Aluminium Profile Type Router - Initial questions

[CNCRY]

Hi Everyone,

First post on this forum and CNC in general. There seems to be a huge amount of knowledge on this forum, and I feel very out of my depth but learnt a lot lurking here for the past couple of weeks, from a base of zero !
Not sure if I should be creating my own build log as its just a couple questions for now.
So I'm looking at building a CNC router for perhaps 75% soft/hard wood, 25% aluminium use. I mainly work with wood on a hobby level.
I don't have any CAD drawings etc yet, all just paper! sketches but the plan is in the style of the 3d tek heavy mill with some inspiration from here.
wasteboard size 1080 x 1000mm

Here are the parts I'm planning to use this far;

Hiwin 20mm Linear Rails/ HGW20CA Carriages @ 900mm (for X & Y Axes)
2005 Ball Screws @ 900mm for X and Y ( 2 on Y axis rails)
BF15/BK15 Mounts
1605 Ball Screw for Z Axis
Z Axis TBC, but likely SC16LUU type bearings
4 x Nema 23 4NM stepper motors

Purchased so far: Aluminium Profile
2 pieces of 40 x 120 @ 1000mm (Y axis)
4 pieces of 40 x 40 @ 1000mm (X axis supports)
1 pieces 80 x 80 @ 1200mm for X Gantry (will rest on aluminium plate on linear rails)

Parts which I need some assistance with (so far...)
I have minimal electrical experience this far, aside from domestic 240V type.
So looking for comments on the below - will it work together etc. I'll probably not purchase any of the below until physical build well under way but getting an idea of design and cost.

Laptop with Ethernet..
AXXB-E Control board - I like this as it works with UCCNC and Mach3/4, plus doesn't require a breakout board I believe. Does anyone have good/bad experience with this?
Stepper Drivers - Would something like CW8060 from cnc 4 you deliver good performance from the motors? Anything better at similar price point? Stall detection sounds like a must but not clear if this has it?
Power Supply - What Voltage to drive the motors at for optimum performance? and any PSU recommended?

Thanks!

[AndyUK]

Hi CNCRY,

Welcome to the forum!

Usual advice is to hold off on further purchases until you've got the design finalised and pretty confident about each electronic component - changes upstream can irritatingly change what you need on the mechanicals, and electronics have a way of updating while you're building, so best to buy once when you need it rather than in advance, so you've got the right idea there.

I suggest you read through the build log section of the forum; you should be able to look at the 3D Tek design and notice where they've done the right thing and where they've skimped. You've already identified that ballscrews are a more appropriate drive option for example - but there are a few design choices about the Z axis that I wouldn't emulate.

Regards your rails and screws choice, the rails seem appropriate, the screws are about right, but in this size I'd also consider 16mm screws and 10mm pitch options, but all depends on if you want to use pulleys. There are a few calculations you can do to help you determine whats best for your use, but you'll pick this up in the build logs.

You're correct that the AXBB-E has a breakout board built in and seems to get a good rep (I haven't got one!), but also compare against other options such as built-in controllers and say UC400eth with a cheap breakout or expensive breakout. Generally I'd say stick with Ethernet based controllers, unless you're going down the LinuxCNC route.

Voltage and PSU specs are again very common topics in the build logs so I won't delve into it here (but bookmark threads and page numbers when you find these things!!). Generally unregulated PSUs are preferred, a lot of users build their own, or purchase AC drives so you only need a toroidal.

Hope that helps!

[Clive S]

Generally I'd say stick with Ethernet based controllers, unless you're going down the LinuxCNC route.

Would you mind expanding on this?

There are plenty of cards for linuxcnc ie .Mesa 7i92,7i96,7i95 ,7i76e .

[AndyUK]

Would you mind expanding on this?

Hah! Yes, "Unless you're going down the linuxCNC route in which case I have no idea" :P

Reading it back it sounds like I'm saying don't use ethernet with LinuxCNC - it was meant as an "if you're using LinuxCNC there may be more appropriate options like onboard PCI cards".

[CNCRY]

Thanks Andy,

I'm trying to figure out whether 2005 or or 2010 ballscrew would be better, or 1605/1610 (which seems to be less available pre machined). I don't plan to use pulleys.
Problem is all calculations seems to lead to other values which I don't know what I'm aiming for!
So filling out a speeds and feed calc for say hardwood @ 20,000rpm spindle I get min/max of 9-11 M/min speed. Assume this is the cutting speed and the rapid people refer to are movement between cuts.
I also used zapp calculator to get critical speed of 20mm diameter ballscrew with 750mm spacing between supports giving 1548 RPM.
From forums someone gives 2010 screw @ 1348 RPM = 2.15m/min, so would the motor need to at appox 6740rpm to achieve that?
But also the motor Voltage driven needs to be known to get an idea of rpm which can be achieved?
I also don't know the final weight of the gantry, I'd estimate say 20Kg based on extrusion weight + spindle and alum plate.

