Second CNC table build, here we go again....

View from front looking at current gantry design.

http://www.mycncuk.com/attachment.ph...id=15648&stc=1

This is for the open-bed style table with the longer gantry arms. The linear guides can be easily covered from the top.

The gantry beam will pass through the side support plates and be welded up and gusseted inside with some nicely cut parts, the outer face of the vertical supports can be strengthened with a couple of ribs to resist side-sway.

Mounting the support rails is harder on the open bed style though, I was going to bolt them on but this style really wants a box frame at bed level.

Every action has an equal and opposite reaction. If you apply 300N accelerating force to the cutting head, the gantry will push sideways with 300N.

Yes I can remember that from school ;)

But those 300N can be countered by applying gusset plates and ribs where needed, most tend to throw a slab of alu at it but that might not be totally needed i think.

Or you could move the rails up closer to the gantry they support. The gantry only needs to be 2" above the bed, doesn't it?

In theory yes but my Z-axis needs a bit of clearance as its a pretty heavy built unit. I am trunk not to lift the rails - that was design idea one, but it means you do not have a clear open bed - you have rails either side which makes loading harder especially if juggling a bigger sheet on your own.

My high-rail option is sketched out earlier, it had no risers at all on the gantry - the beam was right down on the carriages.
http://www.mycncuk.com/attachment.ph...id=15649&stc=1

Quote:

---

Originally Posted by Robin Hewitt

Or you could move the rails up closer to the gantry they support. The gantry only needs to be 2" above the bed, doesn't it?

---

To add scale to my poor sketches, the bottom of the gantry beam is 140mm above the bed surface, not too bad??

Ok, stuff is on order.

The final result is 50x50x2 for the gantry beam and frame legs, 50x80x2 for the motion support/bed frame.

The final capacity is going to be the half sheet 1250x1250 / 4'x4' size.

Direct drive rack and pinion with a view to change to belt reduction drive or even ballscrews later on.

Good man thats what like to see no messing around.:thumsup:

Quote:

---

Originally Posted by JAZZCNC

Good man thats what like to see no messing around.:thumsup:

---

Oh I can mess around, to a point, then I just dive in and go for it.

http://www.mycncuk.com/attachment.ph...id=15652&stc=1
The top box is the gantry beam, 50x50, the motor sits down the bottom on a plate that swivels on the point shown. It will be spring loaded but restricted so it cannot jump out of the rack and chew it up.

A couple of 75mm gusset plates under the beam on the inside and a rib or two on the outside should stiffen things up a fair bit.

Motors, controllers and rack arrived today :)

Steel came Friday, nice and straight and after a few stress tests (me standing in the middle of a 7m long beam) I realised just how rigid this stuff is http://www.practicalmachinist.com/smile.gif

Construction question:

I was thinking of welding on some 100x50x10mm pads under each corner of the motion frame and similar to the top of each leg, then bolting the two parts (welded motion frame/bed and welded leg assembly) together, just to give the opportunity to shim out any variation in leg length etc. Good idea, bad or not worth it???

Next, the bed/motion frame - should I attempt to weld it up as one piece, or weld closing plates flush into the ends of the cross-members, drill/tap and then bolt together?? If yes then how to weld the closer plates in - only way i can think is to heavily bevel the plates then insert and fill bevel with weld tying into the tube sides??? Plates would be 46x76x10mm.

http://www.mycncuk.com/attachment.ph...id=15664&stc=1

Got the metal cut today...
http://www.mycncuk.com/attachment.ph...id=15690&stc=1
Ends squared up and deburred, also cut,drilled and tapped the mount plates for the motion frame and the insert plates for the cross-beams. All ready for a welding session now. http://www.plasmaspider.com/images/s..._e_biggrin.gif

I'm fitting castors and screw-down feet so I can move it about and level it up when assembled.

While I'm gathering info, extraction....

