make your own models

[Ian MacMillan]

AutoCAD is pretty much out the window on this one, hell I remember using it my freshman year in HS. 2 of my clients down here basicly run Solidwords and Autodesk in their CNC departments. One being a general machine shop, and the other manufactures racks for servers.

[Robbman]

Message deleted

[Sokramiketes]

You'll have to ask Mark4Design for the exact details.  I'm not at liberty to discuss the actual RP process used.

It's not much of a secret anymore since Mark has it listed on his site.  The machine is called Perfactory.  Info can be found here:
http://www.mark4design.com/html/prototyping___masters.html, and the company here: http://www.envisiontec.de/02hperfa.htm

Never said it was a secret... just that you'll have to ask M4D.  Nowehere in my contract does it state that I am allowed to freely discuss the process.

I bet your contract doesn't state that you're allowed to eat meat either... but I'm sure you do. 

Perhaps your contract say's you're not allowed to discuss the process, and that's what you meant to say? 

I don't blame you though, most engineers have trouble with the English language.

[Robbman]

Message deleted

[Iain]

I mean things like drawing a new line rather than extending an existing line are not the best of habits in Autocad.

Solidworks IS easy to use.

Now Robbman, a question:  what is the best thickness for hood doors for an N scale model?  If they were done perfectly to scale, would they not show up?  Same thing for rivets and the like.

If you are not able to measure an Atlas/Kato shell yourself then you have no business attempting to create solid models for production.

How am I supposed to measure a shell I don't have?  And if I don't ask about some things then how would I find out save for trial and error, which in this case is not very feasible.

[ljudice]

interesting issues for loco shells, but for me, the possibilities of building the
MOW fleet I've always wanted is interesting...

[Sokramiketes]

I mean things like drawing a new line rather than extending an existing line are not the best of habits in Autocad.

Solidworks IS easy to use.

Now Robbman, a question:  what is the best thickness for hood doors for an N scale model?  If they were done perfectly to scale, would they not show up?  Same thing for rivets and the like.

If you are not able to measure an Atlas/Kato shell yourself then you have no business attempting to create solid models for production.

How am I supposed to measure a shell I don't have?  And if I don't ask about some things then how would I find out save for trial and error, which in this case is not very feasible.

You don't have a single hood unit from Atlas or Kato?  I find that hard to believe.  Who's models have you been showing photos of?

Everyone else has had to do some trial and error... why would you be any different?  Mark does a trial run on everyone's shells.  Chances are you're not going to get everything right the first time.

[RS-27]

I mean things like drawing a new line rather than extending an existing line are not the best of habits in Autocad.

Solidworks IS easy to use.

Now Robbman, a question:  what is the best thickness for hood doors for an N scale model?  If they were done perfectly to scale, would they not show up?  Same thing for rivets and the like.

If you are not able to measure an Atlas/Kato shell yourself then you have no business attempting to create solid models for production.

Mike, that's awfully damned condesending.   And who says that Atlas/Kato has it right in the first place?

************************************************************

the following was an exchange on the nscalekitbashing list on yahoo

At 12:22 PM 1/10/2007, you wrote:

I don't know about the EMDs across the pond. But on the US EMDs the doors on the long hood are proud by 1/8" from the carbody and the
latches are proud of the door bi 1/8" also.  Maybe make a dimple with a drill bit and cover with an etched .002" piece.

Richard
*****************
You're probably right, I've not climbed onto any locos lately,  but you'll have to admit that 1/8" isn't standing very proud.

In fact, it is .00078" on a N model.  I measured a Atlas SD35, the doors were .005" proud, the latch is .004".  A GP38 was a bit thinner at .004" for each.  It's not specified on BLMA's site, but I'd guess the doors are .010". 

My point is, that for emphasis, the thicknesses are exaggerated because they've always been exaggerated and that's what we expect/accept.  Also, could we see the doors if they were scale thickness?  And would painting obliterate them?

Unfortunately, the only aftermarket parts we have are probably worse (thicker).  You probably could get away with using the BLMA parts for all doors, but IF they are .010" thick, then a mix would be noticeable. (to me at least)  added to quoted email:(sorry, Craig)

And how thick and strong is the required adhesive layer?

