View Full Version : Need 6' x 4' bed, 4 axis router design.. Help..!

13-09-2010, 07:34 PM
Hi all..

Well. finally got fed up of mdf, angle iron and skate bearings..! But did manage to make a bit of money with it, problem is, I am now being asked to do some serious (for me) 3d stuff and my home-brew heath robinson setup will just not cut it (literally).. Just not accurate enough plus I want a fourth rotary axis to engrave pool cues..

Starting to get dizzy looking at the enormous amount of designs out there so thought I would ask you guys what you would recommend.. Needs to have a 6' x 4' bed which I would like to be T-track ally, which could be removable to have an open bed..

Also would like to mount a chuck at one end of the x-axis and a sliding tail stock at the other so I can do columns and pool cues...

Will make the base welded steel and was thinking of ally for the gantry and open bed cross members..

My current 5' x 3' cnc can machine ally (carefully!) so I could make parts for the new one as needed..

Anyone point me in the right direction?

Many thanks..

13-09-2010, 08:30 PM
Hi Steve,

Joe's 4x4 is not quite 6x4 but gets alot of good reviews, and could possibly be scaled up. Still uses MDF, but also uses HDPE and bolted together 80x20 ali - might be worth a look to get some ideas:

Or there's the Mechmate with the double gantry which has a good following. Mostly welded steel.

14-09-2010, 06:02 AM

Thanks for the pointers.. Joe's cnc is what I am trying to get away from, really dont like mdf and I am being asked to do some very robust stuff now which might be a bit too much for an mdf machine..

Mechmate.. Wow.. That is some machine, but also some level of manufacture.. Would love to do that but it involves a lot of bending metal and laser cut parts which has got to jack the price way up, especially in this country.. Doing it in the states where costs are much lower no doubt makes it more affordable.. Might try this if I cant find a simpler way to go...

Still trolling though many of the designs on this forum, man there is some nice machines here.. Just toying with the idea of making a hybrid of some of the designs I have seen, although the mechmate way of using a spring loaded rack and pinion motor mount looks a real winner...


14-09-2010, 07:22 AM
Steve ... the Mechmate was born in South Africa,but I gather the 'owner' (Gerald) handed it off when he could no longer afford the time to devote to the User Forum. There are a couple of companies here in the UK who can cut and bend the parts needed for a pretty good price ... the thing that puts many off is the cutting and profiling of the rails ... so a lot of folks fit THK style. The forum is a HUGE goldmine of info and the users are very helpful. If I had the space I would build one ... (over time)

Have a look at http://www.data-cut.com/ it can be modded to fit the size you want ...

14-09-2010, 07:28 AM
Hi Steve,

Here's a sketch I was working on a while ago. Double X axis motors, all ballscrews, profile rail Y and Z, supported rail X. Gantry sides are RHS with spacer block connection to gantry. Y axis ballscrew is underneath to give a low bending moment on the gantry sides. Disadvantage is risk of collision damage and swarf ingress (needs shield).
Maybe some ideas in there.

14-09-2010, 09:41 AM
Hi All..

Mechmate is a very attractive option, I am currently going through the huge amount of data on their site to see if I can do the mods I had in mind.. As this will be my 3rd machine, I have learned some (sometimes painful) lessons that I want to correct for this build.. I have been a cabinet maker but now have a back injury that can often make this type of business difficult.. The idea of a cnc was to allow me to basically sit on my ass most of the time, either drawing on the computer, or being hypnotized by the cnc strutting its stuff..

Things I really want on this build are;

Open bed (with a T-track bed that is removable)
Rack and pinion drive (dual for 2 motor x-axis)
Supported rails for x and y axis
Vacuum bed (using T-track?)
Tool changer

If I can get some of the Mechmate parts done at a reasonable price, and can incorporate these mods, I will almost certainly go that route..


I agree with your design comments.. My first cnc was an endless battle with dust on exposed lead screws and bearings, really have to have a very efficient dust extraction system, or even better, tuck all the exposed stuff out of the way..


15-09-2010, 09:31 AM
Hi Wobblybootie (nice handle)..

