[GeckoSub]

Hi all,
Would love some feedback and apologies up front for being wordy. Will intro the idea and then ask specific questions in the posts to come - but feel more than free to comment on anything



Intro
I've been looking to build a small mill/router for years but stuff got in the way. But now, it will be happening.
I am not an engineer nor machinist but I tend to research a fair bit - that said, I am not the best with deep dives into math so I do end up relying on anectodal "evidence" quite often.

As such, my design so far is a result of what I have picked up here and there and I am looking for feedback on it.

Goal
Goal is to machine aluminium and hold decent tolerances, perhaps +/- 0.025mm. There's a dream to cut titanium one day, but I know it may not be able to (stiffness, lack of flood coolant and wrong spindle) so we can put a pin in it for now, but if it could - even slowly, that would be a bonus.
I have a small product I want to prototype on this machine and make smaller batches of. Speed is not super important as if this takes off, I would outsource the machining anyways.

Budget
Budget is a semi-tight which might confuse you given some of the specs below like the C3 ballscrews but I can source quite a lot of this used and even a few things at very good deals. But think "budget, best bang for buck, but not opposed to spend a bit on good parts".

Granite, really?
I am chosing to have a base and gantry made from solid granite. I used to be based in China, have friends there and can get it to the Philippines where I live now with cheap freight despite it being heavy.

I know granite is not as stiff as steel but please don't fight me on this - this decision is set in stone;-). Why? Because it will be machined to 00 accuracy by a factory that makes surface plates and granite machine bases. They will also glue in all the screw inserts where I specify. So, it shouldn't need any secondary machining and with a granite precision angle I should be able to align all the motion parts myself. Bascially, I need to limit the risks of me effing this up and this is a shortcut to that.
And as has been pointed out - I owe a shout out to Piotr Fox and his surface plate build. Without it, I wouldn't have gone down this path

Size:
Travels
X: 230mm
Y: ~400mm
Z: ~ 230mm (under gantry clearance) but TBD...

Granite side wall and bottom thickness: 60mm

Alu parts (gantry and carriage parts) are 24mm thick

Moving Gantry, really?!
Yes... I have read a lot of accounts on how a moving table/fixed gantry machine would be stiffer, but I am willing to make a compromise here. I don't have a lot of space so the overall size has to be kept down. I also like the idea of the high side walls and plan on having end plates in acrylic to keep the worst of the chips contained. This wont be a high production machine so hopefully this approach might work - also for the same reason I am not too worried about the semi exposed motion parts.

As for the high side walls, the Z-clearance ends up quite generous at ~230mm right now. I will see if I can reduce this but I have a few fairly tall parts that I have in mind to machine. And if I need more stiffness, I guess I can put the vise on a stiff riser block to have the spindle operate at a height where it's assembly is stiffer?

Motion
Linear guides: 25mm
Ballscrews: 2005 (C5 minimum, possibly C3)
Closed loop servos, potentially servos...

Speaking of motion parts, most of this will be secondhand or NOS Japanese.
E.g. it seems like I can find some decently priced C3 ballscrews and have them modified to my specified length.

I was initially going to go with 2.3nm Leadshine closed loop steppers since I was afraid of tuning servos, but from various posts, it seems Deltas tuning is not that difficult and since I can get used B2s on the cheap, I might go with servos after all.

Spindle
For now, your regular Chinese 2.2kw, 24K rpm powered spindle.
Given that dream - with the pin in it - to machine a bit of Gr. 5 Titanium I thought about a 12K or 18K rpm spindle with a bit more torque at lower revs but it would be like sitting between two chairs, I think? For now, 24K Chinese spindle and then later if this machine has made money I can look into a proper servo driven low speed spindle.

Controller
Leaning towards UCCNC.
Was thinking of LinuxCNC for a long time, but too much tinkering in a field I know nothing about.
Masso, Acorn - too expensive.
Mach3/4 - not interested
Chinese standalone controller - I think for now, I would like to have a bigger screen with an easier way of setting up parameters, probing, etc.

[GeckoSub]

More Pics, More Thoughts
Initially, I wanted to package the steppers/servos under the ballscrews (or behind the gantry) but that would need pulleys and belts and to stick with the mantry of keeping this build as simple as possible I am going with direct drive motor brackets in most places.

Here's one side of the Y:


It also hints at the X motor not being a in a bracket but rather mounted on the side plate of the gantry:



Also, that first view shows an aluminium mounting plate holding the ballscrew, motor and brackets to the granite side wall. Not sure this is 100% needed but my thinking was that I would have more freedom to move the mounts around if need be - in case they don't fully align with the holes in the granite that the machine base factory will be making. It's not that I don't trust their precision, but it's more that I am not yet 100% sure of the specs and dimensions of the ballscrews...


