PDA

View Full Version : Router build



Blackrat
17-08-2014, 06:32 PM
so onto my next build ....

my rails have arrived , yay so am just waiting for the motors and drivers
still have to select a spindle, but im quite set on a chinese 3kw water cooled , i had a 1.5kw on my old machine and was quite impressed with it even though i blew the VFD up :S oops

cutting size is 2400x1200x200 so that a std sheet will fit in the machine ..

25mm rails on X Y and Z ..

Z will have a 20mm ballscrew X & Y will be on mod 1.5 rack and pinion

10:1 step down boxes on X and Y with a 26t pinion , Z will either be belt with 1:1 pulley, or direct drive with coupling

rapids will be around 15m/m , will probably speed this up in due time

the frame is quite conservative i think ... and now i dont know if i want to go bigger rectangle section or just stick with the 76mm square
all steel build , gantry sides are 6mm steel plate laser cut , two 76mm box sections for the gantry
Z plates are 10mm with another 10mm to space out for the bearing blocks
epoxy self leveling construction method will be used
where does it stop :D

i will mainly be cutting wood and plastics (corian - acrylic) so i dont need a terribly stiff machine ... for now ... my next build after this will be for aluminum

EddyCurrent
17-08-2014, 07:51 PM
I think you need to go with 12mm steel minimum for the Z and gantry sides.
Take at look at silyavski's build http://www.mycncuk.com/threads/6457-Sturdy-and-Fast-all-Steel-CNC-my-first-build?p=60634#post60634

Blackrat
17-08-2014, 08:39 PM
eish here we go ..... :D

i think while im at it im gonna go with 100mm box section on the gantry ...

Web Goblin
18-08-2014, 07:27 PM
100mm box section! That's massive overkill:tongue:
As you can see 90mm box is enough.

13132

EddyCurrent
19-08-2014, 11:05 AM
@Web Goblin, that looks very nice indeed.

Blackrat
19-08-2014, 12:35 PM
Very nice ... You got a build thread ?

ive been looking at different size rectangle section , but not convinced about stacking two sections on top of the gantry , ideas

obviously it would need to be capped , maybe even try out the damping idea :-P that paper where you stick four tubes inside covered in 3M™Viscoelastic Damping stuff then filled with epoxy ..... looks over at the dynamat shelf ... this is like rocket science

anyone tried to DIY it ?

EddyCurrent
19-08-2014, 10:17 PM
I've mentioned this before but got no response, I think ?
What about using expanding foam for damping ? http://www.toolstation.com/shop/Adhesives%20&%20Sealants/Expanding%20Foam/Expanding%20Foam/d180/sd2857/p27912
That should stop any vibration you would think and it's cheap, lightweight, easily applied.

Web Goblin
20-08-2014, 06:27 AM
Sorry I dont have a build log, didnt really plan on doing one for this build. Stacking two sections of box shouldnt be a problem. Stitch weld them along the seam where they join at both sides and away you go. You will need to get them machined though as they wont be anywhere near straight. If you look at the pic of my gantry you will see flat bar welded to the front of the box section. This was added to be machined to get the front faces parallel and straight for mounting the rail on to. If you also look closely at the bottom of the picture, you cant really see it very well because the box of washers and rag are in the way, but another plate has been welded to the bottom of the box section. This has also been machined at right angles to the front on the gantry. These will sit directly on to machined plates which hold the linear bearing blocks for the x axis and should allow for any small adjustments required to square up the gantry.

Eddy, I dont think that expanding foam would provide any vibration damping or very little because it doesnt have any mass to absorb it. It might provide some noise reduction though.

JAZZCNC
20-08-2014, 03:35 PM
I've mentioned this before but got no response, I think ?
What about using expanding foam for damping ? http://www.toolstation.com/shop/Adhesives%20&%20Sealants/Expanding%20Foam/Expanding%20Foam/d180/sd2857/p27912
That should stop any vibration you would think and it's cheap, lightweight, easily applied.

