Building My First CNC Router, Looking for Advice & Other Info

[AlexDoran]

Hi All,

Just signed up this forum, i am currently designing a small (Large Desktop?) sized CNC Router for our business, it will primarily be used to machine 1mm Intumescent Fire Sheet, but also for producing one off parts for some of our older machines (Aluminium, Steel etc?). We have two much larger CMS Routers already that we use for production of which i create C Based programs for. I have a pretty sophisticated knowledge of building 3D Printers but this will be my first router, i had found this design online and really like how robust it looked:

https://jeremyyoungdesign.com/2016/0...inal-assembly/

So the design is heavily based on that machine.

So far it is looking to be built from the following:

Frame - Genuine Rexroth Extrusions & Connectors.
Gantry - 18mm Aluminium - Bespoke Plates, Would need to have these made (Pricey?!).

Low Profile L1016 Linear Rails & Carriages from Automotion Components - On all Axis

4NM Nema 23's on All Axis ( This Kit: https://www.cnc4you.co.uk/CNC-Kits/S...-4-Axis-Nema23 )

Dual RM2010 Ballscrews on X Axis

Single RM2010 Ballscrew on Y Axis

Single RM1605 Ballscrew on Z Axis

Appropriate Fixed & Floating Blocks on each Axis.

Ebay / Chinese 2.2kw Water Cooled Spindle


So that is the rough spec of the machine currently, the main things i am concerned about so far are:

- Controllers - The motor & driver kit comes with a parralell breakout board, however i want the machine to be compatible with newer Laptops (Windows 10 x64), when building 3D Printers i tend to use Clone Smoothieboards, would these also be suitable for this CNC Build? They perform well and are cost effective, also have stepper signal breakouts.

- Endstops - Mechanical endstops are common practise on 3D printers, i see many Router builds done using proximity sensors or hall effect sensors, are any recommended for maximum repeat-ability?

- Resolution - Will the 10mm pitch and 5mm pitch provide me with a balanced resolution / speed combination, i was advised previously that 2005 screws on the X & Y might hold back achievable speeds - or that i might be disappointed with the feed rates they would achieve.

- The build this is based on uses Aluminium for the majority of the Gantry and Z Axis plates, would steel also be acceptable? I was worried about making he gantry too heavy.

- I was considering buying the ballscrews from china, if they arrived with runout evident, could they be straightened, or might they straighten out once fixed in position?

- Might anyone recommend a company / individual who cold produce the bespoke plates.

Any help / criticism / heads up on missed items would be greatly appreciated.



Thanks

Alex

[Nr1madman]

Hello! I don't have many answers to your questions but I can give you a big thumbs up!
Looks like you did good research and I'm fairly certain you will build an awsome machine!

Ballscrews with 10mm pitch is good for router speeds ie wood and plastics. If you were to use it as mill for steel and such then 5mm pitch is better.

If ballscrews are bent then maybe you could straighten it. Much worse if the end machining for the bearings are off center :)

Steel is good for gantry and plates as well as aluminium, you just don't have to use as thick stock!

[Clive S]

Hi All,

Low Profile L1016 Linear Rails & Carriages from Automotion Components - On all Axis

4NM Nema 23's on All Axis ( This Kit: https://www.cnc4you.co.uk/CNC-Kits/S...-4-Axis-Nema23 )

Dual RM2010 Ballscrews on X Axis
Single RM2010 Ballscrew on Y Axis
Single RM1605 Ballscrew on Z Axis

Appropriate Fixed & Floating Blocks on each Axis.

Ebay / Chinese 2.2kw Water Cooled Spindle


So that is the rough spec of the machine currently, the main things i am concerned about so far are:

- Controllers - The motor & driver kit comes with a parralell breakout board, however i want the machine to be compatible with newer Laptops (Windows 10 x64),

- Endstops - Mechanical endstops are common practise on 3D printers, i see many Router builds done using proximity sensors or hall effect sensors, are any recommended for maximum repeat-ability?

- Resolution - Will the 10mm pitch and 5mm pitch provide me with a balanced resolution / speed combination, i was advised previously that 2005 screws on the X & Y might hold back achievable speeds - or that i might be disappointed with the feed rates they would achieve.

- I was considering buying the ballscrews from china, if they arrived with runout evident, could they be straightened, or might they straighten out once fixed in position?

Any help / criticism / heads up on missed items would be greatly appreciated.



Thanks

Alex

Welcome to the forum Alex.

Some comments: Kits of parts are never a good idea, I hope you have not purchased them yet.
Nema 23 4nm are overkill I would use and have done several times nema 23 3.1nm from the same source now then that kit is using a 36v power supply you really need about 68v to get the best out of the motors so build your own (look up Joe's excellent vid https://www.youtube.com/watch?v=WISDqx-e23M ) go for AM882 drives from China as they have stall detection etc (you will need that if you are putting two motors on one axis)

Linear rails and ball screws can be got from Fred at BST [email protected] China very reliable guy.
Just email him for a quote and mention the forum. You may need to have the F length on the screw to be altered from standard 15mm say to 30mm to fit the pulleys Your screws are fine for what you want to cut.

