Steel tube build

I am planning on making my first cnc router, without much experience. I want to cut wood, and hopefully the occasional sheet of aluminum and some copper.

My draft use 200x200x10mm steel tubes for the X- and Y-axis. Two 20mm linear rails and a ball screw on each section. 20mm steel blocks on top of the rail carriages. Z-axis use 10mm steel plates. I’ll look into adding a house for the Z, and maybe removing the spacers there as well.

The bed rests are 120x80x6mm steel tubes, bolted on to the square tubes as I don’t know how to weld. Total machine size is 1900x1400mm, with the gantry and Z-axis taking up about 400mm. Weight about 500kg without table. I guess I could make it shorter and try to tile my projects. Ignore the long rails on the z-axis, I’m still new to cad.


Currently looking to use levelling epoxy for the rail surfaces, but the surfaces are rather large. Any experience with epoxy for this large an area?

The gantry and Z-axis with spindle will be about 130kg. I haven’t done the math, but could two closed loop/hybrid servo nema34 steppers move it? Not looking to go very fast.

200 x 200 is a bit overkill why this size.? 200 x 100 would work just as well and save weight and money, also easier to epoxy.
Twin rails on the long axis are not needed for a machine this size and usage. Again drop them and save the money.

Regards saving epoxy then just create a strip for the rails by fastening thin lengths of 10x6 steel on the surface and use these as dams.

With the 130Kg and could closed-loop steppers move it then yes easily if you use 230V mains drives. Regards the speeds then you are making the classic mistake by thinking you don't need to move fast. To be honest it's got nothing to do with what "YOU" want to cut at and everything to do with what speeds n feeds the material and tool require to be cut at.
Get this wrong and you'll have poor finishes and wear out tools or worse broken tools.

You need to seriously work on the Z-axis. What you have now, factoring in the rails not being correct, would be very weak. There would be no point building the base frame and gantry so strong only to fit a weak Z-axis.
The Z-axis is THE single most important area of a machine because it holds the spindle. Your's needs a lot of work and seriously strengthing. I would consider moving the top rail from the front face to the top face to give more spacing between the rails and support the Z rear plate more.

For Z axis plate you can bolt some flat bar to the sides to stiffen it up.

I inted to use 25mm aluminum plate on my router and I'll bend sheet metal around the spindle and thus making a tube like shape, making it super rigid.https://uploads.tapatalk-cdn.com/202...7285382022.jpghttps://uploads.tapatalk-cdn.com/202...b21d31f6c2.jpg

Sent from my SM-G970F using Tapatalk

100x100x3mmm tubes works best IMO as most simple the build and could be cut at home using Rage or normal saw. And yes, if you could find sth like 80x120mm even better. The ideal for the gantry is 200x80, but as no such 2x 100x100 or similar will be alright.

Better do the Z with the rails on the front plate and they to slide, not the bearings. makes a stronger Z. I would concentrate on the Z mainly and do the machine around it. As the machine is as strong as its weakest link, the Z is the weak spot most of the times.


Abot the feeds and the speeds:



Most of the time on a home diy machine there is no tool change, to save time, tools or babysitting the job. No toolchange= slower machining speeds. I would say that most of the machines do not need to move faster than 5000mm/min in real life.
For example normally my rapids are 1500mm/min and normally i machine wood at 1500mm/min

But there are cases where i have to make a production run. Say 300 pieces of something. And then i push the machine properly. Say if slow it will do the job for 12 min, when i push it it will do the job for 4 min. And that matters as it could save me a day or two. At these cases i am in the 7-8k mm min.

My machine can do 20k min.when servos at 3k rpm and even faster as my servos are 6k rpm rated , that is at 3k mm /sec acceleration. I have machined at 16k mm/min and while it makes me smile to see my baby move like a bullet shot from a gun that 250kg gantry, it starts to feel and look really Dangerous. The ball screws start to heat up, so i have to grease everything perfectly before hand. The biggest problem is cause wood starts to chip sometimes as its not perfect material like the plastic or similar. So at the end of the day my philosophy is : Less time i spend on the machine=The better , hence the slower speeds give me perfect finish and i spend less time on the machine. Now if i had ATC i would have pushed jobs using roughing bits, normal and then finishing passes, but as its not the case, i am happy..