Have looked at that site, but those cnc'c seem to be the type with the gantry going under the bed.. That would not allow me the open bed style machine I want so cant go down that route.. Am going to post a thread or two at mechmate and see if anyone has tried a flavour of the design in the style I want, if yes then I may get some pointers that way..


15-09-2010, 12:29 PM
Florins machine had some nice features:

16-09-2010, 01:27 AM
Hi routercnc...

Thats it!!!! That is exactly what I am looking for... Thank you sooooooooo much!!! Will jump on him and pick his brains to bits... Yipppeeee!!!

Steve... :-)

16-09-2010, 08:31 AM
Hi Chip...

I have found a few people on the mechmate forum from the UK who have built them, including a guy that made one in his garden in the open air..! I agree that the mechmate is an awesome machine, but also quite costly although having all those people on that forum for support will almost certainly save a few quid and loads of head-scratching..

But florins machine is, at first glance, exactly what I want so I will try and contact him and see what is involved in making that.. Only thing that makes me hesitate is he has used lead screws on his x-axis and I had always thought that was a no no with that sort of length.. Remember reading many posts about 'whip' but florin must have found a way round it..


16-09-2010, 11:22 AM
there is nothing intrinisically wrong with long lead screws as long as they are correctly supported in tension by appropriate supports and of sufficient diameter. Making them larger pitch reduces the revs needed which also alleviate whipping. The issue then becomes one of inertia; a long, thick screw has high inertia needing lots of torque to accelerate it, but again that can be offset by a larger pitch...

16-09-2010, 12:50 PM
I believe that whip occurs when the shaft rotation speed equals the natural bending frequency. So a ballscrew with a 20Hz nat frequency would whip at 1200rpm (=20 Hz[cycles per second] x 60 seconds per minute).

Natural frequency is related to SQRT(k/m) where k is stiffness and m is mass. Therefore in principle a hollow ballscrew, with slightly lower stiffness, but much lower mass, would give a high frequency and could rev higher before whipping, all else being equal. It would also have a slightly lower inertia, allowing slightly higher accelerations etc.

But never seen a hollow ballscrew, so either they are rare, or not worth the expense.

16-09-2010, 02:32 PM
Wouldn't a hollow ballcrew have a greater moment of inertia, which would affect the acceleration?
Otherwise it sounds like a good idea, if you've got a nice long drill :lol:

16-09-2010, 04:25 PM
Hi All..

Cant see why that is not possible.. Would just need to have one thread cut and ground from a high carbon steel tube... I think Jonathan, (aint no expert) that the inertia is a function of mass, which would be considerably less with a tube...


P.S. Am starting to wander away from rack & pinion..

16-09-2010, 04:45 PM
I'm no expert either! But the moment of inertia, I, of a tube is 0.5*pi*(R^4-r^4), where R is outer radius and r is inner radius. From that you can easily see that the further out the mass is distributed the higher I gets. In a tube all the mass is far out...
I'm not sure without working it out properly - what diameter solid ballscrew would be equivalent to what diameter tube, that's the question.

16-09-2010, 04:55 PM
Hi Jonathan..

This may be straying into the realm of the theoretical, not sure such a thing as a hollow lead screw exists.. But I do see your point although the inner mass of a solid would still require movement and account for some of the power required to rotate it.. Just checked florins posted specs and he has used 25mm dia leadscrews, and 2 nema 34 12N motors to power them.. Those motors would probably pull my car up a hill..!


16-09-2010, 06:11 PM
We're deviating from Steve's original topic, but you need to compare like with like diameters. If you start with a solid 16mm dia ballscrew, and compare it to a 16mm dia hollow ballscrew (theoretical) with say a 10mm dia bore, you will find it has less inertia. The reduction is only small because you are taking it away from the middle, where it has less effect.

In your equation R is fixed at 16mm and r varies from 0 to 10mm (in this example). So R^4 - r^4 gets smaller, and the inertia goes down.

16-09-2010, 06:32 PM
In your equation R is fixed at 16mm and r varies from 0 to 10mm (in this example). So R^4 - r^4 gets smaller, and the inertia goes down.

