I ran into the issue you're trying to solve on my build: http://openbuilds.com/builds/c-beam-2-xlarge-cnc-router.4914/

I greatly improved it by removing the MDF platform and replacing it with a 3/8" thick aluminum plate. It's not what I would consider 100% firm and rigid but it seems to be enough better to make a very noticeable difference.

 

I have something similar in the works. I got the idea from Fred:

 

The last picture I uploaded shows then aluminum bed, it's under a mdf spoiler but the ends stick out.

Surprisingly the 30 x 30 inch .375 piece of aluminum was only $65 but I bought it with some other stuff.

 

dogmeatk, I am setting up an arrangement similar to what M90Ranger is doing. My gut feeling (and I have nothing more than that to go on right now) is that the side forces would be mostly absorbed by the Y axis wheels.

 

I can give it a shot, sure. I am using a Dewalt 611 router in my machine; I decided to go that way as I expect that the bulk of the work that I do will be in plastic or wood.

 

Dogmeat, as an update I've not yet got my setup cut; been busy with things for my wife. However I had another thought based on the graphic that is at the top of the page. Rather than attaching a piece of 2020 v-slot to the frame, have a look at the picture in the parts store of how the mini-v gantry is intended to be used. Now envision attaching the 2040 V-Slot to the bottom of the spoil board and mounting the mini-v gantry in line with the router. You'd need to tweak the spacers used since the C-Beam gantry is 6 mm thick and the mini-v gantry plate 0.25", but so long as you can get the bottom of the mini-v gantry plate level with the bottom of the Y-axis C-Beam the geometry should work out, at least in my head it does

 

I was looking at the graphic right at the top. In it you appear to have a piece of 2020 V-Slot attached to the underside of the support for the spoil board, and then a plate holding a couple of wheels mounted on top of the 2060 piece. What I am wondering about with that setup is what it would take to make sure the wheels are aligned with the V-Slot.

What I am suggesting is mounting a 2040 to the spoil board support in place of the 2020 (it would likely be a bit further in towards the C-Beam). Since that is the same height as the C-Beam, if you could mount a plate with its bottom flush with the bottom of the C-Beam then it should be fairly simple to make sure the wheels align with the slot in the V-Slot. I am looking at this from a CBM XL design. I'll try to do up a quick cross section of it tonight and post it.

 

Here is a graphic of what I had envisioned. It isn't the easiest to view (I wish I had done it with the spacers under the wheels being a bit higher).The horizontal 2060 represents the side piece of the CBM. The gantry plate attaches to a piece that supports it; I put a piece of 2020 to support it on the other side, although if the underside piece is thick enough the 2020 may not be necessary. The 2040 attaches to the underside of the spoil board. The 2040 "should" be pretty close to how far you can get the router along the X axis; if it isn't directly under the difference should be able to be ignored.

 

I didn't see the 0.125" plate between the spoil board and the V-Slot in the diagram, and was thinking about making sure that the wheels line up with the V-Slot. My approach was a bit different; rather than split the thickness of the gantry plate I was going to put it all on one side.

 

Nice one, looks great! I built an XL middle of last year, didn't imagine that I would ever cut aluminium with it, and so far I haven't but I'm going to take a careful look at your work later this year when I have time. Cutting wood, I haven't had any issues with accuracy, much better than I had dreamed Images in fact I increased the Z by 4" and still no issues.

 

I've bought a Chinese VFD Spindle that I hope to get 24k rpm from and be able to program speed from vcarve, grbl, xpro etc... I've some fun ahead.

 

Hi Gray,
Nice to hear from you. Yes the VFD Spindle is sitting around my workshop waiting for some bits to arrive before I can get on with that. There seems to be plenty around regarding running the spindle from the Huanyang Inverter, but not so much about the digital connection from xpro to inverter to convert iso Scodes to actual spindle speeds. That said the inverter manual is a large Chinese translation so I'll have to put my best reading glasses on! I'll let you know how I get on.
Cheers, Geoff