Hexapod/tripod cnc

The title says it all... In case you don't know what I am talking about have a look here: http://www.youtube.com/watch?v=LJYhz5aTMnA&feature=channel http://www.youtube.com/watch?v=M9qBbP-1BEA

In theory this type of CNC mill has some advantages over a more typical XYZ setup. What is more, there are programs, such as emc2 that can be used to control it (even the 6 axis version)

My question is - has anyone ever attempted to build one based on arduino and succeed ?

It's certainly a challenge without a floating-point unit. However you could easily use the Arduino to talk to the stepper motor controllers and get a host PC to crunch the geometry...

The tripod is vastly easier to do as the geometry involves +,-,*,/ and square-roots - there is an algorithm for integer square roots that's not too cpu-heavy and there may be some digital differential analyser techniques to help.

Something similar.

http://www.youtube.com/watch?v=quN37YskoaM

Now that's what I call a CNC machine - can go and fetch the tea and coffee as well!

Seems like the software of choice for bizarre/new CNC configurations is EMC2: http://linuxcnc.org/ but needs a bit more power than an Arduino! Sounds like it can adapt to any geometry you want...

I hate when ppl mention that hexapod robot... roughly 50% of search results for 'hexapod + CNC' redirects to a video or an article about it

BTW I am pretty surprised that there isn't (that I know of) one 'official' tutorial (with library + wiring, what programs to use etc.) of the most basic XYZ CNC setup

I expect the response 'why don't you make one yourself?!?', lol but I am not yet there so... :)

That's probably because there really isn't anything basic about even 3 axis CNC.

I am pretty surprised that there isn't (that I know of) one 'official' tutorial (with library + wiring, what programs to use etc.) of the most basic XYZ CNC setup

Not sure what you mean by official but my project is fully documentation with software and schematics:-

http://www.thebox.myzen.co.uk/Hardware/CNC_Conversion.html

I don't agree with the above statement... sure there is a lot of work to be done, but if you can provide a simple to use standardised library, code and a simple tutorial on building a step motor driver (yeah I know you can find quite a lot of information on the internet)...

Getting everything to work well together seems like the biggest problem for most ppl...

What's the point of developing more than 30 or 60 independent projects if one would serve the community better ?

I don't think there is any point in reinventing the wheel each time.... That's what arduino is all about - you don't have to spend hundreds of hours just to read the value of one sensor and send them to processing. Sure all of us could throw away their arduino boards and design a micro controller of their own... lol but is it the point

EDIT ohh yeah sorry mike I must have missed that... I think I've seen your thread about it but it wasn't as informative as that link...

anyway thx for contributing

I don't agree with the above statement..

Well you are entitled to your opinion even if you are wrong.

I don't develop things to serve the community, I develop things because I want to make them. As a service to the community I document them and provide as much information as I can on line. I also answer questions about my stuff.

but if you can provide a simple to use standardised library

One of the major problems with using libraries is that it gives you a false sense that you actually know what is going on. As long as you stick to the original intention of the library creator you are fine. But as soon as you want to extend things you are totally stuck. That is why I prefer not to encapsulate any of my code in this format. I think is is better if all the code is out in the open.

I don't think there is any point in reinventing the wheel each time.

No well you probably would not, but until you have learned to invent the wheel I wouldn't like to drive any card you designed.

That's what arduino is all about

You might think that and again you will be wrong. Is the point of an arduino to bleat because someone has not designed a library simple enough for you to understand? The point of an arduino is not to be able to do any project by cut and paste. The point of an arduino is to learn what you are doing and with a minimum of overheads.

you don't have to spend hundreds of hours just to read the value of one sensor and send them to processing.

Bit of an exaggeration that. However, if you are as stupid as you sound then maybe you do.

Bit of an exaggeration that. However, if you are as stupid as you sound then maybe you do.

lol being a dick doesn't make your dick get any longer :)

No offence but there are things at which I am acctually pretty good and forums where I am a kind of advanced member... though I don't think it gives me right to act like a dick - just because I know more than others and they ask questions that have already been answered 100 times doesn't mean I am in anyway better than them

I don't know how it works here

Though yeah, let say that I get your idea

I guess one could choose either option, find out everything there is to learn or use ready built stuff. Both... have their advantages/disadvantages and often one can even choose something in between.

I must say I love the "Something similar"-video of zoomkat with only a limit on the Z-axis, but I very much wonder how accurate it stays after having made a few steps.

Me, I'm still busy on my first "simple" 3-axis version. It would be great to just place the needed G-code on an SD-card, but getting it to work first has a much higher priority ;)

I'm planning on just using the PC's parallel port and maybe add a micro-controller later.