On another note - have I already got off to a bad start using the non-heavy type of aluminium profile. This was from KJN, I intend to add plate/strengthening to make it more rigid.
Just installed fusion 360 to see if I can learn to create my design in diagram form:)

Ryan

[AndyUK]

I'm trying to figure out whether 2005 or or 2010 ballscrew would be better, or 1605/1610 (which seems to be less available pre machined). I don't plan to use pulleys.

Meh, get a custom quote from Fred and he'll machine whatever you want, it'll still probably be cheaper than the pre-machined, and you'll have a quality product from a trusted vendor.

So filling out a speeds and feed calc for say hardwood @ 20,000rpm spindle I get min/max of 9-11 M/min speed. Assume this is the cutting speed and the rapid people refer to are movement between cuts.

Yep. No feeds and speeds calculator will ever give you rapid speed recommendations. Do remember that it will also depend on how many flutes your cutter has.

From forums someone gives 2010 screw @ 1348 RPM = 2.15m/min, so would the motor need to at appox 6740rpm to achieve that?

Then that someone is a tit and can't do maths. It was probably me. The 2010 means that the outside diameter of the screw is 20mm, and the pitch is 10mm. In one turn of the screw, the ballnut will move 10mm. So 1348 turns per minute equals 13480mm/min = 13.48m/min.

But also the motor Voltage driven needs to be known to get an idea of rpm which can be achieved?

Its less about the voltage, more about how much torque you have left when you get up to those RPMs, which then dictates how quickly you can accelerate and decelerate. Whilst I spent a lot of time doing the calculations, I still ended up with the same motors as everyone else with the machine size, so for starters find a decent build about the same size and steal their motor and PSU configuration :P Work forwards from that.

I also don't know the final weight of the gantry, I'd estimate say 20Kg based on extrusion weight + spindle and alum plate.

Quite machine dependent, but I'd guess you're lowballing it. A 2.2kW Spindle is approx 8kg alone. Rails + Ballscrews + Plates + Motors + Extrusion should be a good ballpark. Maybe nearer 30-40kg?

[JAZZCNC]

Hi Ryan,

Let me just say from the start that the Heavy Mill 3DTEK as got some serious issues and it's far from its name suggests.

First off scrap the Z-axis altogether it's load of rubbish and far too flimsy. The Z-axis is THE most important part of the machine because it's where all the action happens. If it's weak then it doesn't matter if the rest of the machine is built like a tank your cutting will always suffer.
Also, no point using Hi-win profiled linear rails on other axis if going to use a round rail on Z-axis, use either all-round or all profiled. You are wasting your money mixing them.

Next is the Gantry sides, they are also weak and flimsy. They will resonate and flex which again transfers into the cut. To be honest the whole gantry is on the weak side and could easily be made stronger with a design change. Even the Light profile will work if you used an L shaped design.

Regards the ball-screws you will be better with 16mm x 10mm pitch for a wood router at this size. 10mm pitch will easily allow you to cut aluminum without any problems. Whereas 5mm would be too slow for cutting wood correctly.
You don't need 20mm for a machine this size and it will only rob you of power and speed because of the extra inertia of the ball-screw. It takes more power to accelerate and deaccelerate than 16mm and it does make a big difference to performance.

Regards the electrics and controller then always go with Ethernet if you can afford the extra, it's much more stable than USB or parallel port.

Andy mentions Voltage isn't so important but I'm afraid he's wrong, It's very important to the speed and torque you'll get from a stepper motor. The torque you get from a motor is proportional to voltage so if you increase the voltage the torque will also rise, up to a point. To get high speeds from a stepper you need voltage which is why we try to use higher voltage drives so we can boost voltage. Obviously there is a limit dependant on motor/drive specs etc but in general, more volts = higher rpm which is better for a router.

Regards the Motor, drives and PSU spec then there is a well-proven spec for machines this size which has been used for a long time on this forum. Which I'm happy to help you with. However, I'd also advise you to consider looking at Closed-loop steppers as they have come down in price quite a lot and are much better than the standard stepper systems. This could change things a little regards PSU etc depending on which you go for. So go have a look at them if you haven't already and let us know your thinking.?

BIGGEST ADVISE is don't buy anything else until your 100% sure it's correct. Also don't buy electrics until needed as your wasting warranty and things change fast these days regards controllers/software and drives you could end up with a better system by waiting until needed.

Don't be afraid to ask questions no matter how dumb you may think they sound.

