CNC machining for miniatures

[blind pig]

Mate of mine here in Brisbane has just bought one of these:

http://blog.ponoko.com/2010/09/06/user- ... d-printer/

A bit maker-bot-ish?

[mangozac]

Sounds interesting, but still not high enough resolution for what we want...

[Munkey Joe]

its pretty cool though. but yea.... not alot of detail it looks like.

[mangozac]

OK so I'm going to an auction in the morning in which one lot includes this: . It's only a 2.5D machine - my understanding is that the Z-axis has only a few mm of movement and only up or down locations (rather than linear).

Imagine if you will: download templates for a tank. Load them into the software and convert to toolpaths. Set the machine to work and it will cut each one out with precision in styrene sheet. All that's left is to glue them together in the 3D form...

So what do you think it's worth? New they cost AU$5k. This one is obviously used but should be in decent condition. I can buy a new 3D machine (although admittedly not such a professional unit as this) for around $1.5k so honestly it would be a stretch to pay $1k for it...

Oh and don't even bother asking where the auction is Bec - this one is mine

[mangozac]

And I'm now the proud owner of a desktop engraver for only $231

The catch is that it is in fact not the EGX-30A but the previous model, the EGX-30. The only differences between the two are the colour and the fact that the revised model has native USB support. Neither are of any consequence to me and this one came with a parallel to USB converter.

I won't be picking it up until tomorrow morning, so no pics yet. It didn't include drivers or software either, but that should be easy enough to work around...

Should be quite exciting!

[Barn]

That is awesome

[blind pig]

So, good for doing PCB's?

[mangozac]

Not 100% certain yet. It should be, but I need to do some testing first.

[Barn]

[mangozac]

Printed Circuit....... Boards

The name is slightly misleading since these days printing is rarely a part of the process. When the technology was in its infancy the metal traces were screenprinted onto a piece of copper, hence the name. These days photographic process is generally used (or in the case of machines like this, routing, however this is suitable for prototyping only).

</electronics history lesson>

[mangozac]

So after some playing around I'm a little concerned that without the proper Z-axis this machine isn't quite perfectly suited to all of the tasks I'd like to perform. It could be with some basic mechanical hacking, but I'm not sure I want to do that when it turns out that even a used machine like this could be worth as much as $2500. Yes that's how much used ones of the same model are advertised for online!

I discovered that one of my clients has been investigating buying an EGX-30A so I mentioned that I'm interesting in selling my EGX-30. Once I've received a couple of mail ordered replacement attachments I could be in business.

This will however leave me back at square one with no CNC at all, however the idea is that I take the money made from selling the Roland machine and buy something more suited to me. Something like this (). I accept the fact that buying a cheap machine like this is likely to require a degree of mechanical and electronic ability to get it all working properly, but I'm in no short supply of either of those...

Of course that machine is on the cheap and nasty end and theoretically I should be able to get something a bit more expensive for the price I sell the Roland for. I'd be interested to hear the thoughts of our new resident CNC expert Lane

PS. sorry for the sparse replies lately - I've been away on business and spent the last weekend at the Australian Grand Prix. Awesome fun!

[Lane]

[mangozac]

Wow thanks Lane, that is a great little guide!

I should probably have been a bit more descriptive and given a better indication of my level of familiarity with the subject. I was aware of a few of the points you have brought up, however the information was still very helpful!

As a continuation of the discussion about the "Walter 2520" machine:

1. I'm expecting to be doing more work in "flat pack" form rather than machining large 3D objects, for which I feel the 200x250mm working area is acceptable. Slightly larger would probably be preferable but the 300x400mm machine is a big jump in price for what I feel is too much overkill.

2. Repeatability (backlash?) is what I was most worried about. It did sound a little bad. I've read that it's not a big deal when you use the right software that can account for it but ultimately it's still something I'd prefer to minimise.

3. Despite the inconsistent description, judging from the pictures I would guess it's just a trapezoidal leadscrew. Does the leadscrew have much more impact than just on the accuracy/repeatability?

4. I'm planning to cut mainly plastics, but don't doubt I will do the occasional soft metal, in which case I can afford a slower speed because of a lower power spindle. I will be using very small cutters though so the 10k RPM could be a bit low.

Like I said, for such a cheap machine I would be expecting to spend some money to upgrade parts of it (i.e. the spindle motor). I'm just trying to assess exactly how much upgrading will be needed

I had a brief look at the . As a kit it can be had for US$700, but that doesn't include electronics or a spindle. I would expect to be up over the $1500 mark by the time such parts were included. I don't like the idea that the blocks are made of plastic either...

Actually, for that price I'd be best to look at the . I've already got some decent quality/size steppers to use. Shipping will be a bitch though and I'm not sure about the fact that much of it is made from MDF. At least the Chinese machines are Aluminium.

Anybody know of any others?

[Lane]

[mangozac]