Oh crumbs. Here we go!

[hoppo]

I've been wanting to build a CNC router for some time now, and after joining the mycncuk forums some months ago haven't had much time to get my build going. Well, now the days are drawing in fast and the fact that it has been throwing it down with rain all weekend, I got a bit bored so I contacted Chai (from the recommendations found here) and he put together a good package for me, and I've only gone and blummin' ordered it. :congratulatory:

So this is what I have winging it's way via fed ex to me.

3 x supported linear rails SBR20-300/700/1100mm rails with SBR20UU bearing blocks.
3 x ballscrews C7 RM1605-350/750/1150mm with ballnuts and standard end machinings.
3 x sets of BK/BF12 bearing mounts
3 x ballnut housing brackets
3 x flexible couplings 6.35mm*8mm

Now during the wait on delivery I'm going to get into CAD, and try to put what's in my head in to some sort of plan as to what I'm going to do with them when they arrive. :nevreness:

Thanks for all your advice so far, I'm sure Ill be back for more soon.

[motoxy]

Now you've gone and done it!!!!. Just remember to enjoy it.
bruce

[Tenson]

If it is anything like my build, it will be a challenge, but rewarding :)

I think if you a 1100mm axis, you will want to use 2x ballscrews on it, one next to each linear rail. I guess you are thinking of putting one ballscrew in the middle, but this can cause the gantry to wobble side to side since it can resonance and flex like a trapezoid (since you can't brace the gantry in all directions), so it is better to hold it at both ends.

Where in London are you? I'm near Lewisham.

[hoppo]

Cheers Motoxy and Tenson,

Not sure what I'm going to do with it once I've built it but yep I'm sure it'll be a challenge, but hey that's half the fun..

I had thought about two ball screws to move the gantry along the long axis, but sure I read on here somewhere that anything with a gantry over 750mm would require two and I could probably get away with only 1 if my gantry wasn't that wide. Accuracy isn't 100% for me at the moment as it's my first machine, but am willing to consider it, especially as the chinese ball screws are relatively cheap.

Incidentally can anyone clarify the axis's? I've seen so many different ways of representaion of X Y and Z. The Z is obvious, but does the Z connect to the X or Y axis? I assume that from the front of the machine the X axis moves the Z along the gantry from left to right as you look at it, and the Y axis moves the gantry closer or further away from you as you look at it. (Thought this should be a fundamental that I need to get right at the start ;) )

I've finished an initial design of my Z axis which seems to look ok to me, but being new to this thought I'd best load it up and let you guys have a look. Please offer up any improvements that you think I need to do.

Oh and if anyone needs any easy to use design software and uses a mac, I recommend Inventor Fusion which is what I created this in and it's free! Last week I had no experience of CAD packages and I've already managed to create something useful, and it animates! Now if I can figure out how to upload it....

Tenson, I'm in Camberwell, good to know I have a CNC neighbour. :)

[WandrinAndy]

Incidentally can anyone clarify the axis's? I've seen so many different ways of representaion of X Y and Z. The Z is obvious, but does the Z connect to the X or Y axis? I assume that from the front of the machine the X axis moves the Z along the gantry from left to right as you look at it, and the Y axis moves the gantry closer or further away from you as you look at it. (Thought this should be a fundamental that I need to get right at the start ;) )....

Z connects to Y which connects to X.

X = Bed/Base of the machine
Y = Across the Gantry
Z = Vertical Slide

Some folk swap X and Y around, but generally not on this forum.

[hoppo]

Brilliant, that clears up my axis issues. :)

I've managed to figure out how to create a picture from my CAD package, so here it is. Z axis version 1! To be honest I'm not 100% happy with it, mainly due to the width of the plate that the spindle will attach to, I'll have a re-think, but I think I have good base to work from.

[irving2008]

Usual comment. Mount the bearing blocks as far apart as possible on the fixed part and the (longer) rails on the moving bit. Gives more rigidity as the bearing blocks can be further apart for the same amount of travel and relatively small increase in cost.

[motoxy]

As far as I can make out the americans have the z connected to the x while we have z connected to y.

[Ger21]

Not for this American.:smug:

Normally, the longer axis is the X axis, which typically leaves the gantry as the Y axis, which is where the Z is attached to.
I believe that some people call the gantry the X axis because they stand facing the gantry, so the X axis moves left to right. Having the X axis move left to right is desirable, as that's how you'll see it in your CAD and CAM software. I stand on the side, with the gantry moving left to right.

[hoppo]

Well I've managed to restructure my Z Axis. I've had to compromise on the travel to space out the linear bearings a bit more. But hopefully it should now be a bit more rigid. I've also swapped the bearings onto the stationary back plate and moved the rails onto the moving part. I've also been able to shave about 20mm off the width too. It's now approx 160mm wide by 330mm high with 150mm of travel. I don't really want to go less than 150mm of travel. I suppose the I could always add longer rails at a later stage if I ever really need that extra travel, and could also then move the bearings even further apart. Lesson number one learnt, design the bloody thing before you order the parts!

Now to start looking at connecting it to the Y axis. Thanks all for clearing that one up.

[hoppo]

Blimey a big thumbs up to Chai, :thumsup: unbelievable service, guess what turned up on my doorstep this morning, all the way from China, less than a week after ordering?
7146
I was like a kid at Christmas this morning. Only problem now is that I was expecting it to take at least a couple more weeks to arrive and hopefully I'd have been well on the way to completing some sort of design! Which I haven't!

What I have managed to do is think a bit about my YAxis, and have come up with a couple of ideas, nothing revolutionary and not anything that you'll not have seen before, and this is where the questions start to flow.

