PDA

View Full Version : BUILD LOG: My 2nd version of a 2.5D CNC milling machine



tribbles
27-07-2009, 10:49 PM
I've now started the second version of my 2.5D CNC milling machine (many thanks to Gary's advice on parts) - I've got 10 days to build it (holiday from work). The basics are designed, but there's a lot of flexible design involved (i.e. "make it up as I go along" :) ).

I'll be (hopefully) updating my website every evening with the activities of the day.

http://www.tribbeck.com/steel/stlcutter2/

bikepete
28-07-2009, 04:28 PM
As a newbie I hesitate to weigh in but are you sure one block either side of the router on the Z will be sufficient? Looks like a fairly heavy unit and four blocks total would much better resist the torque of its weight hanging out the front, let aside any cutting forces. If you want to stick with one each side to retain travel then maybe the double-length ones would be a good idea e.g. these (http://www.worldofcnc.com/products.asp?recnumber=25)?

tribbles
28-07-2009, 07:06 PM
Thanks for the comments - it was something I considered, and I still may - I've arranged the ball screw nut quite a bit lower than the two blocks, so that adds some stability. There's some space below which can take another pair of blocks if I find there's some play.

Lee Roberts
29-07-2009, 12:10 AM
Nice to see you are using the supported rail for the z-axis Jason, i'v also used the same for my second machine:

pavlo
29-07-2009, 12:13 AM
Why have all that travel on the Z axis? If you are working with the router raised by more than a couple of inches the bottom end of the linear rails will clash with the work piece. I'm a newbie at this and have yet to build my Z axis so maybe I'm wrong, perhaps one of the more experienced guys will give their thoughts?

Lee Roberts
29-07-2009, 01:14 AM
Hi pavlo,

Well I have given allot of thinking, after the z axis is complete I will have about 170mm of travel. When you look at the different length cutters and so on it’s really best to have about 100mm of travel so you can accommodate different lengths, you also need to think about how far the collet holder will hang out of the bottom.

The other main reason is that the supported rail is only fixed at 150mm intervals, so yes we could cut the rails/travel down but then we would also need to machine the rails with new fixing points (drill, tap).

Given the choice of + 70mm extra vs. the work and loss of cutting everything down I went for having extra travel.

If I was to bring the lengths down I would be wasting money on the Rail, Ball screws and time spent building. I went with a pre machined ball screw for this z axis and that is 400mm long, the design I have gone with allows me to keep the full length of the ball screw and the rails, now if at any point I need to machine a tall part I also have the travel to do so.

Its better to have it, then not. Even if you may never use it !

pavlo
29-07-2009, 09:20 AM
Thanks for the reply Lee but I was referring to tribbles build blog. Your one makes sense to me

Lee Roberts
29-07-2009, 06:20 PM
Why not use some of these (http://www.worldofcnc.com/products.asp?recnumber=76) and pocket the carriers (plates) to get the rails as close together as possible? Archimedes had it right ;)

For me price, £32.20 (http://www.slidesandballscrews.com/15mm-profile-rail-p-128.html?cPath=40&osCsid=3e99e99e9199891be4f9201066aaaa0c) per bearing block ? How dare you :surprised:

tribbles
29-07-2009, 09:07 PM
pavlo:

I think I've got about 230mm of travel (will need to measure it properly tomorrow when I work out how high it needs to go).

It'll be cutting 100mm tall foam to start off with (using a 100mm long tool), so that's the minimum movement I wanted; however, I may be using cutting heads with different lengths, so I wanted the option of going between 40mm and 100mm long cutting heads, which means 160mm travel. I've got the option of even longer tools if I find more suitable ones at a later date.

As Lee said, it's better to have too much than too little (although I don't have the full length of travel for the ballscrew [at this moment in time]).

Kip:

My original plans were to use profile rail for all of the axes, but I didn't want to spend "flipping great wodges of cash" :biggrin: Although I did want to spend more than the minimal amount I wanted on the first version.

The router isn't that heavy (with its base-plate, it's about 6.3Kg; without it I've no idea since I don't have a set of scales in the workshop), and the width of the whole Z-axis mechanism isn't that important (the rails are 1600mm long, the Z-carrier is 160mm wide, so movement is a maximum of 1440mm; 1200mm is the minimum I waned).

The X-axis will be using rails (with a pair of carriages on each side), but that's mainly because the Y-axis mechanism is very heavy (I can see an aluminium version being created at some point in time).

Oh, and I quite liked your thought of a train-wreck - why the edit? :smile:

Lee:

I think mine's a bit more "agricultural" than yours (which isn't surprising as it's being built in a workshop where they make crop sprayers).

templecorran
30-07-2009, 12:49 AM
Sure is agricultural! :clap:

I'm looking at your X axis drive arrangement and wonder if the z carridge will bind in the rails. You have a turning moment there with the screw at the top and the cutter at the bottom, the point of rotation will be somewhere in between the X rails.