Doesn't Work very good Eddy it doesn't have enough density.!!

Blackrat
20-08-2014, 04:00 PM
a 10mm inside boundry of epoxy with a filler adds about 10 kilos to each beam ... i think it will made a huge difference

im no engineer , and got no idea where to even start to try to calculate the difference in damping .. but it can only make it better

i thinks im gonna give it a try ...

JAZZCNC
20-08-2014, 07:28 PM
a 10mm inside boundry of epoxy with a filler adds about 10 kilos to each beam ... i think it will made a huge difference

im no engineer , and got no idea where to even start to try to calculate the difference in damping .. but it can only make it better

i thinks im gonna give it a try ...

Can't beat filling with dry sand and bonus is it can be removed when need to move the machine.

Blackrat
20-08-2014, 07:29 PM
On the gantry ? What's the density ? Wanna see what it's gonna weigh :-)

JAZZCNC
20-08-2014, 09:31 PM
On the gantry ? What's the density ? Wanna see what it's gonna weigh :-)

Approx 1250 - 1500 Kg/M3

Don't be afraid of weight in the gantry it can be your friend and give far better surface finish and allow deeper depths of cut. Unless your doing very high feed directional changes then having heavy gantry really does help not hurt.!

Boyan Silyavski
21-08-2014, 12:13 AM
While i agree about the weight, after all i managed to make 1000x400x200 machine to weight 200kg, something that i don't see on this and many designs are the ribs. Apart from the careful listening to advice and using quality components, the machine in question is build so that there is not a single part of it that is not insured against bend in any directions. By ribs. Hence the precision achieved.
Also there is not a single detail which gives doubt about weakness. Where it looked weak on the initial design, i designed it so that the weak places were reinforced by smooth transition and angles made so to transmit forces in the direction i liked.


Point that look not right to me on your design:

Ok, you will weld 2 beams together for the gantry, i did the same , but where are the side plates that are soldered to them at both sides. They will bend and if not, they will vibrate. They can not be left open.


Gantry legs. i see one plate in the middle that will reinforce them . One is enough against left right bend, but not against twist. i would put 2 on each leg so they form a box.


Web Goblin, your gantry looks very nice. What you intend to do with the gantry legs? Why you did them like that? You plan to fit the screw there? I contemplated this design for compactness, but for my liking it seems weak for left right forces. I would have reinforced this place with closing the beams and drilling holes from above to fit the screws to the bearing blocks.

Web Goblin
21-08-2014, 06:36 AM
I did the frame this way because I thought it would give me the best strength while still getting full travel of the gantry rails. The join between the cross rail and foot plate is fully welded and should be very strong.
The bearing blocks are mounted below the "foot" plate. The bmp from my cad drawing below shows how the blocks will be fitted. The top cross bars are not shown here but they will add more strength for left/right forces. The position for the ballscrews is shown as well. These will be covered as well to try to prevent swarf getting to them.

13157

Blackrat
21-08-2014, 06:05 PM
Ok, you will weld 2 beams together for the gantry, i did the same , but where are the side plates that are soldered to them at both sides. They will bend and if not, they will vibrate. They can not be left open.


Gantry legs. i see one plate in the middle that will reinforce them . One is enough against left right bend, but not against twist. i would put 2 on each leg so they form a box.

.

heya silyavski

i agree ! end caps are a must ! i intend to have a few internal webs to prevent twist

im not convinced on this one ... if there was only one arm, then yes for sure , but the other arm of the gantry prevents the twist

i think for now im going to go with filling the frame with sand ... but the gantry , im going to try out a epoxy fill

Blackrat
21-08-2014, 06:47 PM
sand with a 1500kgm3 will add 60kgs to the gantry :S

an idea ....

JAZZCNC
22-08-2014, 11:29 AM
sand with a 1500kgm3 will add 60kgs to the gantry :S

an idea ....

