View Full Version : My second router design

17-04-2013, 04:50 AM

I have now started work on my second CNC machine! these things are very addictive!

I have drawn up some plans on Google sketch up and thought I would post them here to see what other people thought of the designs and to see if there was any problems that I haven't spotted as my last machine has many 'faults' should we say.

The basic features are:

a working envelope of 700 x 550 x 200

it uses SBR20 supported rail throughout along with RM1605 ball screws all purchased from chai in china as I am there now so I can cut down shipping cost :playful:

the spindle is a 2.2kw water cooled model that I already have from my other machine

The main construction material is eco cast aluminum tooling plate which is machined flat on both sides and the main gantry and black piece is 20mm thick

all the parts are to be cut out on a CNC mill, hopefully the proto track at my university so this should ensure that all parts are square and this should mean that on assembly the machine is more acurate

I have not decided on motors and drivers yet but I am thinking the high voltage kit that is supplied from Zapp which uses 3nm motors and 60V+, it would be good to see what people think about this.

the main aim of this machine is to be able to cut wood, plastic and aluminum, my hope is to be able to cut aluminium at a reasonable speed for a DIY machine but I am not sure if the machine will be rigid enough.

any constrictive criticism is welcome, and I look forward to hearing what you think about my design. I may have not put enough info about the design so any questions just ask



21-04-2013, 05:00 AM
Ball screws have now arrived along with the linear rails, just need to get them home in one piece !

I have also been doing some research on stepper drivers and I am still unsure what to do, my ball screws have a pitch of 5mm and will be direct drive from the stepper motors so does that mean that I will have to have high voltage drivers to produce the speed required for a decent feed rate?

I have found these on eBay and was wondering what anyone else thinks about them compared to the high voltage kit supplied by zapp
CNC Wantai Digital Control 4Axis Stepper Driver,24-80VDC/7.8A www.wantmotor.com | eBay (http://www.ebay.co.uk/itm/180574418300?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1423.l2649)

Also any thoughts any one has on the machine design would be greatly appreciated


21-04-2013, 07:50 AM
It looks very simular to my first router, i reckon you should beef up the gantry some more,run something across the back of the y axis plate and swap the z axis rail and bearing blocks around so the rail is on the back of the front plate,
not sure what zapps high voltage kit consists of as ive not been on there a while.but on both machines i knocked up ive used zapps pm752 drivers and their SY60 steppers,cant fault them.

21-04-2013, 08:17 AM
Thanks for your reply

Yes I thought that the y axis plate may flex to much, I could make a structure from 20mm box section steel which would be say 70mm thick in total and then fly cut the surfaces flat for accuracy but this would obviously be heavier, or just put a few box sections bolted to the back of that plate. As for the Z axis I thought about doing it the way you mentioned but does that then make the whole thing to heavy for the z axis motor to move as it has the weight of the rails + plate + motor + spindle or is it fine as the pitch of the screws if quite fine so it does not load the motor to much.

can I ask what sort of feed rates you can get with those drivers?

21-04-2013, 09:17 AM
Hi, too late for this comment being of any use, but a working area of 800x600 would have been useful to enable 1/6 sheets to be used. My machine has a 545 travel, and as I work in mdf and ply I end up with more waste and inferior nesting than a 600mm travel would have allowed. Good luck with your build. G.

24-04-2013, 07:53 PM
I have now changed the design slightly and I have beefed up the gantry more, well I hope so anyway!

It is now made from box section steel, the biggest wall thickness I can get, probably 3mm. It is the part in blue just so that it can be seen more easy. The plan is to weld 2 sections of 50 x 25 together with a piece of 25 x 25 to create a sort of I beam shape. The frame would be clamped together and then spot welded to try and retail some squareness before it is fully welded but I know this process will put the whole part out of square and distort it, so it will then be fly cut flat on all faces so it can be mounted square to the rest of the machine. The idea behind this was that steel is far cheaper and I have the facilities to then make it flat after, but I am still not sure if that structure will be stiff enough to cut aluminum at reasonable speeds.

As for the z axis I have reversed the bearings and rails so that the rails are now on the dropper plate but because I can not mount the bearings back to back the ones on the z axis are much closer together with only 80mm separation between centers, do you think this will be a problem?

Any thoughts would be greatly appreciated :)

24-04-2013, 09:44 PM
Looks really good joe. X axis bearings are spot on. One at the back and one in line with the cutting tool.