Yes, of course. I was tying to compare like with like stiffness not diameter since I was assuming you'd use a bigger screw to compensate.

16-09-2010, 08:19 PM
the fallacy is that a 20mm screw on a 5mm pitch has a minor diameter of 15mm which means the inner bore couldnt be more than 10mm, so the stiffness would be the same as a 19.5mm solid screw and the inertia similarly.. i.e. boring a 10mm hole reduces the inertia by 7%... hardly worth the effort.

On a 40mm dia, 40mm pitch screw the thread minor dia is 35mm so a 25mm bore might be possible... about a 16% reductoin in inertia... might be worth it in some scenarios

but overall, not worth the trouble, which is why you never see them

16-09-2010, 09:40 PM
Hi Irving,

The inertia benefit is marginal, but it's interesting to compare a 20mm dia solid with a 20mm dia tube with 15mm bore which shows ~28% increase in natural frequency, so potentially a 28% increase in rpm before whipping . . .

16-09-2010, 10:26 PM
I would have thought that the hollow section, being less stiff than the solid one would start whipping at a lower speed.... like for like on external diameter. I am sure the critical speed is related to the stiffness in some way.

16-09-2010, 10:51 PM
I am sure the critical speed is related to the stiffness in some way.

That would seem logical since tensioning the screw sort of makes it stiffer, and also raises the critical speed.

17-09-2010, 08:04 AM
Critical speed is related to stiffness, but dynamic stiffness not static stiffness. The difference is the mass term. Nat freq is related to SQRT(k/m), so although the stiffness drops for a hollow section, the mass drops more. Therefore the frequency increases, as does the whirl speed.

17-09-2010, 09:34 AM
Cant help thinking of my ex-girlfriend now... One christmas she put on over 2 stone in weight... Then tried to startve herself slim but after a couple of days kept whining on about feeling hollow...

Whether she was full, or hollow, there was definite sagging in the middle.. And it did take a huge amount more force to get her moving...!!

All this interesting stuff about hollow lead screws aside, I think i had better check the real critical aspect to this.. Cost! Had originally thought of dual drive rack & pinion, which would have allowed less powerful motors (was looking at http://www.oriental-motor.co.uk/uk/sites/produkt_4.php?ida=3&idser=69&idpr=1240&sgid=72 which is a mechmate european recommended motor), but the rack itself look expensive.. If anyone can point me to good sources for R&P, and I am assuming Zapp is as cheap as anyone on lead screws and ballnuts..?

Better get my calculator out and work out some prices..


17-09-2010, 10:17 AM
You could always consider chain drive...


A chain and sprocket solution could work out very cheap to do.

17-09-2010, 10:39 AM
or belt drive?

17-09-2010, 11:37 AM

Am using a chain drive now, which works ok, but as this is my 3rd build I was looking to upgrade wherever I could.. Have a feeling that motors and electronic are going to be the biggest buy but those will be very dependent on the type of drive mechanism. One thing I will say about chain drive, get a good quality chain! I was shown the difference between a cheapo one (which I have) and a decent make and was very surprised at the accuracy difference.. 2nd thing is, keep them tucked away cause they attract dust like flies to poo..


17-09-2010, 12:06 PM
Why not use ballscrews on the X-axis, but spin the ballnut instead of the screw. I know it's a bit more work to set this up, but it removes the resonance issues and greatly reduces the rotating mass...

I'm doing this on my router, just with threaded rod not ballscrews:


One of those on each side of the gantry.

17-09-2010, 12:11 PM
Hi Jonathan..

That looks well funky.. I would like to see the completed unit included how you attach the motor..


17-09-2010, 12:56 PM
I'm attaching the motor just to the side of the gantry on a sliding mount thingy....much like I have on the Y-axis (see by build log). I'll mill a slot in the gantry side for the belt to pass through. The outer rings of the tapered roller bearings will be held in some 20mm aluminium plate, with bolts between the plates to apply preload to the bearings.
It'll be a while until I make that.

Note how I've made the delrin (might use brass...) nut in 2 parts, squashed together to eliminate backlash. That's the theory anyway.