Does anyone know which CADCAM software Hornby, Bachman, Dapol use?

[EdwardNo2]

Hello all

 

Just joined RMWeb, and getting back into model railways after 40 years! Things have changed a bit!

 

I am very interested all to learn all I can about modern CAD Cam design methods used in modelling these days, but haven't seen any articles about how the modern models are produced by Hornby, Bachman, Dapol, et al. I was originally trained as an electronic engineer, so fascinated to see these methods have revolutionised the hobby.

 

Can anyone tell me what CAD software the manufacturers use or point me in the direction of any 'how it's done' articles I have missed?

 

Would be interested to see if any of the big boys techniques might scale down into small run production methods.

 

Thanks and cheers

Mike

[backofanenvelope]

Mike,

 

Although I don't know I suspect they will use Solidworks which is not cheap if you have to buy it. The majority on hobby users tend to use the free or cheaper versions on the industrial packages that the like of Hornby et al use. The learning curve is not easy but there are a numbe of apps that make life easier for the modeller on being Google sketchup. You will find there are people that have used it on RMweb to great effect and all packages have there quirks which you have to learn but I would trawl the net if you are interested in taking this up but don't be too disheartened when you can't produce a purfect Class 47 or something in two clicks.

 

Others with more knowledge and experince will be along to provide more..

Tom

[kipford]

Mike

Model Rail No 23 September 2000 has an article on 'Developing Hornby's new Clan Line'. Although over 11 years old the techniques shown have not altered significantly except the use of more up to date rapid prototyping systems and materials.

 

Regards

 

Dave

[spamcan61]

Some former work colleagues of mine went on a training course for Pro/ENGINEER, which included videos showing Hornby/Scalextric at work, so I assume they do or used to use that.

[EdwardNo2]

 

Model Rail No 23 September 2000 has an article on 'Developing Hornby's new Clan Line'. Although over 11 years old the techniques shown have not altered significantly except the use of more up to date rapid prototyping systems and materials.

 

 

Many thanks! I will endeavour to track down a copy.

Mike

[EdwardNo2]

 

Although over 11 years old the techniques shown have not altered significantly except the use of more up to date rapid prototyping systems and materials.

 

 

Oops, replied too quickly. Are you able to cast any more light on the rapid prototyping systems and materials which they are now using? Have they adopted 3D printing or do they do something else?

 

Best

Mike

[richbrummitt]

Have they adopted 3D printing or do they do something else?

 

I would be surprised if they haven't

[Atso]

Hornby and Dapol do use 3D printing to produce at least some of their prototypes/proving models. Stereolithography seems to be the favoured 3D printing method here - the Hornby model in the above mentioned article uses this as did the Dapol Q1. Also CNC machining would likely be used to produce some of the chassis components.

 

Shapeways and iMaterialize (and others I've not tried I'm sure!) offer a good printing service at a reasonable cost and although the quality isn't quite up to the standards of some of the more expensive methods (look up Fineline printing for some truely awsome prints), it is improving all the time!

[Rabs]

This is an interesting area - I've been building myself a mini-mill/FDM 3D printer (similar to Makerbot: http://www.makerbot.com/) to be able to design and print custom models for my N gauge model railway (also as a major hobby project in its own right).

Most budget projects us a process called Fused Deposition Modelling (FDM), which is a bit like building a model by squeezing molten plastic through a nozzle, like icing a cake. Most serious rapid prototyping systems use stereolithography (either SLA or SLS). These are capable of much higher resolutions, but are seriously expensive bits of kit to buy and run. Also, the material they use is quite brittle - so you need to be careful with the finished item. There are lots of companies out there that will build to order on their machines. I use them regularly for prototyping in my job. As an estimate, the going rate for something the size of an N gauge tank loco would be about £50 from the supplier we use most often. For a given resolution model you pretty much pay for the volume, as stereo lithography inherently doesn't care how complicated the part is. I suspect that it probably wouldn't be cost effective for larger scales unless you have a one off model that you simply must have. Also, a good prototyping workshop may well have a 5 axis mill which could machine a model from solid. This might work out cheaper than additive methods at present - but it does limit the geometry a bit, internal features possible in SLA or injection moulded parts are difficult to reproduce with machining.