For the same reason, I have some elongated screw bores on the gantry plates to allow some adjustment:

[Momentz]

Sounds like you've got it all sorted.

So basically your building what Piotr fox has already built.

Sent from my SM-G781B using Tapatalk

[GeckoSub]

Sounds like you've got it all sorted.

So basically your building what Piotr fox has already built.

Sent from my SM-G781B using Tapatalk

YES!
And cudos to Fox BUT I felt the need to take the huge shortcut of having the base done for me. Less hassle and I think I have found the right factory to make the base but I am the guinea pig. Worst case scenario, I still need to do all the grinding and sanding he did...

But yes, yes, Piotr is without a doubt the reason, I am venturing down this path. I spent a long time thinking extrusions were viable, then thought I should do Epoxy-Granite but solid granite it is - I see no other, easier way to obtain the precision I am hoping for.

Also, those gantry plates are very much copied from him... haha.

[GeckoSub]

Question - Z-assembly, Moving Rails or Moving Carriages?
I had drawn up a regular moving carriage design but I guess moving rails would be more stiff given that long clearance I have. I try to draw up the Moving Rails and post it up as soon as I have time.

It's also to do with the overall Z height which I am still a bit undecided on. I think I need to think up the longest part I would need to machine and then see if I have room to reduce the Z a bit.
90% of what I will be making is not tall parts. But there will be the odd one. E.g. some waterproof battery holders for 18650 cells which would ideally only be machined from one end.

[GeckoSub]

Question - Ballscrews, Linear Rails and Base Stiffness?
I mentioned most of my choices come from reading and seing what others have done and not math... So, the ballscrew choice of 2005 and 25mm rails are a result of that.
Does that sound about right from what I want to achieve (aluminium at a decent speed) and as a match to the machine base/frame? I mean, I could go up in size on both but maybe the frame is not stiff enough to warrant that at all?

The wall thickness of the granite is 60mm. Gantry is 60mm thick, too. I guess I could go to 70mm if that helps a lot in terms of stiffness. I am honestly just assuming 60mm is more than enough for aluminium but with that titanium hope lingering (yes, with coolant and another spindle) would 70mm possible somehow make a big difference - the only reference I have is "the baliscissors guy" machining titanium on a Tormach 1100 and not sure how stiff that machine is in relation to mine...

[GeckoSub]

(Not So Much A) Question - Limit Switches and Homing
I am hoping to get secondhand Metrol switches which have 5um accuracy to use for homing. Don't know if they are too banged up, but I will likely give them a go.
But almost all of the ones I can get are Normally Open and I read that NC is vastly preferred if a cable breaks, a connections goes bad, etc.
Would you risk it with an NO switch or is there some really easy curcuitry magic that can be done to make an NO switch behave like an NC?

I was hoping to home to the zero pulse of the servos - and actually even the Leadshine CL steppers I have in mind have zero pulse - but it seems UCCNC does not offer that feature(?).

[GeckoSub]

Question - Delta B2 Servos, Which Size...?
B2s are abundant on the secondhand market and affordable but I see they come in low, medium and high inertia models... I have zero idea what would work for a small, but heavy for its size machine like mine?
Same goes for their power.
Again, anectodally, I felt like 400w would be alright, perhaps even just 2 x 200w for the tandem Ys. I'd be surprised if 750w was called for and I know I am asking a lot not having done or being great at the math but could one of you possibly stick your neck out and venture an educated guess on my behalf?

[GeckoSub]

Question - Z Counterweight, Gas Struts, Springs or Motor Brake...?
As the title says
Is motor brake easy enough to implement?
How does it actually work? The brake clamps locked when power is cut...?

If not brake then what's you favorite alternative?

[GeckoSub]

Question - Three or Four Feet For Support?
For leveling this thing on top of a sturdy table, bench or stand, what's preferred, three studs or feet or four...? I see some big machines use three and maybe many/all big surface plates, too.
(I have to add this to the design of the base so the factory can drill the holes and glue the inserts in)

[GeckoSub]

YES!
And cudos to Fox BUT I felt the need to take the huge shortcut of having the base done for me. Less hassle and I think I have found the right factory to make the base but I am the guinea pig. Worst case scenario, I still need to do all the grinding and sanding he did...

But yes, yes, Piotr is without a doubt the reason, I am venturing down this path. I spent a long time thinking extrusions were viable, then thought I should do Epoxy-Granite but solid granite it is - I see no other, easier way to obtain the precision I am hoping for.

Also, those gantry plates are very much copied from him... haha.