So, nope. You don't need a fast machine . You need a precise machine that is free of vibrations and gives you perfect finish.

Quote:

---

Originally Posted by Boyan Silyavski

My machine can do 20k min.when servos at 3k rpm and even faster as my servos are 6k rpm rated , that is at 3k mm /sec acceleration. I have machined at 16k mm/min and while it makes me smile to see my baby move like a bullet shot from a gun that 250kg gantry, it starts to feel and look really Dangerous. The ball screws start to heat up, so i have to grease everything perfectly before hand. The biggest problem is cause wood starts to chip sometimes as its not perfect material like the plastic or similar. So at the end of the day my philosophy is : Less time i spend on the machine=The better , hence the slower speeds give me perfect finish and i spend less time on the machine. Now if i had ATC i would have pushed jobs using roughing bits, normal and then finishing passes, but as its not the case, i am happy..

So, nope. You don't need a fast machine . You need a precise machine that is free of vibrations and gives you perfect finish.

---

That's a crazy way to work. So your cutting wood at 1.5mtr/min when you could and should be cutting at 7mtr/min and causing excess tool wear just because you don't want to change the tool one time and do a finish pass. . . .A manual tool change takes 30seconds.!!

I cut aluminium faster than 1500mmm/min.! :thumsup:

Quote:

---

Originally Posted by JAZZCNC

That's a crazy way to work. So your cutting wood at 1.5mtr/min when you could and should be cutting at 7mtr/min and causing excess tool wear just because you don't want to change the tool one time and do a finish pass. . . .A manual tool change takes 30seconds.!!

I cut aluminium faster than 1500mmm/min.! :thumsup:

---

I guess realy depends on what kind of job i am doing. When i cut general stuff and cleaning passes yes i cut at 7-10k mm min, but when i am fabricating for example the music instruments i am making i keep slow, because going fast chips edges from time to time and is very difficult to repair if at all, as size must be correct. So i dont speed then and cut at finishing speeds. As at the end of 3-4 days work if i have chipped say every five from 100 pieces it would be like 50 pieces to repair. Which is tremendous amount of time lost and material, if its unrepairable.

Same with the signs i make. I aim at perfect cut of Plexiglass or Dibond. The other day i cut PLexiglass and made a video, even wanted to post it here. 18k. spindle speed and 70 ipm/ at 0.5mm deep using 1/8 single flute. I played like 3 hours to find the perfect ratio. basically went to the conclusion that could go 2 times as fast and 3 times as deep, but from time to time there was crap on the bit and i had to be constantly there. Having in mind had to cut 8 hours and the material is 100e/m2 i have decided to go at said speeds and air blow from the fog mister and that let me have the job unattended. So simple.


PS i cut aluminum at 2k mm min also

Quote:

---

Originally Posted by Boyan Silyavski

I guess realy depends on what kind of job i am doing. When i cut general stuff and cleaning passes yes i cut at 7-10k mm min, but when i am fabricating for example the music instruments i am making i keep slow, because going fast chips edges from time to time and is very difficult to repair if at all, as size must be correct. So i dont speed then and cut at finishing speeds. As at the end of 3-4 days work if i have chipped say every five from 100 pieces it would be like 50 pieces to repair. Which is tremendous amount of time lost and material, if its unrepairable.

---

Yes exactly, which was my point. That's the material dictating the feeds not you deciding to go slow. I'm sure if you could with same finish etc go faster, then you would.

End of the day it's mostly the material and tools that dictate the feeds n speeds required and not something the machine should dictate because it's under specified or poorly designed. Like wise the user's choice in feeds n speeds is limited to a small range if they want to cut correctly and not destroy tools or materials.

Thanks for all the good feedback! I need to implement this for the next draft.

My own welded steel router is built from 50x50 and 100x50 steel box, all 3mm wall thickness (search for "AVOR" on this forum for a bit more detail). Overall size roughly 1800x1000 with a cutting area of about 1500x750. I'm happy with the overall strength of the machine but if I were doing it again, I would use at least 4mm, maybe 5mm, wall thickness steel. Not for strength but mainly to control vibration and resonance in the tubes. It also makes it easier to bolt on the rails, etc, as 3mm is a bit too thin for a decent tapped M5 hole. 10mm wall thickness seems way too much for the uses you suggest - that's heavy stuff!