When I redid my rockcliff in alu angle I had a load of trouble with bearing blocks binding with the smallest possible movement in the assy. I still have trouble with the Y axis table (fixed gantry version) when I remove the upper part of my table with the T slots in. :redface:

I think you might have a problem with all that space between the uprights of the Z carridge letting the assy flex. To prevent that, you might want to make a bracing plate to stiffen up the four corners of the rectangle formed by the bearing blocks. And while you're at it, put the X drive nut in there too.

Any way. Thanks for reading my 2 cents worth. I hope it isn't needed. :whistling:

John
:geek:

tribbles
30-07-2009, 10:20 PM
The movement of the axis is free - I attached the ballscrew nut to the Z-axis earlier today (just uploading the pictures, and then I'll write it up). There is a small amount of play, but I've not put in all the bolts, and of the ones that are in, not all are tight (I think only 8 are tight, and there's 22 in total).

I know what you mean about the brace; I'll have to see what happens when everything's tight (I do need to put another plate in anyway to push the router vertical; this could be made a triangulation piece).

Note that I'm considering that axis the Y axis (based on the previous version and the orientation of the computer models I've made).

tribbles
30-07-2009, 10:39 PM
You call that axis whatever you like mate....you made it so you get to name it :toot: My X has an X on the back so I don't forget lol .......I'm going to figure how to mark the Y :naughty:

Ahh - excellent.

The axes are now "Jim", "Fred" and "Shiela". :joker:

templecorran
30-07-2009, 11:17 PM
You might want to fix the Mondeo, or fit a tent to the Aeon; heavy rain forecast for the rest of the weekend!

I'm kidding - about the tent!

You know -- the more of this you blog, the more we all want to see Jim Fred & Shelia do their thing!!

John
who is mighty envious of that wonderful workshop you have.:heehee:

tribbles
30-07-2009, 11:28 PM
I'll fix the Mondeo over the weekend; I've got a large gazebo that I used while building the Aeon, and that'll cover me while working on it. The AA guy thinks it's the same problem I had 18 months ago, except the other side (which is lucky, because I bought a pair of 2nd hand drive shafts at the time, and I didn't get around to using that side).

If I can't find the shaft, then I'll have to see what the weather's doing next week; I'm not working on it over the weekend (it's 60 miles away).

It's also not my workshop (which is a shame), but they do have a large amount of steel which I can just pick and choose (the 1500mm lengths of 80mm box section I'm using were described as "offcuts"). And their welders are much better than mine.

I do have a better selection of smaller nuts and bolts though. And especially washers, of which I've only found 4 M8 ones in their entire collection.

Monday I should be wiring it, and Tuesday they should be moving around with any luck...

templecorran
31-07-2009, 08:33 PM
Hi, me again, from soggy wet Carickfergus.

I am amazed at how quickly this is taking shape! and such a monster too. 7ft x 5ft

How free was the gantry to move pushing it at the point where the ball nut is attached? I can't help think you might need another drive screw on the other side ...
It's that dreaded turning moment thing again. and the far gantry is (what?) 1.5m over.

What a busy life you have!! Building CNC Routers, fixing Mondeos, family ...

We are really enjoying this blog.

Ye'll sooon be suukin deezal - keep 'er lit there bi


John

tribbles
31-07-2009, 09:21 PM
(I'm going to give everyone a thanks later on)

It is my second go at it :)

There is practically no turning play in the beam, and there's only 4 of the 16 bolts screwed into the carriages. The distance between each of the carriages is as high as I could get away with (to assist in reducing turning play). I have a slight thought that there could be some skewing in the steel, but I hope I've used thick enough steel (which was the problem I had in the first version).

The level of accuracy I need for a car body is not much - 3mm is probably good enough (most kit cars are built from by hand, and they're accurate to around 10mm). My worst case aiming point is 1mm, but theoretically I should be getting 0.0625mm (half-stepping) - the controllers will do 10,000 steps per revolution, so 0.0025mm is the best with the ballscrews I've got.

My priority at the moment is speed though - the car will be made out of blocks of foam, and if a whole block needs cutting, then I've estimated 1 hour per block for milling. There's about 64 blocks to cut, which means 8 days (although some "blocks" are just thin strips which won't take much time). The original machine we were looking at over 1 day per block.

Mondeo - I found the other driveshaft this evening when I got back, so it should be fixed tomorrow (probably 3ish hours - that's how long it took me the last time, although most of my tools aren't in my possession [either in the workshop, or at a mate's house]).

Family - I'm not married, and I don't have any kids, so that doesn't really take up too much time :smile:

tribbles
07-08-2009, 08:20 PM
Okay, my 10 days are up - and it works (but needs a bit of tweaking).

Here's some pictures of it in action, and also the finished test item.

ETA: Or rather as finished as it could be before the tool fell out (it wasn't in very far into the collet [deliberately to give me some extra length]). That's why the front isn't finished.