 

On a closely related note (not worth a separate thread) - I've been looking for source 3D CAD models of locomotives to print. I thought about trying to extract the 3D model data from one of the railway simulation games - although I'm not sure what the copyright situation would be with this. It might be ok for add-on locos produced by fans if I were to contact them and ask permission, but I doubt the developers would be willing to give out their 3D models, as it seems to be one of the main differentiator between the different simulation games. Does anyone know of any resources where people have models available (I'm a 1940s GWR modeller)?

[Pugsley]

This is an interesting area - I've been building myself a mini-mill/FDM 3D printer (similar to Makerbot: http://www.makerbot.com/) to be able to design and print custom models for my N gauge model railway (also as a major hobby project in its own right).

Interesting stuff, but I think you'll have a problem with the resolution for N gauge with the Makerbot device. I've looked at them, but they're not really suitable at the moment, IMO.

 

This device looks like it has real potential, if he can carry it off without infringing anybody's patents:

http://3dhomemade.blogspot.com/

[EdwardNo2]

On a closely related note (not worth a separate thread) - I've been looking for source 3D CAD models of locomotives to print. I thought about trying to extract the 3D model data from one of the railway simulation games - although I'm not sure what the copyright situation would be with this. It might be ok for add-on locos produced by fans if I were to contact them and ask permission, but I doubt the developers would be willing to give out their 3D models, as it seems to be one of the main differentiator between the different simulation games. Does anyone know of any resources where people have models available (I'm a 1940s GWR modeller)?

 

I agree that 3D model data from one of the railway simulation games would be a rich source of CAD models. They might need some tweaking. Maybe the answer would be for a repository of CAD models to be set up/built up as 'Creative Commons' resources? If we all worked on them it would ensure authenticity. This sort of thing works well in my work field (web development) where we use Open Source software for all our services and solutions. The 'Creative Commons' licence can be set to allow free use for personal usage but with limitations/licence costs for commercial use.

 

Rabs you might want to make contact with www.rail-sim.co.uk who did a lot of the work on the Trainz Classic UK games - I'd be interested to hear how you get on if you do.

 

I'd like to thank everyone for their inputs this week, it's been very helpful in pointing me to the right sort of Google searches to make, I found the rapid prototyping sites very interesting. Rabs, you mentioned a cost of £50 for an N gauge loco body - would that be using stereolithography? Any idea for 4mm? I haven't been able to find any sample costs on the websites yet.

 

Regarding CAD software, from my researches this week, Blender looks like a very exciting solution. It's free (open source), can exchange data with the proprietary programs and outputs .stl files for rapid prototyping. I think it has been used fairly widely for railway simulation games design. It would be interesting to hear from someone who is familar with Solidworks and Blender to know how they compare in terms of capability. On first look Blender looks very powerful and has a high level of spec.

 

Best

Mike

[EdwardNo2]

Some former work colleagues of mine went on a training course for Pro/ENGINEER, which included videos showing Hornby/Scalextric at work, so I assume they do or used to use that.

 

I also think that this seems likely too after a bit of digging online. PTC market Creo Parametric as the 3D design standard here (http://www.ptc.com/product/pro-engineer) and they are running a schools competition with Scalextrix (http://scalextric4schools.org/) with a special schools licenced software product (any school teachers here?).

[Red Devil]

I'd like to thank everyone for their inputs this week, it's been very helpful in pointing me to the right sort of Google searches to make, I found the rapid prototyping sites very interesting. Rabs, you mentioned a cost of £50 for an N gauge loco body - would that be using stereolithography? Any idea for 4mm? I haven't been able to find any sample costs on the websites yet.

 

Like anything else, depends who you use, how quick you want it and how much resolution you want to pay for.

 

 

 

These are 4mm models done in Prime Grey by I Materialise which is an SLA material, cheaper than an n gauge loco body....!

[Rabs]

Very interesting Red Devil. I'll have a look at them.

 

Interesting stuff, but I think you'll have a problem with the resolution for N gauge with the Makerbot device. I've looked at them, but they're not really suitable at the moment, IMO.

 

This device looks like it has real potential, if he can carry it off without infringing anybody's patents:

http://3dhomemade.blogspot.com/

 

Absolutely - I'm thinking buildings and scenery items more than rolling stock. Also my design is a bit smarter than the standard makerbot. I'm mounting one of their printing heads onto a device which also has a CNC micro mill. The idea being that I will be able to print a layer, tidy it up with the mill and then continue printing the next layer. Then I get the best of both worlds - the complex features of printing with the accuracy of a decent CNC mill. The downside is that it will be very slow to produce items - but for one off custom models that will be insignificant to the design time so I don't think I'll care too much.