[CNCRY]

Thanks for bearing with me, now I have the whole idea in my head I'm keen to start building , but want to do it right first time so remaining patient, and re-thinking the design.
Is Fred the BST automation aliexpress store? What sort of lead time do people get with him and is there a ballpark % amount of import duty etc to be expected?
That RPM calculation was from a U.S forum in IPM, and was clearly wrong then. Thats clear now, so with the 10mm pitch I'd need to aim for 1000rpm to get 10m/s cutting speed, which should be about right for a 2 flute cutter in hardwood. While Aluminium would need to aim for 5m/s or 500rpm at the motor?
So from most similar builds I have seen everyone ends up with 3/4NM Nema 34 motors, but there is variation in how they can be driven.
Trying to pin down some aspects so I can then design around those.
You are right on gantry weight . Profile is 8kg + 8Kg spindle + perhaps 10kg rails and carriages + aluminium plate + motor - more like 35Kg.
Ryan

[CNCRY]

Hi Ryan,

Let me just say from the start that the Heavy Mill 3DTEK as got some serious issues and it's far from its name suggests.

First off scrap the Z-axis altogether it's load of rubbish and far too flimsy. The Z-axis is THE most important part of the machine because it's where all the action happens. If it's weak then it doesn't matter if the rest of the machine is built like a tank your cutting will always suffer.
Also, no point using Hi-win profiled linear rails on other axis if going to use a round rail on Z-axis, use either all-round or all profiled. You are wasting your money mixing them.

Next is the Gantry sides, they are also weak and flimsy. They will resonate and flex which again transfers into the cut. To be honest the whole gantry is on the weak side and could easily be made stronger with a design change. Even the Light profile will work if you used an L shaped design.

Regards the ball-screws you will be better with 16mm x 10mm pitch for a wood router at this size. 10mm pitch will easily allow you to cut aluminum without any problems. Whereas 5mm would be too slow for cutting wood correctly.
You don't need 20mm for a machine this size and it will only rob you of power and speed because of the extra inertia of the ball-screw. It takes more power to accelerate and deaccelerate than 16mm and it does make a big difference to performance.

Regards the electrics and controller then always go with Ethernet if you can afford the extra, it's much more stable than USB or parallel port.

Andy mentions Voltage isn't so important but I'm afraid he's wrong, It's very important to the speed and torque you'll get from a stepper motor. The torque you get from a motor is proportional to voltage so if you increase the voltage the torque will also rise, up to a point. To get high speeds from a stepper you need voltage which is why we try to use higher voltage drives so we can boost voltage. Obviously there is a limit dependant on motor/drive specs etc but in general, more volts = higher rpm which is better for a router.

Regards the Motor, drives and PSU spec then there is a well-proven spec for machines this size which has been used for a long time on this forum. Which I'm happy to help you with. However, I'd also advise you to consider looking at Closed-loop steppers as they have come down in price quite a lot and are much better than the standard stepper systems. This could change things a little regards PSU etc depending on which you go for. So go have a look at them if you haven't already and let us know your thinking.?

BIGGEST ADVISE is don't buy anything else until your 100% sure it's correct. Also don't buy electrics until needed as your wasting warranty and things change fast these days regards controllers/software and drives you could end up with a better system by waiting until needed.

Don't be afraid to ask questions no matter how dumb you may think they sound.

Appreciate the advice JazzCNC
Heavy Mill was my initial idea as I could see how it was constructed, plus I assumed must be a sufficiently good design to sell:) I'll not be copying it's Z axis in that case, and will design on the same rail setup instead. As I just mentioned in reply to Andy - im's trying to establish some "fixed" ideas and design around that, but X changes so does Y!
So one thing I did notice on heavy mill were weak side's so my design there is similar to Andys log = http://www.mycncuk.com/attachment.ph...3&d=1573338796 . That is rails on top on the extrusion, then a carriage and aluminium block which the gantry rail site on top of,and fixed using side plates .

I'll be looking into 1610 ballscrew based on both the replys now.
Do closed loop steppers work with standard nema motors ? I'll take a look.

I'll be trying to draw this in a shareable format soon. But design steps wise, can I design/build the X/Y axis and gantry to a good standard, then consider the Z axis as the next "module" or is that a bad idea!
If I know the motors to be used, I can pretty much leave electronics until the build progresses, and certainly won't buy anything related to that yet.
Will my existing 120 x 40 profile with 40x40 cross pieces suffice for X/Y axis with extra bracing?
And with the L design for gantry - just thinking how I can incorporate the 90x90 profile, perhaps a 60x60 attached to it at the back?

Ryan

[Kitwn]

Is Fred the BST automation aliexpress store? What sort of lead time do people get with him.
Ryan

Ryan,
Yes, Fred is BST Automation. I fell of the sofa when my leadscrews and rails arrived in remote western Australia only 4 days after I paid for them. It often takes longer than that to get stuff from the state capitol Perth. You do pay quite a lot for delivery though and it's worth checking the different total cost for item + postage and 'free' postage for the same thing.

Don't forget that many Chinese businesses are closed at present because of the virus so delivery times at the moment are anybody's guess.

I don't know about import duties to Pommieland but make sure you know the total cost before comparing suppliers.