Up until now I've been hoping to use aluminium extrusion, for ease and simplicity, but having another browse through the other builds got thinking about using aluminium plate. Is there any advantage of using plate? On my second design in the pics attached to this post I think this would work better with aluminium plate, instead of using the extrusion and then I was thinking I could move the ball screw onto the same side as the rails. My only concern is the extra cost involved in getting the pieces made up. What are the comparative costs in buying a 750mm length of extrusion and say a 750x300x20mm plate? Is 20mm thick enough as a back plate for a Y Axis?

Of my two initial ideas below I'm leaning towards the first one as it just looks stronger, but am all ears as to any thoughts, advice on improvements, amendments, materials etc etc.

[Jonathan]

From the picture, unless it's a twin start screw, it looks like you got RM1605 for the longest screw, not RM1610. Assuming the ballscrew is 1150mm long, as per the original post, this will limit you to about 8.5m/min due to the critical speed. That speed is plenty, but I doubt you will get it with 3Nm motors unless you have a 2:1 ratio between the motor and the ballscrew to get 10mm 'effective pitch'.

20mm is plenty thick enough as a back plate for the Y-axis, so long as you have some box section or extrusion bolted to it to increase the strength parallel to X and the torsional stiffness. It will be much more rigid if you have the ballscrew directly between the rails, so on the same side is good.

[D.C.]

I'm wondering the same as I design my gantry hoppo, a top & bottom mounting for the rails I think would provide a slimmer gantry so the router bit will be closer to the gantry but it also seems to sacrifice some gantry stiffness when compared to side mounting the rails.

Assuming all other materials and dimensions etc stay the same, does a top/bottom mounting of linear rails give better real world performance or side-mounted ones?

[hoppo]

Hi Jonathan, thanks for your post. It is indeed a single start RM1605 and 1150mm in length. I was planning on trying to drive it directly with a Nema 23 3Nm stepper but after your post I'll now somehow implement using pulleys to drive the x axis. So much yet to learn, but hey that's half the fun! On the subject of ratios will the RM1605's be ok for the shorter Y and Z axis's being driven directly? I saw a post on here that recommended this site Bearing Station - Adhesives, Bearings, Belts, Chains, O Rings, Oil Seals, Pulleys & Sprockets (http://www.bearingstation.co.uk) so I'll start to have a look at what would work, unless you've got any other recommendations. I was also beginning to think of using two screws, instead of one along the x axis. Would this make a difference or if I did go down this route would you still advise on using a 2:1 ratio but on both screws?

I've decided to definitely use aluminium plate for the Y axis, it just looks and feels more solid, and the prices are comparable, having looked on aluminium warehouse. Would I need to use ecocast for the backplate or would 6082 do the job? I'll probably be looking for someone to machine the sides for me, and then I'll drill it tap it and bolt it together. Anyhow I've attached my idea for the Y axis. As usual any comments appreciated. And also as usual many thanks for all the help so far. I'm sure there'll be loads more questions.

[martin54]

Learning myself so could be wrong but from what I have seen & read ally plate on it's own won't be enough for the gantry & you will need some extrusion or box section fixed to it.

[hoppo]

Hey Martin, yes I think you're right. Jonathan also mentioned it in an earlier post on this thread. I plan on bolting a bit of steel box section across the backplate of the gantry to give it that extra rigidity. Do you know if that applies to the sides too?

[hoppo]

Was just about to order some aluminium plate to start building my z axis. Had a bit of a panic. Is 15mm thick enough or would I be better off using 20mm for the sliding part and the back plate?

[Jonathan]

Do you know if that applies to the sides too?

It would certainly help, but even better would be to lower the gantry as much as you can and have the tool operating below the level of the X-rails.


Was just about to order some aluminium plate to start building my z axis. Had a bit of a panic. Is 15mm thick enough or would I be better off using 20mm for the sliding part and the back plate?

15mm is OK, but most use 20mm as you don't gain much by going above 20mm. In the whole scheme of things it's not going to cost too much for the extra 5mm...

In answer to an earlier question - standard 6082 is fine for supported rails. Tooling plate is only really worthwhile for proper linear guides.

[Robin Hewitt]

No easy way to say this but your Z axis looks particularly -er- "unlikely". I think you may be just about to make a big mistake.

Basically, the fixed part of the Z above the bed should be at max material thickness + a tadge for clearance.

The Z axis travel should be max material thickness + max tool length + a tadge for clearance.

Do not over-estimate these maximums because you think it will limit the machine. Over-estimation leads to overhangs that will weaken it so much it may not want to cut anything harder than cheese.

OTOH I haven't read the whole of this thread. If you are planning on cutting foam patterns with a 6" tool it looks fine and dandy apart from the minimal vertical separation on the Z blocks.

[hoppo]

Cheers for the tips Robin, I'm still a while off finalising the plans yet and very much in the design stage still, and when I get round to picking up the design again I'll try to implement your good advice. In the meantime I have made progress in other areas. Namely I've managed to pick up 3 drivers from Wantai (DQ860MA) and built myself a power supply to power them which I'm quite proud of. Testing it out with a few Nema 17 steppers that I have for a different project all seems to work rather well. I've only been driving them back and forwards using an arduino to supply the pulses so far, but all seems good so far. I'll step up to using that old pc that I have lying around eventually but I'll need to find a hard drive for it first and install Mach 3 or similar. To be continued....

[hoppo]

The Z axis travel should be max material thickness + max tool length + a tadge for clearance.

Picking my design up again as I have a bit more spare time on my hands. How do I work out my maximum tool length? Would this be the amount of end mill sticking out of the spindle? I'm hoping to get 150mm material thickness, is this wishful thinking? Going from what Robin suggests I take it therefore that I'll need 150mm from bottom of fixed z axis plate + tadge (say 5mm) assuming endmill protrudes 50mm out of spindle does this mean I'll need 205mm of travel?