How do you work that out.? What size boxsection are you using and how long.?
Looking at your pictures of design the box section looks like approx 120 x 80 and not much more than 1mtr long (and your proportions are wrong 120x80 won't have square internal rad they will be approx 5mm.)

So according to solid Works, I've just took the volume of the internal of 120 x 80 x 5 x 1mtr length rectangle section and working on 1470kgM3 which is the closest material I could find in Density which happens to be Melamine resin. The density comes out at 11.29kg per 1mtr length. That's just the internal volume not the Steel.

The same volume with Unfilled Epoxy comes out at 8.45Kg but that's without any filling which you would need. SO I reckon Epoxy/sand or gravel mix would end up weighing much more.!

Blackrat
22-08-2014, 11:36 AM
150x100x4 1600mm wide

if you go to the bottom of the materials list, you can specify your own material so you dont need to look for a similar material with the density you require ...

the epoxy is only 10mm thick ... working as a sandwich between the two layers of steel

JAZZCNC
22-08-2014, 12:43 PM
150x100x4 1600mm wide

Why so Big.? . . . . You'd be better off using smaller size with thicker wall.! . . . Besides Your Gantry design is the weak link here not the material.

To me your focusing on the wrong areas of the build and not looking at the Basic design and foundation. Without good foundations best resonance handling material in the world don't mean diddly squat.!!

Another thing to ask your self is do you really need massive strength.? Silyavski (Boyan) machine is a credit to him but IMO it's massive OTT for anything but aluminium. SO Unless your mostly cutting aluminium do you really need to go that far.? . . . . Like every thing in life Strength comes at a cost in both monetry and performance terms so why waste both if your only using 100% of it 2% of the time.?

All round machines are great but they take much more to build so they do it well and even then they NEVER do it has good as a Machine specificly designed and optimised for that sole purpose.
Remember a good design is built just strong enough to perform it's role with enough safety margin to ensure reliabilty and anything more is waste.!

Blackrat
22-08-2014, 12:59 PM
i want to cut fast ... i cant sit around all day watching the machine scrape about ....

so what would you recommend to cut 2400x1200 MDF minimum ... my initial design was 76x76 tube which was too small ... so 150x100 is too big ...

im not too bothered on non critical design details right now , if the radii is 5mm or 6mm its not gonna make a difference to the whole concept .... when i receive the steel i can measure what the exact detail is and amend the drawing to be sure any bolts/nuts/washers dont sit on the radius of the tube

Boyan Silyavski
22-08-2014, 02:18 PM
I don't see why you would not push this build just a bit up and finish with aluminum cutting machine. I mean a very nice aluminum cutting machine.

I am in the process of making now (http://www.mycncuk.com/threads/6619-Quite-an-Unusual-one) exactly the same working area machine and one of the ideas is to cut Aluminum precisely. Ignore the third bearing blocks in the middle everywhere.

By pushing a bit up, i mean on the long axis 2510 screws from china (http://www.aliexpress.com/store/314742) and rotating ball nut (http://www.mycncuk.com/threads/3340-Rotating-Ballnut-design-ideas) . I am not expert here, before some time Dean told me not to go RP if i want precision cuts. You would not believe how comparably cheap are 2x 3m / screw + nuts, for what they will offer.

2x 100x100x4 for the gantry , 100x100x3 for sides , front and back. All rest cross sections could be 80x80x3

just received 7x6m 100x100x4mm which i will use on all machine. 500eur with transport here in Spain. So basically my design uses 42m of main material. It doesn't matter if you make it on the floor or raise the legs, its the same.

i am expecting from my machine 20000mm/min, servo driven 1:1.5 , with very good resolution. That at 3000rpm of the servos. Not that they can not be driven at 4500rpm. There were panasonic servos at ebay, that have the same torque up to 6000rpm. that should be pretty frightening though

if you want fast, and heavy you will drive it AC servo, yes? Not that it can not be step driven, though i cant imagine what i would feel if it looses steps when i am doing 2m ali sheet

JAZZCNC
22-08-2014, 08:06 PM
i want to cut fast ... i cant sit around all day watching the machine scrape about ....

so what would you recommend to cut 2400x1200 MDF minimum ... my initial design was 76x76 tube which was too small ... so 150x100 is too big ...

