CNC machine build

[paul2004]

Jmenge, if cost is the key factor, then yes, you can save a lot of money on a DIY machine.
But please keep in mind, it's a time consuming process. I did that route, my big "lesson learned" is to find proven plans from someone that
has already worked out the kinks. Experimenting with design as you go (as I did) gets expensive.
That said, I have a nice DIY CNC machine for less than a shopbot or something like that.
Is it as good as a shopbot? I am not going to make that claim, I don't know, but it serves my needs well.
You learn a ton while building your own machine too. It's nice to be able to debug problems when they happen.
Overall , a fun experience. If my house burnt down, I would probably do DIY route again.

[JDuke]

Jmenge, if cost is the key factor, then yes, you can save a lot of money on a DIY machine.
But please keep in mind, it's a time consuming process. I did that route, my big "lesson learned" is to find proven plans from someone that
has already worked out the kinks.  Experimenting with design as you go (as I did) gets expensive.
That said, I have a nice DIY CNC machine for less than a shopbot or something like that.
Is it as good as a shopbot? I am not going  to make that claim, I don't know, but it serves my needs well.
You learn a ton while building your own machine too. It's nice to be able to debug problems when they happen.
Overall , a fun experience. If my house burnt down, I would probably do DIY route again.

As mentioned if cost is the key factor, DIY is generally the way to go.

I’ll definitely second the idea of using someone’s solid plans, as experiments get expensive very quickly.

You may also want to scour Craigslist or any other local market as lots of people start a build and either don’t finish it, or don’t use it much once built, and may sell for an attractive price.

EG, my first ‘builds’ were control boxes to get x-carves up and running.

I bought 3 of them in some state of non functioning and built control boxes for significantly less than you can buy a replacement for, with more robust electronics than are OEM. 

GRBL based systems are a great starting point, as the principles translate well to more robust systems, but the cost of entry is quite manageable.
You could build an arduino based control box for less than $200, the software for sending G code Universal G code sender is free, and GRBL is free as well.
For generating simple 2.5d g code, there were several free options as well.

In contrast
Even a Mach3 license is around $200
The hardware to run it on older computer with parallel port isn’t expensive but if you want anything later than old Win XP machine you have to get an Ethernet smooth stepper or similar for another $200

Duke