On my little table I built an air-box that sits along the front of the table between the now unused water tray and the slats...
http://www.plasmaspider.com/download...p?id=36632&t=1

In the front i cut a slot with an area bigger than the pipe cross-section and also slightly tapered towards the middle in an effort to even out the air flow. It works well as long as the unused parts of the table are covered - i get no muck in the shop at all.

It's connected to this radial blower...
http://www.plasmaspider.com/download...p?id=36631&t=1


I know I can get more out of it by dumping the corrugated pipe and fitting solid duct but it's not going to do a table with an area four times as big.

I don't have any spare power capacity so simply sticking a big-ass blower in there won't do it, need to get smart here. I could maybe run two of these blowers, they are pretty cheap to buy and with external motors they are immune to metallic hot muck.

I had thought about a zoned system - a four zone would work ok with the same blower. Question is, how would I do it???

Any smart ideas on extract??

Don't want water - its only half as effective and makes the machinery in the shop rust.

Welded the end plates into the cross-beams...
http://www.plasmaspider.com/download...p?id=36647&t=1


The leg joint plates onto the main beams...
http://www.plasmaspider.com/download...p?id=36648&t=1


And the foot plates onto the legs...
http://www.plasmaspider.com/download...p?id=36649&t=1


Fairly confident in the welds, did a quick destruction test on scrap first, nice and hot concentrating on the thicker parts and washing into the 2mm tube. Joints are ground flush now. Casters will be added soon, to the side of the feet, the feet can be screwed down to lift it off the wheels.

Got two welded side frames - standing side by side, there is no visible warp, I'm pretty chuffed with that as it was a concern. I levelled the frames between the bench and sawhorse using a spirit level http://www.plasmaspider.com/images/s...con_e_wink.gif wedged them up, shimmed the bolting points between frame and top beams then tacked the outer and inner faces, seam welded the lower and upper faces fully, then seam welded the outer face, inner face is left tacked as there is a cross-bream to go in.
http://www.plasmaspider.com/download...p?id=36759&t=1



Its a bit bigger than my existing one http://www.plasmaspider.com/images/smilies/icon_eek.gif http://www.plasmaspider.com/images/smilies/icon_eek.gif Looks good though.
http://www.plasmaspider.com/download...p?id=36760&t=1


Upon assembly i realised i could not weld in the lower cross-beams as planned, I'm not really up to vertical mig yet, cannot get it on the bench, and the underside welds could not be done at all. Answer - make some more bolt fitments out of 10x50mm bar, weld them on and bolt the lower cross-beams in, this will make disassembly easy too.

Good welding, you deserve to succeed :thumsup:

Quote:

---

Originally Posted by Davek0974

how to weld the closer plates in - only way i can think is to heavily bevel the plates then insert and fill bevel with weld tying into the tube sides?

---

If you are trying to weld thin steel and thick steel together you need to run the weld pool down the thick piece of metal and then quickly move across to the thin metal, back to the thick metal, down a bit across just enough, back again to the thick piece, down again etc etc

If you try running a bead equally down the joint, the thin metal will blow out or the thick metal will not get proper penetration.

A CNC machine is never going to generate enough force to cause a properly formed weld to fail, it will probably never get above 1% of the failure point. But if you decide to do nice 'stack of pennies' welds that would make an instructor happy you will be introducing a lot of heat into your frame that will cause stress and distortion over time.

Clamp it, tac it, clean it, back weld an inch at a time and drink lots tea so the heat doesn't build up.

This is an awesome video for demonstrating welding heat distortion, well worth 10 minutes.


http://www.youtube.com/watch?v=0pf-qQDslhU

Amazing, thanks.

Is it the same for Mig though which is a lot faster, I guess it would be???
I would love to be able to Tig like that ;)

Quote:

---

Originally Posted by D.C.

If you are trying to weld thin steel and thick steel together you need to run the weld pool down the thick piece of metal and then quickly move across to the thin metal, back to the thick metal, down a bit across just enough, back again to the thick piece, down again etc etc

---

Exactly how i do it, always bias towards the thicker part and set power for the upper end of the thinner part.

