring of spokes in turn. In theory the modifiers could be stacked without being applied. However, this kills my computer (blender crash work lost). So after several attempts (just to be sure) don't do that any more. Instead hit apply after each ring of spokes. For 4 of the rings it worked brilliantly, somebody really clever at both maths and software works on Blender.

 

On the other two rings it worked less well.

Basically one or two of the spokes went missing leaving just holes in the rim potentially useful for later but not what I needed now. The picture shows the result of difference rather than union. all the spokes should be little short segments but 2 aren't. As it was only a limited number of instances I guessed it might be due to an unfortunate alignment of nodes in the two meshes. A minor shift of location worked. A point to note for the future perhaps.

 

After repeating for all sets of spokes I got:

 TA DAH!

 

Looks pretty though a bit green but is it right?

 

[sharknose156]

at first glance looks very good.  amazing....I will peruse it later comparing it to the real wheel.

 

i wonder if using a 3d printer, one can print another 3d printer hardware!! 

 

this being said, i just saw a small piece of metal in a very expensive and light metal being produced by a 3d industrial printer....

so the spokes would be a piece of cake i was told by the engineer (  a PhD ) using the printer i saw...

 

Very well done Nick !!.. push it as far as you can.

Edited October 26, 2016 by sharknose156

[Jnkm13]

6 hours ago, NickD said:

 

@Jnkm13 Agree on casting hub and rim as a route. though I would like to know jsut how much they would cost for the complete wheel so I will probably run that to completion though it is not a sensible option. A couple of questions. How successful was sanding he nylon material? What route are you taking on the Bentley?

 

 

Nick,

 

I am not sure what material Shapeways is using, but it was quite difficult to sand. After a layer of primer and paint, I noticed that more sanding was required so I soaked them in IPA. This softened the material making sanding much easier. Had me really worried. Good thing is after leaving the shock absorbers overnight, they regained their rigidity.

 

For the Bentley, I will go for the aftermarket wheels by Fernando Pinto. They cost quite a lot but if I were to try lacing the wheels myself, the Bentley will end up wheeless. So after all the effort put into the Bentley, it's a small fortune that I am willing to part with. 

 

Regards,

Jeremy

[NickD]

Next step is get the mesh ready for printing. The printing process likes to have a nice continuous mesh with no holes, duplicates, or massively distorted geometry. Blender provides an Add On called 3D tools that checks the mesh for a range problems. I had done this before but still learned some things this time.

 

The wheel mesh above has nearly 600k faces, which sounds like and is a lot. Basically I used the same sub-surf settings for both the spokes and the hub and rim. As a consequence. The hub and rim have a nice smooth surface and the spokes have a really really smooth surface.I looked at remeshing, which I will cover later.

 

3D Tools identified the following

 

Volume: 20.83cm**3

15400 Manifold faces – free edges – 3%

3155 Intersecting faces

500k zero faces – mesh size is too small on the spokes - probably should go back and fix - not difficult just change the modifier settings and redo the boolean

237 zero edges

9000 non-flat faces

11 sharp edges

140k hanging faces

 

Some are just because the wheel is so small that it triggers length or thickness limits.

All the rest were fixable apart from Manifold Faces. These are faces that have a free edge. Normally they are an issue because free edge = hole.

 

I tried to fix them and found the limits of my laptop. Wheel of doom for about an hour while it thought about it and then...

ta dah - it went a bit wrong to say the least.

 

So plan B - Upload anyway and see what Shapeways made of it. Answer was: mesh was good. So much for the diagnostics.

 

 

Several really useful things. The site is really intuitive and clearly shows what needs to be fixed. As expected its the spoke size.

 

Cost for this wheel would be a whopping £14.38. I'd put the conversion rate but with sterling the way it is it would be misleading.

 

Cost is driven by volume and the wheel is solid. I could reduce the cost if I worked out a way to make the model hollow and include an exit route for the removal of trapped but unused powder in manufacture.

 

This was about the cheapest option. Frosted ultra-detail, high def acrylate and castable wax would print as drawn but with prices north of £62. Not as bad as platinum at £32,000! and it would print properly - won't be doing that then.

 

For those of you who have not tried to print anything yet, this is not untypical of the journey. Models usually have some sort of issue and savings can be made by reducing the enclosed volume.

 

Next steps are increase spoke size (Sorry VT), reduce the face count and separate into two pieces to reduce the internal volume.

 

Thanks if you are sticking with this

 

Regards

Nick

[Roy vd M.]

Nick you've been providing some interesting information here and I look forward to learning how much you will be able to bring the price down. 

 

Don't forget that a 1/12th scale model (especially filled with metal parts) is quite heavy so I definitely wouldn't make the remaining material too thin if I were you. 

[NickD]

@Roy vd M.- good point well made.

 

I have a cunning plan (as Baldrick might say in Blackadder). Since the last post, I've been thinking. For me the tyres are OK. So my next step will just be the rims, spokes and hubs which should get the volume down. If needed one could print the tyres in any desired pattern in a cheap material that could be used as a pattern for a resin mould.

 

Plan A would be to print them in the resin acrylate - because they will and because I have not used it before. My concern is exactly as you state - subjected to load the spokes might ping like dry spagetti into expensive fragments. Hopefull there are enough that this will not happen.

 

Plan B is increase the gauge (sorry VT) and print in the nylon. Likely to be durable but roughness and painting issues.

 

Nick