Ok well if your wanting to cut fast then your concerns are not just strength.? Your main concern will be spindle power and tooling.
You Can build the machine like a tank but ultimately the spindle power and tooling will dictate the speed you can cut at. Any machine using 100x100 steel section for the gantry will more than match most spindles that would be attached to a DIY machine.

Building a successful machine is about balance. No point wasting money on expensive servo's, ballscrews and massive steel structure etc if your going to attach a 1.5Kw spindle at the sharp end.! . . . . . Equally no point attaching 10Kw ATC spindle if the machine is flexing like a jelly fish. The results while differant in practise will still be the same.? . . . Disappointing.!

Balance is required so you need to define what your going to cut with the machine and be realistic regards the feeds/speeds/DOC. Everything needs to match and suit to get a balanced machine and this is where it gets hard and expensive to achieve with a Jack of all trades machine. Esp if your chasing speed.!!

Edit:
so what would you recommend to cut 2400x1200 MDF minimum ... my initial design was 76x76 tube which was too small ... so 150x100 is too big ...

I never said 76x76 would be too small and 80 x80 would make a decent gantry if design was good so I wouldn't be too concerned using that size section provided it had decent wall size. . . BUT . . Your box on box with no bracing and open ends isn't a good design IMO so I'd be looking to change things in this department.

Blackrat
10-09-2014, 08:51 PM
wahoo

my rails and servo kit has arrived

but now to finalize the design to get building

Blackrat
10-09-2014, 09:01 PM
oops ..

the problem now is how do i do the Z axis

mount the rails so that they move ? or mount the bearings so that they move ?

i realize that mounting the rails so that 'they' move is better as the forces on the rails are consistent and that the bearings are fixed ( as per the drawing) and .. that the bearings are maximized as they are directly bolted ontop of each other

but ... the biggest problem is that the travel is now limited ....

if i google z axis pictures, the majority is showing the rails fixed, and the bearings moving ... is it really worth the effort to do the visa versa ?

Blackrat
10-09-2014, 09:15 PM
i think this render explains it a little better

the gantry bearings are directly onto the Z bearings which is great ... but i can only lift the Z rails until they are level with the bearings, which stops Z clearance .... sheesh i feel like im speaking Chinese now

Clive S
10-09-2014, 10:04 PM
i think this render explains it a little better

the gantry bearings are directly onto the Z bearings which is great ... but i can only lift the Z rails until they are level with the bearings, which stops Z clearance .... sheesh i feel like im speaking Chinese nowWhy would you want to go higher than that you have shown, you can't go higher than the gantry so lower the bed. The Z rails (in my opinion) are better fixed on the plate that the spindle is fixed to as it helps to stiffen the plate etc. ..Clive

Blackrat
11-09-2014, 09:08 PM
to clear the work piece ... the bit will foul if it doesnt retract far enough

dont really wanna lower the bed (or raise the gantry) as this will induce more flex , vibration , resonance and the list goes on

Clive S
11-09-2014, 10:43 PM
to clear the work piece ... the bit will foul if it doesnt retract far enough

dont really wanna lower the bed (or raise the gantry) as this will induce more flex , vibration , resonance and the list goes on
Well how about moving the z bearing blocks higher up the plate. ..Clive

Boyan Silyavski
11-09-2014, 11:41 PM
Its a part of my designs also to be able to move the spindle up to clear for a large work piece. But in real life that would mean realigning vertically the spindle, and its time consuming process. The only proper way to do this without fiddling is to make a custom mount and machine the spindle so they fit perfectly, as i make now on my machine now. But its expensive.

having in mind at which stage are you and the prices of ball screws and rails for Z, if simply you made the error and bought them short, spend now 100 quid and buy new ball screw and rails, instead of compromising the whole investment. If don't have them, all the better. The Z should be made properly, not to speak that you want a high performance Z.