Again probably a bit late but due to the unequal load capacity of the open bearings on the supported rail it is best to mount them back to back, so on top and underneath. This also pulls the Z axis closer to the gantry or gives you more room to widen the steel rails which will improve overall rigidity.

This would also mean you can move the lower blocks on the z axis down level with the under side of the gantry to make the most of them. wider spacing reduces transmitted load from the lever arm of the z axis.

This is only my point of view and others may disagree,

If you are planning to cut lots of ali it will be worth getting the gantry down as low as possible and then packing up the work piece to reduce z travel. (edit just looked again and realised you have.)

Happy building

25-04-2013, 01:36 PM
Thanks for you input, its great to hear what other people have to say

what you are saying about mounting the Y axis rails makes sense and I have not re designed it so that they are over each other. but during this I have changed the gantry design. It now uses 2 sections of 25 x 50 with an aluminum plate between them which is 126 x 725 x 16. My problem is that I am unsure weather to use aluminum bar or steel bar on the sections highlighted in red.

The only reason I can think about not using steel would be weight so I did some calculations to find a estimate of the whole Gantry weight and it came out to be 45kg using 5083 aluminum with a density of 2650 Kg/m3.

If I was to use steel it would be an extra 17.9 Kg onto the over all weight compared to only 8.7 if it was aluminum so that would be 17% heavier.

I don't know if this would make a noticeable difference as it would be heavier but it would defiantly be stiffer but would the aluminum be stiff enough on its own?

25-04-2013, 02:41 PM
I'm just in the process of doing a post to explain the reasons why the rails work better that way. Looking at the Thompson guidance the load capacity is reduced by a quarter if the force is pulling the bearings away from the rail. whilst you probably wont exceed the max. load of the bearing it is the rigidity that is important.

Nice touch adding the wrap around z axis carrier, should be rock solid. I noticed on the Thompson site that they use a small bolt to squeeze the open ends onto the rail to pre-load them and remove any clearance to further stiffen the carriage. you could easily add this if needed ( note only very small amount and does mean the rails must be perfectly aligned as the factory clearance is to over come minor rail alignment problems.)

With regard to the the red sections I would stick with ali as you will then be able to have decent tapped threads for the rail mounts and backing plate. Normally solid bars are not the most efficient way of making beams as the centre part offers little in the way of torsional and bending resistance (parallel axis theory), but as you want a solid machine for Ali then the extra weight will help damp out vibrations. if you went for steel I would go for box section to reduce weight but these are known to 'ring' as certain frequencies, add the need to grind and its probably best to go for the Ali bar.

I would still be tempted to widen the y axis to create a squarer beam, as this will improve the torsion resistance when the cutter is in he middle of the beam. you could just add some angle to the back plate.

Again just my opinion, not saying you have to do it this way.

25-04-2013, 08:40 PM
I would change the joint for the top Y-plate that holds two of the Y-axis bearings to make it easier to adjust the rails and bearings so that they are parallel. Currently you're relying on the plate at the back being both precisely the right height and having parallel edge faces, which is risky. If you change the joint round so the back plate is bolted to the aforementioned Y-bearing plate, then that plate with the bearings will self align to the correct position.

I strongly advise using timing pulleys (most likely 5M HTD) at the very least on X and Y since as you have chosen 5mm pitch ballscrews, you may need to use a 2:1 ratio between the motor and ballscrew to get decent feedrates for cutting wood. Since you're planning on mainly cutting aluminium, that's not so big a deal, however pulleys also help damp vibrations, so reduce resonance problems. Using pulleys also means you don't need motor couplings, which are often a weak point as you can see from the number of threads about which ones to choose. Pulleys easily tolerate a lot more misalignment than oldham couplings, so it arguably makes mounting the motors easier.

Have you considered not having gantry sides and making the machine frame such that the tool operates below the level of the X-rails, rather like your first machine? This tends to be much stronger as it replaces the flimsy gantry sides with the steel frame, which can be made as strong as you want since it's not moving. This is especially important since you want to cut aluminium, which as you no doubt know requires substantially higher stiffness than your previous machine. If you stick with the current design, then most of the time the Z-axis will be operating at near full extension, so the force imposed on the gantry due to cutting forces will be significant. Even if the material is raised up, you still have gantry sides which aren't as strong as a steel frame.

I would still be tempted to widen the y axis to create a squarer beam, as this will improve the torsion resistance when the cutter is in he middle of the beam.

I would too.