I had to cut a 10mm plate off the end of a 50x50x2mm tube yesterday as i welded it on 90deg out :thumbdown: and i was impressed with the penetration i was getting, took some cutting to split the joint.

Had a quick mess about with the Tig, i was semi impressed but seems a lot hotter, probably moving too slow?
http://www.plasmaspider.com/download...p?id=36791&t=1

I'll stick with the Mig http://www.plasmaspider.com/images/s...on_e_smile.gif


Got the lower cross-beams fitted, then after a few tests it seems that a good way to fix my spacer/support strip was to use good ole pop-rivets, worked nicely. Also managed to get one of the racks fitted.
http://www.plasmaspider.com/download...p?id=36792&t=1

Ok, linear rail came today http://www.plasmaspider.com/images/s..._e_biggrin.gif

Using a taught line to check my beams, I have a slight bow in both sides, as you look at the table they would bow towards each other at the centre, the amount is about 1 to 1.5mm on each side so thats pretty bad and would give me a 3mm variation from centre of motion to either end.

Question is - how best to tackle this????

Quote:

---

Originally Posted by Davek0974

Ok, linear rail came today http://www.plasmaspider.com/images/s..._e_biggrin.gif

Using a taught line to check my beams, I have a slight bow in both sides, as you look at the table they would bow towards each other at the centre, the amount is about 1 to 1.5mm on each side so thats pretty bad and would give me a 3mm variation from centre of motion to either end.

Question is - how best to tackle this????

---

You can always use the epoxy method that is quite often used on here.

I had thought of that but seems a possible method may be to fit the rails as-is and then fit a stretcher bar under the slats to force the beams out a little, 3mm should be easy.

Could even use the middle slat support possibly, not got that far yet though.

Got the Y axis motion fitted and the gantry supports tacked up.

http://www.plasmaspider.com/download...p?id=36922&t=1

http://www.plasmaspider.com/download...p?id=36923&t=1


Even without the beam fitted the side supports are very rigid, most encouraging.

Got the gantry supports welded up - very little distortion which was great, impressed myself with that one http://www.plasmaspider.com/images/s...con_e_wink.gif

One thing I can now say is, that gantry beam is damn rigid, i put a dial indicator between the gantry and side beam and it's practically a solid object http://www.plasmaspider.com/images/smilies/icon_eek.gif http://www.plasmaspider.com/images/smilies/icon_eek.gif

http://www.plasmaspider.com/download...p?id=36926&t=1


http://www.plasmaspider.com/download...p?id=36927&t=1


http://www.plasmaspider.com/download...p?id=36928&t=1


Also got the X axis and THC mount welded up, came out nicely.

One question - I had to move the x motor rearwards, placing the rack on the back face of the beam - no biggy but, I had forgotten the rack bolts go across the rack face so are now screwing into nothing. Should i fit some short support angles like 10x20x4mm 100mm long and welded to the beam, and then mount the rack to them????

Or something better???

Quote:

---

Originally Posted by Davek0974

One question - I had to move the x motor rearwards, placing the rack on the back face of the beam - no biggy but, I had forgotten the rack bolts go across the rack face so are now screwing into nothing. Should i fit some short support angles like 10x20x4mm 100mm long and welded to the beam, and then mount the rack to them????

Or something better???

---

Looking good Dave, I presume that you mean Z axis motor not X? I'm not sure how much force the stepper will exert on the rack but I would think a long piece of angle tacked on to the back beam would be ok to mount the rack, just have to be careful that you don't distort the beam welding it but you seem to have done a good job of not distorting things so far :applouse:

No definitely X-axis, the Z-axis is mounted on the THC unit, X-axis on the carriage.

I have opted to try the 3M VHB double sided tape for this - no distortion and easy to fit, seems to be getting good remarks on other builds.

Hah...i meant Y axis not Z but I guess you call it the X....confuses me everytime!!!!

Ha :)

Depends on the table i guess, my one is x=left/right, y=front/back, z=torch

Reclaiming some motion distance....