Sorry, didn't mean to say "YES! I have it all sorted", haha. Far from it.
I meant yes to Fox's build being the main inspiration for mine

[peteeng]

Hi Gecko - I think 5mm pitch is fine for Z but too small for XY. 10mm or bigger for XY is the go. You will need more feed speed with the fast spindle to get correct chip thicknesses than 5mm can provide. I use braked motors for the Z and they are really good. If you use balancers, weights etc they take up too much space... Peter

re feet use 3 for levelling but have a couple for stabilising once levelled. in regard to gantry ht and gantry motion you won't regret extra height. Once you have a vice and some material and a tool the height is needed. Stud=y7 some commercial mills to figure out what they do...

If your machining parts on a commercial CNC do not allow for adjustment. You should be able to make it exactly as required off the mill. Also look at using dowels to locate parts vs relying on friction only. Dowels make a huge difference in stiffness. Something like this.

https://henryeng.com.au/product/dowel-pins-5mm/

[pippin88]

This is a servo level build, don't waste it with steppers.

5mm pitch is fine on screws less than 1 metre long and with servos (servos go to 3000 to 6000rpm)

Z Axis moving cars vs moving rails.
Beware the clearance issues of moving rails
You need to get the tool tip up over the work piece.
If you have a 200mm tall workpiece, and a 100mm tool stick out (distance from spindle nose to tool tip), then you need 300mm of Z travel / space - you need 300mm from the table to the bottom Z car. So side walls need to be taller / gantry higher above the table.

It is very easy to have fixed Z rails with moving cars go taller, this allows you to bring the spindle nose above the bottom of the gantry and thus get the tool up over tall work pieces.
Essentially moving cars means you can have shorter side walls.

[GeckoSub]

Hey Pete,
Great to see you weigh in, was hoping you would as you are in a lot of the threads I have been studying.

Leveling and support
With you on the 3 feet for leveling and once there, a bit more support - that should be easy to incorporated into the design.

Dowel pins and need for adjustment
I think I can call in a favor from a friend at a factory in China with some big CNC mills to make the alu gantry parts for me so they should come out with pretty tight tolerances.
Hadn't thought of dowel pins at all - will put that on the to do list

Ballscrew pitch
As for the pitch on the ballscrews, if both 5 and 10mm would work well enough with servos, I may just let the decision be driven by which one I can get the best deal on in terms of price and grade. I have only been searching for 2005 but maybe I have more options in 2010. I intend to look for used Japanese ones if they aren't too beat up, but I can buy them oversized if need be as the fee for modifying them is very affordable.

[GeckoSub]

Likewise to Pippin,
Happy to hear from you

Servos vs. Steppers
Yes, once I saw the price on used Delta B2s and factored in, I don't need a power supply for them like I would the steppers, I was +90% making the switch to them. I just recall scary stories of tuning, but I have since read that the name brands ones tune easily enough. And I am glad the machine warrants using servos.
Would you say 400w is ample? Perhaps just 200w each on the Ys? (That said, 400w are everywhere and 200w a bit more scarce on the secondhand market in China.)

Moving Rails vs. Moving Cars
Perfect with your examples, just need a bit of time in CAD/on paper and/or more sleep or more coffee to picture this and the relation to the side walls. Will get back once I have it drawn up and visualized

I have a self-imposed deadline of about two weeks to finalize the design on the machine base before I send off the drawings. That said, it's a pretty simple design. It's more about making sure I don't mess up anything or have any glaring omissions.

E.g. I am thinking of extending it just a tad in Y, just enough that the machine can cut new Z-assembly parts down the line if needed. And the Y can be extended without the machine losing any rigidity.

[joeavaerage]

Hi,

Would you say 400w is ample? Perhaps just 200w each on the Ys? (That said, 400w are everywhere and 200w a bit more scarce on the secondhand market in China.)

There is a correct size for servos, its the momentum equation, and is highly dependent on the diameter of the ballscrews, and only moderately dependent on the pitch, and even less dependent on the mass
of the machine.

I use 750W Delta B2's. I might possibly have got away with 400W servos, but for an extra $40USD I could have 750W....so I got them. They sparkle.

I use 32mm diammeter double nut (BNFN) 5mm pitch C5 screws by THK, they are 750mm long. The axis mass is 150kg (cast iron axis bed, plus part plus workholding).

The momemtum balance is:
Total momentum = 80% rotational momentum of the ballscrew + 12.5% rotational momentum of the armature of the servo + 7.5% axis mass.
Inertia ratio =100/12.5= 8
So despite the axis mass being fairly heavy (150kg) its still only a small fraction of the overall momentum which is dominated by the ballscrew.

You need to do the momentum calculation, anything less is a pure guess, and that's not good enough when spending big money.
The dependency of rotational inertia of the ballscrew is proportional to the fourth power of diameter. The axis mass momentum is proportional to the square of pitch.

I just recall scary stories of tuning, but I have since read that the name brands ones tune easily enough.