25-04-2013, 09:17 PM
Thanks for your input

I have now changed round the Y axis plate as you said, this is one of those little things that will make putting the whole thing together much easier.:wink:

Ross what exactly do you mean by make the Y axis wider? doesn't this mean that the distance between the cutting tool and the back of the Y axis will be further?

As for the 2:1 belts, I had considered this but I wasn't sure if they would flex to much as in my old machine the belts stretched quite a lot under load. I was wondering what your experience of this was. And could I ask what sort of feet rate is possible from a 5mm pitch screw using 72V NEMA 23 drivers.

Yes I considered making the gantry sides much smaller but I was doing it this way so that the whole frame that the machine sits on could be fly cut flat so that I don't have any issues with the bed not being flat or to far off flat. but your right this would make the whole machine stiffer. But thinking about it just now as long as the surface that the rails sit on are flat compered to each other then the the machine could level its bed and it should then be fairly flat.

I will change the design slightly and put up a few more pictures

Thanks for all your help everyone

25-04-2013, 09:42 PM
Ross what exactly do you mean by make the Y axis wider? doesn't this mean that the distance between the cutting tool and the back of the Y axis will be further?

If you look at the last image of post #8 then you have dimensioned the beam as 126mm deep and & 70mm wide (guessing back plate is 20mm) the 70 mm is the bit increase. yes this will increase the distance from cutting tool to back of the y axis but you have gained a bit of space by rotating the rails to the top and bottom and you could allow the very back to over hang the rear gantry bearing a bit as the resultant twisting force from the z axis on the yaxis will act in the centre of the beam. as long as the rear X axis bearing on the gantry sides is behind the centre it will be ok.

If you are going to belt drive the z axis motor then it might also be worth mounting on the back of the z carriage plate so it is lower on the machine, whilst it is not going up and down with the Z axis it is still being accelerated back and forth on the Y axis, bringing all the mass closer to the centre can only be a good thing :o)

25-04-2013, 09:46 PM
Ross what exactly do you mean by make the Y axis wider? doesn't this mean that the distance between the cutting tool and the back of the Y axis will be further?

Ross means increase the spacing of the Y-axis bearings. The deflection due to the bearings is proportional to the cutting force squared, so increasing them a small amount makes a surprisingly large difference.

As for the 2:1 belts, I had considered this but I wasn't sure if they would flex to much as in my old machine the belts stretched quite a lot under load. I was wondering what your experience of this was.

When the belts are used to drive the ballscrew the error due to belt stretch will, in most cases, be negligible compared to other errors. The difference with your other machine is the belt drives the axis directly, so the cutting force is applied directly to the belt. When driving a ballscrew however, the cutting force is divided by the effective ratio of the ballscrew drive, which put simply is a big number so the belt doesn't stretch much at all.

And could I ask what sort of feet rate is possible from a 5mm pitch screw using 72V NEMA 23 drivers.

Try this (http://www.mycncuk.com/forums/faqs-problems-solutions/1524-what-size-stepper-motor-do-i-need.html).

03-05-2013, 01:27 PM
Right I have now changed the design quite a bit

The gantry is now much smaller and I have lifted the sides of the machine up to compensate. The main of the frame will be from box section steel, either 100 x 50 or 50 x 50 with the thickest wall I can get. I will try to weld it using clamps to keep it square and flat the best I can but it will then be fly cut to make sure that the top face is accurate and then the machine can face its base plate to ensure that is flat.

My plan is to have a second false floor that can be bolted into the frame which will be around 100 mm thick for machining parts which require less z travel, this should then make the machine much stiffer as the z axis will not be as far out.

As for driving the ballscrews with belts I am going to stay with direct drive at the moment to save some cost, but I am sure for what I am doing the slower rapid rates will not be to much of a problem.

If anyone else has any suggestions please let me know :friendly_wink:

03-05-2013, 06:07 PM
If anyone else has any suggestions please let me know

Well you could move......and then also......and put that over there..:friendly_wink:

Looks good. I know its a welded steel frame but you might want to look at the junction between the side walls and the bed, you might get flex there. Also can you add any bracing the ends or will that hinder your access? maybe look at at a removable end braces?

To help with the resonance it might be worth unequally spacing the bed supports or at least move the central leg off centre, if that makes sense? vibrations have harmonics at multiples and divisions of the base frequency so by having equally spaced connections you can unknowingly increase the vibration.

What is the bearing spacing on the y axis?