Spindle again China be careful with the vfd as there are two types that look the same to the untrained eye. So post a link here before you buy.

proximity sensors are good and cheap about £3.50 china or ebay

Finally keep all your questions in one place then people can see what you are doing and chime in
Good luck with the build

Edit go for 20mm rails all round 15mm are fine but much harder to deal with.
Edit Just noticed the screws 20mm!! I would use 16mm all round but it depends on the length.

[RobC]

Welcome Alex! I'd personally be rather worried about how the weight from the gantry is being applied to the X-axis bearing blocks at such an angle. I'm not sure if anyone else would share the same worry but It's certainly a great design! Now as someone who recently within the past two months purchased ball screws from china there is two things you need to know customs will sting you for import tax and the quality of the screws is fairly decent (20mm screws 5mm pitch).

[RobC]

Also to some more of your questions. My gantry was built from steel and I've regretted doing it ever since. Far too heavy and aluminium would of done the same job with ample structural rigidity. Proximity sensors are much better than mechanical switches, break out board wise if it has a parallel port if you want to connect to modern computers just buy a UC100 connector and all your issues will be resolved but make sure it's the legit one and not a Chinese fake one otherwise it ain't going to talk.

I have got a post talking about my gantry on my build post if you want to see why steel was a bad choice by me.

[paulus.v]

Welcome Alex! I'd personally be rather worried about how the weight from the gantry is being applied to the X-axis bearing blocks at such an angle. I'm not sure if anyone else would share the same worry but It's certainly a great design!

I'm not sure I understand what angle you refere to... If it's about the gantry weight distribution to the X axis glides assemblies, the answer is at a few SW clicks distance. Make sure you have set the actual material/ mass of the components then look at the center of gravity. It should be close to the center of the X axis bearing blocks assembly in the vertical plane perpendicular to the X axis.

[RobC]

I'm not sure I understand what angle you refere to... If it's about the gantry weight distribution to the X axis glides assemblies, the answer is at a few SW clicks distance. Make sure you have set the actual material/ mass of the components then look at the center of gravity. It should be close to the center of the X axis bearing blocks assembly in the vertical plane perpendicular to the X axis.

Yes i was referring to how far out from the blocks the weight is being applied.

[AlexDoran]

Hello! I don't have many answers to your questions but I can give you a big thumbs up!
Looks like you did good research and I'm fairly certain you will build an awsome machine!

Ballscrews with 10mm pitch is good for router speeds ie wood and plastics. If you were to use it as mill for steel and such then 5mm pitch is better.

If ballscrews are bent then maybe you could straighten it. Much worse if the end machining for the bearings are off center :)

Steel is good for gantry and plates as well as aluminium, you just don't have to use as thick stock!

Many thanks, i appreciate the advice, i will bear that in mind about the end machining i did not consider that before, only that they might be bent!

Welcome to the forum Alex.

Some comments: Kits of parts are never a good idea, I hope you have not purchased them yet.
Nema 23 4nm are overkill I would use and have done several times nema 23 3.1nm from the same source now then that kit is using a 36v power supply you really need about 68v to get the best out of the motors so build your own (look up Joe's excellent vid https://www.youtube.com/watch?v=WISDqx-e23M ) go for AM882 drives from China as they have stall detection etc (you will need that if you are putting two motors on one axis)

Linear rails and ball screws can be got from Fred at BST [email protected] China very reliable guy.
Just email him for a quote and mention the forum. You may need to have the F length on the screw to be altered from standard 15mm say to 30mm to fit the pulleys Your screws are fine for what you want to cut.

Spindle again China be careful with the vfd as there are two types that look the same to the untrained eye. So post a link here before you buy.

proximity sensors are good and cheap about £3.50 china or ebay

Finally keep all your questions in one place then people can see what you are doing and chime in
Good luck with the build

Edit go for 20mm rails all round 15mm are fine but much harder to deal with.
Edit Just noticed the screws 20mm!! I would use 16mm all round but it depends on the length.

Hi Clive, thanks for the pointers! I have as you said decided not to buy the kit, the saving was insignificant and i did not want to be using that breakout board they were supplying anyway. I also questioned why they supply a 36v PSU with the kit, when you look at their higher volt PSU's it states that they are not compatible with the Drivers supplied in that kit...? Seems iffy to me, maybe the drivers are cheap and can't run at the stated specs. I will probably go for those drivers you have mentioned if they come with good recommendation.

Not sure i feel comfortable building my own PSU haha, i will try to find a higher quality alternative.