After building the base frame and motion, I decided to build a new floating head using Hiwin and a single carriage - pretty much a standard design. However, this moved my torch 30mm closer to the front of the table than I had planned which left very little over-travel on a full 1250mm sheet - not good.

The first pic is a cross-section of the average torch mount with the Hiwin at the bottom and the circle being the torch clamp.

http://www.plasmaspider.com/download...p?id=36943&t=1


Now, as I am using wide/stable Hiwin rail, could this idea work?? It regains a valuable 30mm of spare travel on the Y axis.

http://www.plasmaspider.com/download...p?id=36944&t=1

Here we go, down to 27mm distance now, thats a little less than my existing unit http://www.plasmaspider.com/images/s...on_e_smile.gif

http://www.plasmaspider.com/download...p?id=36960&t=1


The side effect is that to get the torch back on centre, I have to shift the Hiwin carriage to the left a fair bit, luckily this will be over the machine side beams and will not affect anything at all - perfect http://www.plasmaspider.com/images/s...on_e_smile.gif

That should be fine Dave I would have thought, it's not carrying any weight and ther are no forces acting on it after all. For complete ease of maintenance could you not turn the torch mount through 90° (gap facing the front) and use one of the fixing bolts (now coming in from the right and tightening into tapped holes in the 5mm steel plate) to tighten the clamp? I'm not sure from the drawing if there's ease of access to get to the torch mount?

Yes, but then there would be a risk of losing the torch alignment when the torch clamp is released to remove the torch. Hopefully with the clamp mounted firmly from the base, it will stay fixed once aligned.

Got the floating head mount fitted to the Z-axis, probably work better when i fit a means to stop the carriage and torch falling off the bottom http://www.plasmaspider.com/images/s..._e_biggrin.gif

http://www.plasmaspider.com/download...p?id=36980&t=1

Dave, I'm probably being a bit thick....but how are you moving your Z axis up and down? From what I can see you've got your Z stepper mounted on the profile behind the fixed Z front plate? Is there something on the bolttom of your Z front plate that connects to the stepper behind the plate?

No, the round object sticking out of the top is the coupling for the stepper, it's just not fitted yet :)

It means i can move the z up and down by hand easily.

Inside its a standard screw type affair.

Quote:

---

Originally Posted by Davek0974

No, the round object sticking out of the top is the coupling for the stepper, it's just not fitted yet :)

It means i can move the z up and down by hand easily.

Inside its a standard screw type affair.

---

OK I'm still being thick :stupid:....but I'm still not seeing how you're connecting your stepper/screw to the front of the Z axis, your torch holder?

Quote:

---

http://www.plasmaspider.com/download...p?id=36960&t=1

---

That's the front of your Z, how's that connected to your stepper/screw or whatever??

Ok, looking at the diagram you quoted, the long 5mm slab at the base is the large flat vertical plate shown in my picture :)

The Hiwin rail/block in diagram is shown on the bottom left of my picture - the block is not fitted yet.

I get that Dave, but not how the block and torch mount connect to the stepper?

Ah, right.

Behind the plate lies one of these -
http://www.ebay.co.uk/itm/NEMA23-75m...item4881441268

Its a standard Z-axis unit, screw drive, 2mm pitch, 75mm travel - the motor sits on top and drives the screw direct, the middle block is fixed to the gantry carriage and the upper/lower blocks move with the motor/screw.

Works well.

Ah ok, so why do you need the rail and block on the plate that's moving up and down? Fine adjustment? Surely mounting the torch holder on the bottom of the plate will be enough?

Quote:

---

Originally Posted by njhussey

Ah ok, so why do you need the rail and block on the plate that's moving up and down? Fine adjustment? Surely mounting the torch holder on the bottom of the plate will be enough?

---

The second rail/block you mention is the floating head, it only moves a few mm, just enough to trip a switch as the torch probes the metal surface before cutting.

You can't do initial height sensing without one.