If you use Delta servos, or another brand like DMM, or any of the top tier servo brands like Yaskawa then all have setup and tuning software.Do not settle for any brand
the DOES NOT have set up and tuning software. If you've not used servos before you need it. The nightmare stories you've heard are newcomers whom try to tune servos without the benefit
of set up software.

Almost all software has an Autotune feature. For inertia ratios of 10:1 or less Autotune is near perfect. For inertia ratios of 10:1 or greater, and especially 20:1 or higher then Autotune is at best a start,
you'll have to manually tune. As the ratio gets higher the difficulty increases. You need to do the inertia calculation to avoid walking into an impossible tuning situation.

If there is one thing to take away from this post is the the DIAMETER OF THE BALLSCREWS IS CRITICAL, because it determines the rotational momentum of the ballscrews which commonly dominate
the momentum balance equation.

Posted a pic of my machine, its cast iron and steel, because I wanted it as stiff as I could make it affordably. Nearly finished my trunnion fifth axis. Weighs about 85kg, so will add to the axis
mass I used in the momentum calculation, but even were I to double it the total axis mass momentum (now including the trunnion fifth) would still only be a modest fraction of the total momentum.
Just goes to show how large diameter ballscrews dominate.

Craig

[joeavaerage]

Hi,

Question - Z Counterweight, Gas Struts, Springs or Motor Brake...?
As the title says
Is motor brake easy enough to implement?
How does it actually work? The brake clamps locked when power is cut...?

Do yourself a favour, If you are of the opinion that you need a Z axis brake due to the mass of the Z axis, buy a servo with a built in electromagnetic brake. The Z axis 750W Delta B2 servo
pictured above has such a brake. Its magnificent and so easy. Its natural state, ie un-powered is 'brake engaged'......so you might say its fail-safe. Non braked 750W Delta B2's cost me $438USD each (new,
and the braked 750W Delta B2 cost me $542USD (new), so an extra $104USD, money well spent in my opinion.

No counterweights, no gas struts, no nothing, don't need it. The mass of the Z axis approx 50kg, requires 50/125=0.4Nm of torque from the Z axis servo to hold it up, and the servo is rated for 2.4Nm, so only
a small fraction of the available torque is required. This is one distinct advantage of 5mm pitch screws, great mechanical advantage, ie high thrust for minimal torque.

Craig

[joeavaerage]

Hi,

I was hoping to home to the zero pulse of the servos - and actually even the Leadshine CL steppers I have in mind have zero pulse - but it seems UCCNC does not offer that feature(?).

I use Mach4 and have done for eight years (light years ahead of Mach3). It does have 'Index Homing', and was my intent years ago to use it. I use Omron roller plunger snap action microswitches for Homes and Limits.
As it turns out with them I can get 0.02mm repeatability while homing......and that is good enough, or at least the extra work and wiring required to use Index Homing is not high on my list of things to do.
I rather suspect that with good roller plunger microswitches you too could get 0.02mm or better, and in which case Index Homing is not really required.

https://nz.element14.com/omron-indus...15a/dp/1500340

I do rather suspect that I will use index homing for both the trunnion and fifth axis previously pictured. I'll want very accurate and repeatable way to zero the fourth and fifth axes.
I'm hoping that if I can zero them well and repeatably that I can still use the mill as three axis, with the vise still sitting atop the fifth axis. The majority of the parts I make a small and
mostly brass, aluminum and plastics. If the parts are small enough I would not have to remove the trunnion off the bed of the mill.....and that would be a major time saver.

Craig

[joeavaerage]

Hi,
IF you are using servos, then B2's are rated to 3000rpm but have a maximum of 5000rpm.

With 5mm pitch screws that means 25m/min rapids. At rated torque my machine can accelerate all axes at 0.27g (rated) and 0.75g(overload).
That is what I set my machine to initially....but its too God-damned scary fast. I have detuned it to 15m/min rapids and 0.15g accels...and that's entirely enough.

If you used 10mm pitch you could go faster and potentially even accelerate quicker.....but personal experience tells me that would be just over the top for a hobby machine.
If you stick to 5mm pitch with servos it will be fast enough and you get better thrust.

If you are using steppers and they perforce go much slower then 10mm pitch screws make sense.

Craig

[GeckoSub]

Hi,

(...)You need to do the momentum calculation, anything less is a pure guess, and that's not good enough when spending big money.(...)
(...)You need to do the inertia calculation to avoid walking into an impossible tuning situation.(...)

Craig

Craig, I almost have the full band gathered now.
There are a handful of people I have been seen popping up in threads over the years that I was hopping would stop by here and you are certainly one of them.

It's early morning out here though and I am on my way to the ocean for a freediving session so will respond to your comments when I am back.

But for now, you have convinced me to do the math to check the servo power so thanks for the formula