Ok so, when building my last 3D Printer i had used a chap on Aliexpress to provide me with the ballscrews and everything else associated with them. I compile a list of what i needed this time and first priced it from the UK, so i had a list of what i needed which was going to come to about £690 + Delivery here in the UK, went to aliexpress and returned to the guy who supplied me before and sent me this quote back:

2 x SFU2010- 1200mm with 2pcs nut with end machining=$42
3 x Set BK15 BF15=$24
3 x DSG20H=$6
3 x XB30x40-8x12=$8.46
1 x SFU2010- 712mm with 1pcs nut with end machining=$16.46
1 x SFU1605- 362mm with 1pcs nut with end machining=$10.2
1 x DSG16H=$1.21
1 x Set BK12 BF12=$6.22
1 x XB30x40-8x10=$2.82

Price is $117
Shipping costs is $100 by fedex to UK
Total is $217

Now i honestly could not believe how cheap that was, if i ordered today it would be £166. I got a price from another supplier on there and they quoted ~£340 delivered, would your guy be anywhere near this do you think? Even if they were bent there would be lots of room to play in terms of having them straightened at that price.

Spindle wise, i was going to go for this, unless someone has experience with another:

Click Me

Thanks for the other tips about the motors etc, i like overkill so i think i will stick with them in case of any future needs.

Welcome Alex! I'd personally be rather worried about how the weight from the gantry is being applied to the X-axis bearing blocks at such an angle. I'm not sure if anyone else would share the same worry but It's certainly a great design! Now as someone who recently within the past two months purchased ball screws from china there is two things you need to know customs will sting you for import tax and the quality of the screws is fairly decent (20mm screws 5mm pitch).

See Below

Also to some more of your questions. My gantry was built from steel and I've regretted doing it ever since. Far too heavy and aluminium would of done the same job with ample structural rigidity. Proximity sensors are much better than mechanical switches, break out board wise if it has a parallel port if you want to connect to modern computers just buy a UC100 connector and all your issues will be resolved but make sure it's the legit one and not a Chinese fake one otherwise it ain't going to talk.

I have got a post talking about my gantry on my build post if you want to see why steel was a bad choice by me.

Yes i like the idea of Aluminium so i think i will definitely go with that after hearing that, proximity sensors are good for me and they are cheap too so i will incorporate them.

I have found a controller that i think i am now going to use:

DDCSV1.1 Controller

I like the idea of just being able to run files external from the laptop so if it's as good as some of the reviews say it is (for the money) then i will probably go with this. I also liked what this guy did with it:

Brainright

I'm not sure I understand what angle you refere to... If it's about the gantry weight distribution to the X axis glides assemblies, the answer is at a few SW clicks distance. Make sure you have set the actual material/ mass of the components then look at the center of gravity. It should be close to the center of the X axis bearing blocks assembly in the vertical plane perpendicular to the X axis.

Ok, so the Axis is orientated as is to try and get the maximum workable area within that size of frame, i could have made it more central "looking" but in reality its not really central because on one end of the axis the spindle can reach past the machine, and on the other end you have a few hundred mm of area that you cannot reach, this is why i have tried to balance it and make the most of the space available. I'm nt sure of the weight of the spindle but this is obviously very important in order for Solidworks to calculate the centre of mass, i had read online that in general a 2.2Kw spindle will weigh in the region of 5Kg. So i set that as the mass of the spindle and run the evaluation, and it came out pretty well imo, more than acceptable to me, the linear rails and carriages im using are supposed to be high quality so im hoping that they will be fine:



EDIT: Just realised that picture isn't so great, the Centre of Mass appears to be just in front of the left side carriage block, and it also shows a slight angular (right word?) pull, twisting towards the rear of the X Axis. Possibly i could move the right carriage block towards the middle some? This would seriously hurt my OCD of having them space equally.

Thanks Again for all the replies i've already learnt alot so keep that criticism coming :D

Alex

[Clive S]

Not sure i feel comfortable building my own PSU haha, i will try to find a higher quality alternative.

One similar to this http://www.zappautomation.co.uk/elec...er-supply.html

Re the spindle make sure it is the correct vfd check the picture and make also 4 bearings

http://www.ebay.co.uk/itm/WATER-COOL...kAAOSw0e9UtOrX

Re the controller will it home two screws with independent homing? Will it cope with full Gcode G2 G3 , canned cycles etc. What size gcode file can it handle? (I have no experience with these but they have been debated on the forum)

[AlexDoran]

Hi Clive,

Thanks for the PSU link, says they are out of stock but i can look for a similar item based on that.

Sorry for my lack of education, what do i need to look out for when it comes to the VFD? What differentiates between the correct & incorrect one?

I am not sure about the independent homing, i will need to check that for sure, is this common practise for machines with multiple screws on one axis? Is this to ensure they stay parallel? I believe it can handle full GCode list, need to research canned cycles, apparently it can handle any file size because it reads from the USB stick as it goes.

Unconvinced with my result from the Centre of Mass evaluation i done, i got on the phone to Solidsolutions, who provide our solidworks license and also provide support, indeed i had done it incorrectly. Here is where the actual centre of mass is on the gantry. It's hard to see, but where it says "Point 1".



Not sure how much of a problem this poses, are the carriages likely to be affected by this? Not how i might remedy this other than reducing the Gantry to a single 40x40 extrusion as opposed to 40x80 and then centralising over the Gantry Base Plates further. This in turn might affect how rigid the gantry is surely?

Thanks

Alex