 

I've also been following that website closely - it's a neat solution and might become my next project if, as you say, it can be done legally.

[richbrummitt]

I also think that this seems likely too after a bit of digging online. PTC market Creo Parametric as the 3D design standard here (http://www.ptc.com/p...ct/pro-engineer) and they are running a schools competition with Scalextrix (http://scalextric4schools.org/) with a special schools licenced software product (any school teachers here?).

 

PTC changed the names of Pro/Engineer and CoCreate to Creo Parametric and Creo Direct. (Actually they rebuilt the entire software) So Pro/Engineer will become a legacy application now.

[lyneux]

I agree that 3D model data from one of the railway simulation games would be a rich source of CAD models.

 

I'm not sure that they would. I had a look at a view of the wagon models on the LTSV website but a lot of them use texture mapping for e.g. bogie details which is not "proper" 3D modelling. Take a closer look at the model and you can see that only the very basic shape of the model is actually 3D data and that there is an image overlaid onto the model to provide the "detail".

 

Regarding the creative commons idea... I think this is a good thing and should be encouraged. However, in the railway modelling world I'm not sure it will work. A quick search on the term "copyright" in the box on the top right will show you how paranoid people are about intellectual property round here. People get very worried about copyright over photographs (surely one of the easiest things to "share") let alone something that might have taken several months of work.

[Rods_of_Revolution]

I'm not sure that they would. I had a look at a view of the wagon models on the LTSV website but a lot of them use texture mapping for e.g. bogie details which is not "proper" 3D modelling. Take a closer look at the model and you can see that only the very basic shape of the model is actually 3D data and that there is an image overlaid onto the model to provide the "detail".

 

Regarding the creative commons idea... I think this is a good thing and should be encouraged. However, in the railway modelling world I'm not sure it will work. A quick search on the term "copyright" in the box on the top right will show you how paranoid people are about intellectual property round here. People get very worried about copyright over photographs (surely one of the easiest things to "share") let alone something that might have taken several months of work.

 

Lyneux is completely correct, as someone who has produced railway models for 3D Printing as well as being the leading 3D Artist on many of Oovee's Railworks addons I can tell you that the type of modelling used is completely different.

 

For starters models used in computer games do not have to be "watertight", you deliberately avoid including any polygons on a model which are not visable to the user as they won't get any benefit of them being there but they will still use computer resources. So any model for use in computer games will be like swiss cheese with regards to trying to print it. When modelling for games another method of reducing the poly count are things known as "smoothing groups", this means you can have a cylinder with 12 sides, you can then tell the computer to render it smooth, so it will look completely smooth to the user when in fact it isn't. If you were to print this seemingly smooth model you'd just get the 12 sided cylinder.

 

Additionally you have no constraints on thickness of parts in computer games, if the window frame is 100mm wide then you produce it 100mm wide, with 3D Printing you have a minimum thickness which can be printed, this may mean handrails, window frames etc have to be chunkier than in reality. So many parts of a computer model would be too thin to print, even if the mesh was not riddled with holes. Not to mention lots of the detail on a game model is actually just a texture, it's effectively painted on.

 

When modelling for games you are trying to use as few polgons and as simple mesh as possible to use as few computer resources as possible whilst using various tricks from smoothing groups to bump maps to make the model look more detailed than it actually is. With 3D Printing its the exact opposite, you want to use as many polys as possible to get as much detail on a print as possible.

 

The bottom line is you might as well start from scratch, at the very best you could "trace" around the computer game model to get the same overall shape and dimensions.

 

Kindest Regards,

 

Jack

[lyneux]

A very good summary Jack!

 

at the very best you could "trace" around the computer game model to get the same overall shape and dimensions

 

Not knocking you 3D computer artists, but I'd much rather use an accurate 2D drawing based on prototype dimensions to trace around than rely that someone's 3D model has the correct dimensions.

[Rods_of_Revolution]

A very good summary Jack!

 

 

 

Not knocking you 3D computer artists, but I'd much rather use an accurate 2D drawing based on prototype dimensions to trace around than rely that someone's 3D model has the correct dimensions.

 

So would I, all the models I have produced for railworks are from 2D scale drawings. You can't beat a good old fashioned drawing at the end of the day, after all a computer is merely a tool, what's a saw without the dimensions to cut!?

 

Cheers,

 

Jack