Blackrat
12-09-2014, 07:39 AM
+1 clive ... i hear ya ... but then you go down the whole road of tuning fork ... and defeat the whole purpose of having the rails move

im going with bearings moving , like i drew on the original concept, its much easier and i doubt there will be any loss by going with this

silyavski ... i have enough travel on the rails and screws ... its just the design now. I have about 230mm between gantry beam and table top. but only having 230mm of movement isnt enough , i need another 120mm minimum to pull out a long series bit

for example - if i were to machine a block of foam for a plug that is 160mm thick .... and lets say i needed to machine right to the bottom of the foam , so i would need a bit of 120mm long ( the router clears the piece for arguments sake)

then the travel required would be the piece 160mm + 120mm bit + clearance 50mm = 320mm right ? if the move the block up like Clive suggested by 90mm then i would have sufficient clearance

going to redraw bearings the other way and have a scrutinize

JAZZCNC
12-09-2014, 04:49 PM
+1 clive ... i hear ya ... but then you go down the whole road of tuning fork ... and defeat the whole purpose of having the rails move

im going with bearings moving , like i drew on the original concept, its much easier and i doubt there will be any loss by going with this

This is ok if your only ever going to use long tools that reach the bed or material base but what happens if you want to use short stubby cutter.?

In this case if you have to drop the spindle in the clamp this then gives you a long fixed length lever that is a long tuning fork.!! That never changes it pitch no matter if you raise the Z axis.
This does make a differance and in some materials you will clearly see it compared to having rails moving. You can certainly forget cutting aluminium and expect any kind of decent finish.!!

Blackrat
13-09-2014, 09:54 AM
Jazz , I do agree ....
Do you feel I should just raise the bearing a bit to have a touch more clearance like Clive suggests ?

I don't find many examples of moving rails ... Got any links ;-)

Clive S
13-09-2014, 10:13 AM
Jazz , I do agree ....
Do you feel I should just raise the bearing a bit to have a touch more clearance like Clive suggests ?

I don't find many examples of moving rails ... Got any links ;-)A lot of people put the z rails on the plate that the spindle fixes to as it helps to stiffen the plate up. 13372 ..Clive

JAZZCNC
13-09-2014, 10:30 AM
Jazz , I do agree ....
Do you feel I should just raise the bearing a bit to have a touch more clearance like Clive suggests ?

I don't find many examples of moving rails ... Got any links ;-)

No I suggest you have moving rails.!! . . . .The whole idea with any area of the machine is to try and reduce vibration at the tool and moving the bearing higher will just make this worse. Yes you'll have more clearence but you'll get more vibration.

Most machine builds on this forum use moving rails so Check them out.!

To be honest if your just building a Router for cutting wood and soft materials then it's not a problem but if you want to cut harder materials which often require shorter tooling then it does become a problem. Moving rails helps get around this without messing around raising the work or having moving table.

If you want good combination so can cut all materials then have a moving table along with moving rails.

Blackrat
03-12-2014, 06:55 PM
a little progress ....

i thought id try out using epoxy between two steel sheets and i gotta say it works great ! there is no ring from hitting the the beam , and its not 'that' heavy

wish i could have done some before and after tests ... but i dont have the time right now to play

did a quick test run of the X and Y axis today , its super smooth and so quiet, man you gotta love servo's !

mitchejc
17-12-2014, 08:36 AM
Looking great! How did you manage to weld those plates with the oval holes inside the tubing

Blackrat
18-12-2014, 07:00 PM
i cut the inner tubing into sections, then welded the oval plate piece and two rectangle sections together , repeat for the entire length

EddyCurrent
18-12-2014, 07:12 PM
Did the inner tube have to be metal or was it part of the strength too ?

Blackrat
18-12-2014, 07:19 PM
yea must be metal as this completes the sandwich ..