This morning I have replied to a post and also introduced myself there, but I think I should go for a proper introduction in the right place, so here we go ...
I am Italian (well, no one is perfect!) and currently live in Kent. I come from a software development background (web and mobile apps) with occasional drifts into DIY building stuff. Three years ago, wanting to learn arc welding, I built a trike bicycle using steel tubes and parts cannibalised from old bicycles I found at a local scrap metal place in Italy.
Fast forward to last September 2015 when I bought an OX-type machine from a UK supplier. It's sturdier than the openbuilds standard model in that it has double plates on the Y axes. My experience with it is not too bad, but I believe I have exaggerated with the size (1000 x 1500mm). Its limit is visible in the design, especially the direct-coupling belt drive, where you have your steppers rotating very slowly compared to when using acme screws, for example, or at least some gearing on the pulleys (I really appreciated the excellent post by JAZZCNC on http://www.mycncuk.com/threads/3823-...med-CNC-router about this and other aspect of CNC design). This specific shortcoming is my main concern, as it does not only limit resolution, but, because the steppers would be running slower than their optimal speed, it seems to me that this is also the cause of unwanted vibration at low feeds.
However, the entire exercise of building a ready-prepared mechanical kit has been an invaluable (albeit expensive) learning experience for me. I feel I have learnt a lot about basic CNC. I am controlling the machine with the xPRO CNC v2 controller board from Spark Concepts, which integrates four stepper drivers. I am using Fusion 360 as CAD-CAM; not only is this now effectively free and, but, although it takes a bit more time than other programs to get to grips with, the effort is paid back in learning experience, as I find the CAM module amazingly detailed. I use Universal G-Code Sender to send the code generated by Fusion to the actual machine.
I have recently started renting a workshop where I plan to develop lamps for children of my own design (mainly woodwork, but I foresee adding some metal parts, too). Whilst I'll be using my current machine for the time being (I am using my existing DeWalt 611 router on it, again not the best option ...), I am seriously thinking of building my own. I could design my new machine from scratch, and have now had a look at the very many building ideas in this forum, but I feel that I could do with some general advice on initial building choices for a strong, steel-framed (or better aluminium?) design that could be used as the basis for designing an industrial-quality DIY machine?
I am looking at a machine in the size range 1000mm x 1500mm, capable of managing an entire plywood sheet 1200x600, and the budget would be in the range £1,000-£2,000. I would re-use my existing controller and possibly my existing NEMA 23 steppers. Eventually, I would also like to add an engraving laser head (max 3W) aside the spindle (perhaps a liquid-cooled Chinese one?).
This could hopefully be my spare-time project this year. Is there any 'consolidated' knowledge on DIY CNC design, such as this is better than that, do this, don't do that? Or the point (and fun) of it all is that there is not such a thing?
It's a fascinating subject, and I can see CNC DIY-type machine are growing in popularity. They are more complex to manage than 3D printers or laser CNCs, but the ability to mill 'real' materials has no comparison!
Thank you all !!!
Ciao // Enrico
Hi Enrico - I too have only just joined here. You can't be far away as I'm just over the border from Kent as I'm in East Sussex - where abouts are you?
Hi Andrew, nice to meet you, I am based in Tonbridge. I have seen your welcome-post ... wow! You have quite a shop!
So essentially you're just up the road then ! (My farm's on the A21)
Originally Posted by Enrico
Uhmm ... yes, quite close I guess! I should pop in at some point to see your CNC machines!!
I have seen you have in mind to work with Fusion 360. It's a great piece of software (you would expect nothing less from Autodesk!). Compared to, say, SketchUp, it requires a little bit more planning when drawing. You can draw parametrically, defining variables to define specific parts of your drawing; this allows you, for example, to resize the drawing properly, not just scaling the entire thing. The CAM module is quite sophisticated (not that I have a huge experience about CAM software, anyway, but it seems to me quite detailed compared to some other programs I have seen). There are plenty of tutorials and video resources about it, and that is one other reason I chose it (apart from the price tag!).
Enrico, yes it would be good to meet - it's only a few miles down the road.
Fusion360: yes I'm hearing good things about it. Can't download it at the moment as it needs a 64 bit machine and all my current PCs are 32 bit. I've won a suitable base unit on ebay which I expect to be delivered next week, and the 24" monitor and SSD drive came on Friday :) - so shouldn't be too long before I'm up and running.
The CAM side versatility is a big plus to me. I need customisable Fanuc G code for the Wire EDM, Heidenhain conversational for the Partsmaster, and 'sort of' G code for the Traub lathe. The first two I've managed to customise in FeatureCAM sucessfully, but I've not attempted the Traub - mega complex having two spindles, a moving turret AND powered tooling. It has pretty good coding input 'at the machine' but it's all "mandraulic".
Last edited by AndrewMawson; 17-01-2016 at 03:41 PM.
This is part of a screenshot of the huge list of post processors available by default on Fusion. It contains a number of scripts for Fanuc and Heidenhain machines. You should be able to work on those, or even get help from Autodesk (I read somewhere that they were available to provide help on scripting).
Hope this helps!
Looks good to me - thanks for that. I'm not the most patient of people when it comes to waiting for goodies to arrive