Bob in IDaho, who keeps looking for the magic material/technique.

[Mark4]

I bet your contract doesn't state that you're allowed to eat meat either... but I'm sure you do. 

Perhaps your contract say's you're not allowed to discuss the process, and that's what you meant to say? 

I don't blame you though, most engineers have trouble with the English language.

"say's" Skibbe?! Careful, the apostrophe police could be watching! ...and please, don't feed Robb meat...at least not red meat. As well as being specifically forbidden in his contract (clause 14, subsection 12, paragraph 2) it makes him too excitable.

Some reflections on scale model design:

Hood door thickness in N is invariably overscale. The customers like to be able to see the doors and feel that they are getting value for money . The same applies to rivets.

3D solid modelling software can be easier to use than the old 2D CAD. Part of the reason for this is that 2D engineering drawing standards/codes have a lot of conventions to allow 3D objects to be represented in 2D - section views, center lines, projections etc etc. As well as knowing how to use the software you had to know the relevant drawing standards and had to analyse the task to be able to accurately and unambiguously represent the part or assembly with the minimum number of views. Each year at a local university I present a lecture series to final year engineering students on modern prototyping techniques and it's great to see the way that parametric solid modelling software frees the students up to concentrate on the design rather than the mechanics of "engineering drawing" that we waded through back in my days as a student.

3D solid modelling software packages such as Solidworks, Inventor, Alibre, Catia, ProEngineer etc have limitations though. Most railroad prototypes are a collection of relatively simple shapes. When you consider automobiles with the compound 3D surfaces you really need to look at software that works in a different way; surface modelling software. Easy it is not.

[Iain]

Thanks Bob and Mark.  At least someone is helpful.

Has it occured to anyone that I may not own a set of Vernier calipers (or any calipers at all)?

[Chris333]

Harbor Freight has a pair for $16.00 and it should be your next purchase if your thinking about drawing or building models. I have this same exact set along with 2 others. It is not the most sophisticated piece of equipment, but it will do what you want it to and you will use them all the time.
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=47257

You want cheaper, I'll give you cheaper ($7.99)
http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=93293


Order them now

[Sokramiketes]

Mike, that's awfully damned condesending.   And who says that Atlas/Kato has it right in the first place?

I'm glad I was clear.  I suppose you haven't had the honor of receiving emails from "Iain" fishing for freebies... be it models, kits, or information.  If you want to encourage that behavior, go ahead.  I will not.

I never said Atlas/Kato had it right, but they are acceptable and easily accessible.  If, after measuring one, someone wanted to go closer to scale with less pronounced detail, then a design decision would have to be made.  If someone wanted something that looked good and was a known dimension, then just copy an Atlas/Kato shell.  There's some art involved, and it's not a simple answer.

[KelNScale]

So let me ask a question. What is easier, drawing to full size and scale down, or drawing to scale size with 3D software? No freebies or whatnot, just a design question.

Kel

[Craig Martyn]

So let me ask a question. What is easier, drawing to full size and scale down, or drawing to scale size with 3D software? No freebies or whatnot, just a design question.

Kel

Although there may be different answers to this, my personal opinion is that it depends on what you are drawing.  For instance, if you drew something that had numerous rivets and other very fine detail at 1:1, you will NEED to bump those things up in size to make them visible in 1:160.  That said, you'll do less work drawing the model in 1:160 from the get go (keeping in mind different tolerances and what not) than drawing it 1:1 and having to go back and re-work things that simply will not come out right (or at all) if drawn to scale.

This thought by the way goes for anything from plastic injection molding to etching to rapid prototyping...  etc.

[Mark4]

So let me ask a question. What is easier, drawing to full size and scale down, or drawing to scale size with 3D software? No freebies or whatnot, just a design question.

Kel

For file interchange between different organizations full size is handy. Key elements can be set up with parametrics so that they can be tweaked later. For example rivet hieght and diameter can be set up as 2 user-defined variables. Changing the variable changes every rivet on the model. This is especially useful if you want a 3D model to be used in more than one scale. As for what is easier - it really depends on the task at hand. After a while a lot of the design is based on earlier work so it depends on what scale the original work was in.