Thread: Good morning.
Hello, yet another new member. I'm a scratch build model maker, I create museum standard, 1/8th usually, models of classic racing cars fom the 50's, and 60's. I work with a small model maker's lathe, and mill, but I wish to start making parts via a cnc milling machine. Since I am now retired my budget is sad to say the least, proof that model making as a profession does not make you rich. AutoCAD is my preferred medium for creating drawings for the models, and I am well into turning all my 2D drawings into solids, and thence into .stl format. I had two items built up for me in wax via a SolidScape machine, and I realise that the granulation of the surface was so noticable that any castings would show the surface markings so badly that I would not be able to clean up without losing the profiles or shapes. So, I am now going down the subtractive machining route. I have no experience whatsoever in the use of a cnc mill, so please give me some latitude for stupidity. I have considered constructing a small mill, and to this end I have looked at all the methods, that is to say, extruded aluminium sections for a gantry type of mill, modification of an existing small mill, or the modification of one of the small educational mills that sometimes come on the market. I have to say that for the very small parameters that I need for cutting, would point to an educational type that has ballscrews, and may have all the electronics then gutted, allowing for the retro fit of a 4axis system. If anyone out there has any information that could point me in the right directions, I would be more than grateful; also, if I can likewise help anyone please ask.
I'd go for a mini mill CNC conversion if you want tiny detail, routers aren't accurate enough. Tiny tools take a long time to cut and you need a good concentric grip on them. You can get quite intricate if you get it right.
This pic shows a smallest possible letter 'S' I cut into aluminium using a desktop mill with a 5 micron step. This is extreme close up, spot the thread of tobacco and ballpoint pen tip. The cut on the right was a 1mm tool, the cut on the left was the 0.5mm tool (see pic 2). Tool turning at about 3000 rpm.
To do this required true zero backlash and a good flow of suds to carry the cuttings away. How small do you need to go?
Thank you so much for your message, much appreciated. Can you tell me what are the dimensions of the letters, and what sort of mill was used ? This looks like the direction that I need to go in. What was the cutting language, or program used, and was a primary drawing created for the cutting process? I am aware of the problems of backlash, this is why my assumption is that I should find a machine with ballscrews. I would like to work from .stl files, if only because of the fact that I have turned my drawings into solids and exported them as .stl's. This was because my first direction was to have the parts built up in wax for investment. As I said in my introduction this method has nasty flaws, I have inserted a photograph of an inlet manifold and a supercharger casing, and you can see the problems with the surface finish.
The manifold is approx 40mm long. Also I should add the fact that this is the first time I have used a forum, I truly am learning how it all works as I go along and I am trying very hard not to lose anything. I hope the photograph arrives okay.
Back to the machinery. When you suggest a mini mill, are you talking of the types of mill one sees at emporiums such as Axminster Power Tools? I have talked with a few model makers in the past and they were for the most part using cnc for only the most basic of machining, and as you can see my ambition is to be able to perhaps create parts on a small, very rigid, 4 axis mill.
The manifold is the sort of thing where I have ended up fabricating in brass, using complex jigs, silver soldering, and then finishing by eye with a host of small files and scrapers, most of which I made or modified myself. 16BA tapping is the norm for this sort of model making, and the last models I made, a 1959 and a 1960 Ferrari 250 Testa Rossa sports racing cars, had steering boxes with hand filed worm and gear inside. This was when I realised that there had to be a better way.
Once again many thanks for your reply.
Last edited by Lee Roberts; 06-04-2012 at 09:39 PM.
Just as the printed article had flaws in it the subtraction method with have tool marks that may not appeal to your accuracy without hand cleaning up or a very light bead blast.
That's just the nature of the beast and just making you aware of it.
Looking at your examples you will need a 3D CAM program to produce much of this work so allow around £500 to 1K for a decent program otherwise you are limiting yourself too much.John S -
Thanks for your reply John S. Is it possible that you could enlarge on your comments re 3D CAM program. Do you mean the type of software often applied to Roland machines?
The need to clean up after machining is not a problem, this being a normal task undertaken after my hand operated machining, but the printed objects really do have an awful surface finish, and that was when the SolidScape machine was printing at its highest resolution. I heeded the story told to me by a person who makes scratch kits for model locomotives and rolling stock. His process is briefly, draw part in solid and export as .stl, build via wax deposition, send wax pattern for investment pattern, repair the new brass pattern and make good missing profiles etc, send repaired brass pattern back to foundry for RTV moulds to be created, make multiple wax patterns from RTV mould for casting, for customers. A frightening process. So that's why I wish to go go down the path of computer guided cutting; also the model on which I am working at this moment has a great deal of aluminium castings, which would benefit from very accurite production.
Robin, a sudden thought. Would it be viable to make 3axes cuts and having made a frame for the item being machined, flip it through 90degrees so as to cut each face, having prepared suitable new drawings for each face's cutting pathway. Lastly where have all the CAMM 3's vanished to?
Being on a shoe string [ nothing wrong with that - it may alter ] take a look at Cut3D from Vectric.
Read the blurb and then you can download a working demo package and do the tutorials.
This will give you a very good insight to what it can do.
The demo package will generate code but only for the models supplied, you can use your own model to see what happens but it won't generate code for it.
It will also do 4 sided machining as you mentioned to Robin. A 3D CAM package that will run 4 axis is very spendy.
From memory the Roland packages were very crude and only to show students what was achievable, not what was achievable as quality. Remember some of this file could well be running for 5 to 8 hours, something not possible in schools.
BTW do you have a signature or name ?John S -
The CAMM-3 is distinctly weird, not recommended, I only bought one on a nostalgia trip.
NEW MEMBER: Morning all.By Roadrunna in forum New Member IntroductionsReplies: 5Last Post: 24-01-2014, 12:03 PM
NEW MEMBER: Morning AllBy MAW in forum New Member IntroductionsReplies: 7Last Post: 07-03-2012, 02:53 PM
NEW MEMBER: good morning from menlo parkBy Douglas114 in forum New Member IntroductionsReplies: 1Last Post: 06-12-2010, 12:56 PM
NEW MEMBER: morning allBy kris251 in forum New Member IntroductionsReplies: 1Last Post: 12-11-2009, 06:37 AM
NEW MEMBER: Good Morning from a very windy NorfolkBy greybeard in forum New Member IntroductionsReplies: 1Last Post: 31-01-2009, 10:37 AM