Hi all!  
I thought I'd post my new mk2 cnc machine build log, for general interest and comment.  My hobby is designing and building radio controlled model aircraft, so my mk1 cnc machine is ideal for cutting out the parts from liteply, ply, and balsa.  I also use the machine for occasional bigger stuff like 18mm ply, plus engraving wooden signs, but this often shows up the design limitations.  Therefore, time for an upgrade !

Ignore the date on some photos - the battery is playing up.

Before showing some initial progress on mk2, here is a photo of the original mk1.  It had many good features including 25mm hardened steel rails, ball bearing blocks, 16mm ball screw with 5mm pitch, 3Nm steppers on x&y, 1.8Nm on z, and a 33v 10A supply.  However, the plywood construction and long gantry gave rise to too much flex.

The replacement mk2 machine uses alot more aluminium, including profile sections.  Attached is a render of the mk2 machine.  Also attached are a few photos showing the progress of some of the parts on this new machine.  Last photo shows inside the control box, which I intend to carry over to the new machine.

Quick overview of improvements for mk2 are:
1.  y & x axis changed to carraige and rail type bearings rather than unsupported bar, much stiffer
2.  Base frame 80x80mm aluminium profile, much stiffer
3.  Gantry much shorter (spans Y axis instead of X), much stiffer
4.  Gantry uses aluminium profile for top and bottom of gantry
5.  Power cables and data cables run totally seperately (2 sets of cable guides), reducing interference and false estops (hopefully)
6.  Cable guides run under main table, keeping machine width down (almost fits on table now!)
7.  Router replaced with 1.5Kw watercooled spindle, plus Moeller invertor drive (quieter!)

If people are interested then there will be more to follow . . .

I would love to see the progress as you finnish your project. It's looking nice so far.

Sorry about the size of the photos, scaled these next ones down a bit!

Here is the VFD invertor from Moeller.  I decided not to go down the Chinese route for the VFD and paid a bit more. 
Starting at the top and going clockwise are the instructions and mounting bracket.  Next is the invertor itself.  Base model came without the front panel, so I added this as well.  This can be unclipped from the front and mounted elsewhere, then connected back to the invertor using a network type cable.  Next is a white panel to cover the electrical terminals, followed by a blanking plate (bottom right) which was fitted where the control panel is now fitted.

The large rectangular block is a 200W braking resistor.  Without this the spindle can take a while to slow down so I'm told.  Whether this is strictly necessary in a hobby environment I'm not sure, but I was in the spending mood so I bought one.  The silver box in the bottom LH corner is a filter.  This connects between the mains and the invertor, and I think it should reduce interference back into the mains.  Finally the CD containing the manuals etc.  Not pictured is a 4 core shielded cable from the invertor to the spindle.  I'm concerned about RFI from the invertor and spindle (causing false estops or interfering with our house appliances) so have purchased some ferrite clamps for trial over various cables.  I'm also ready to try different earthing set ups between the components, but this will have to be done when it is all running, so is some way off yet.  If you look at the RFI requirements for commercial installations they seem to be quite exacting and I intend to follow these where I can.

Next photo is the spindle.  This is a 1.5Kw water cooled unit, and is from China.  In the end the low price of these swung it!  I've checked the resistance across the terminals and got about 2-3 ohms, so there are no shorts.  It spins freely by hand, and I intend to hook the cooling circuit up to a Thermaltake 850i unit (used on water cooled PC CPUs), unless I see something better.

All for now, still much to do  . . . .

To keep noise from the VFD causing problems, never run the line and load next to each other or in the same conduit.  Use either shielded cable or a metal conduit/raceway for the load from inverter to motor. 

Vince

Hi Vince,

Thanks for the info.  I'll follow that advice.  When I get nearer the stage of running everything together I'll no doubt need more help on RFI.

The old machine has been dismantled and the base for the new mk2 machine has been put onto the bench in it's place.  It's looking good so far, but getting things aligned and square is taking time, but no sense rushing this stage.

In the near future the ball screws will be sent away to have the ends machine to sit in the angular contact bearing block and have a blank shaft machined to attach a coupling to the stepper motor.  Mk1 used unmachined ball screw ends (i.e. threaded right to the end) with a complicated arrangement of thread clamps, thrust bearings and rubber tube!  It worked, but I intend to do better this time.

All for now . . .

Barry

Right then, I've not posted for a while because my fears about the X axis stiffness have been proved right.  I'm using the long 1200mm rails from the mk1 machine, including the pillow blocks which run on it, and they are just not stiff enough over this span.  Although they are 25mm hardened rails, they have too much flex.  I can deflect them a mm or two by hand in the centre, which means that the main gantry will sag in the middle, and also move sideways under cutting loads.

If you look at the render in an earlier post you'll see the long X axis rails.  During the design stage I had considered supported rails, and also the rail and carriage system (which I have on the Y axis), but they are expensive for this length, and I have spent alot on this hobby already!

I have therefore decided to remove the rails altogether and use ball bearings running directly on the top, bottom and outside of the base frame, which is 80x80 profile.  You can see some guide bearings in the original render, but these will be moved and added to to make the system only run on these.  I've checked the profiles with a straight edge and they appear straight enough to do this.  Bearings I have to hand are ID 5/8" (15.88mm) OD 1 3/8" (34.93mm) and width 11/32" (8.73mm) which I think will do the job.  They will mount to the gantry legs.  There will be a top pair to take the weight, a bottom pair to stop the machine tipping, and side pairs (outside) to stop the lateral movement.  They will all be slotted to allow adjustment, and I would like the side pair to be able to apply a preload by means of a thread or similar, so that the gantry hugs the rails.

Before I go down this route, what are the pitfalls of this solution?  Will it just keep racking?  Does anyone have any photos of other machines successfully using this approach?

I'm considering adding a steel (stainless or hardened) wear plate to the top of the profile section, since this would quickly wear.  Any other thoughts?

Thanks
Barry

Sag and lateral location sorted!
Took a bit longer than expected, but I have made the modifications to allow the gantry to run directly on the aluminium profiles.  I'm pleased with the result, especially the side location bearings which I can adjust the preload on.  The gantry location is very rigid - so much better than the old unsupported rail.

Checking for racking
The gantry ballscrew loads the centre of the gantry, so the next thing I was worried about was cutting at the edges and the racking which would result.  Whilst much, much better than my previous machine, if I restrain centre and push the gantry at the edge there is a small amount of movement.  A double ballscrew is an option, but more money.  We'll see how it goes.  By the way, has anyone ever added a pulley and cable system like that used on drawing boards (remember them ?!!) to synchronise the 2 legs?

New renders and photos
The renders show the new bearing arrangement.  A pair of bearings on top take the gantry weight, and another pair underneath hold it in place (not fitted at time of taking photos).  The side location is controlled by a pair of single ball bearings (dia 3/4") retained in a plate.  These press against the outside edges of the base frame, gripping it.  The RH side is fixed to the RH leg, and the LH side is adjustable by bolting in the middle of the mounting plate on the LH leg, and a pair of pressure bolts which press against the outer edges of this plate, pre-loading the balls into the frame.  See close up photos.

Also shown is a general photo with the gantry in place, z axis starting to take shape.

On Tuesday the ballscrews will be machined to fit into angular contact bearings.  Then stepper mounting brackets need to be made.  Starting to come together now.