[joeavaerage]

Hi,
I have made repeated purchases from this company in Korea. They specialise in CNC motion parts. I bought the 32mm screws I used in my machine. They were second hand, but when they turned up I'd swear they had never been
fitted, and even had matching serial number THK measurement certs, perfect in every regard. I paid $1000USD for three including six FK25 support bearings and three day Fedex to New Zealand, exceptional buying.
It made my new build possible, I just could not have afforded these quality parts new.

https://www.ebay.com/str/industrialp....m47492.l74602

This is a 20mm diameter C5, 5mm pitch by THK, five of them available:
https://www.ebay.com/itm/19448004260...cAAOSwWLphflwu

This is about the same price as a new C7 rolled screw....and a genuine ground C5 is sooooo much better.

Craig

[GeckoSub]

Thanks for the links Craig,
Looks like some good deals.
I have seen you mention that shop in other posts but I have the fortune of being able to dig into Chinese-only secondhand trading apps where the price is even a tad better (I don't speak Chinese but I have friends who are helping me with this). But selection and track record is of course not as solid as your Korean shop. Still, I think I can find a vendor or two that I can trust. Especially if I find a machininist who can do a good job of modifying them to length.
That said, I am aiming for C5 at least but keeping my eye open if a great deal on some C3 turns up.

[GeckoSub]

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

Quick question to clear something up. Do I understand it correctly that the 3000rpm is where the motor has "all" the torque which would be sufficient in cutting scenarios (with a well matched motor) and then for rapids you can make use of the 5000rpm?

I just grabbed some numbers from a Haas SuperSpeed and a Syil X5 and they rapid at 36 and 30m/min respectively and have max cutting feeds of 21m/min and 10m/min. Given that my machine is just a rather glorified hobby router, I do think the speeds I can get from 5mm sounds alright.

[joeavaerage]

Hi,
there are a number of Chinese suppliers claiming C5 and C3 but at prices little more than rolled C7's. Total BS. If you can find some suppliers that you can trust to supply
genuine C5's or C3's, but quite frankly China is not the best place for high end ballscrews.

Do I understand it correctly that the 3000rpm is where the motor has "all" the torque which would be sufficient in cutting scenarios (with a well matched motor) and then for rapids you can make use of the 5000rpm?

Yes. Many AC servos have that feature. It is a quirk of Feild Oriented Control motors, its called Feild Weakening. In the case of Delta B2s servos they retain rated torque up to rated speed. In the specific case of
my 750W B2's, 2.4Nm at 3000rpm. Beyond 3000 rpm and up to 5000rpm it has lineraly reducing torque.

Torque at 5000rpm =2.4 X(3000/5000)
=1.44Nm

That would be enough for a rapid traverse (g0), but probably not have the thrust for regular cutting (g1) I would say my machine has max rapids of 25m/min but max cutting of 15m/min.
Its somewhat academic as I very seldom ever use the highest speeds and accelerations my machine is capable of...its just too scary.

For a small to medium sized hobby machines I would say 10m/min to 15m/min is fast enough and accelerations of 0.1g to 0.2g will ensure good toolpath following.

For production machines these would be considered tool slow and anemic, where 50m/min to 90m/min and 1g and higher is considered normal. Mind you those
machines weigh many tons and cost hundreds of thousands if not millions of dollars.

I rather doubt 60mm of granite (E=70GPa) could reasonably contain the forces associated with 1g accelerations, it's just as likely to fracture. In fact, I'm not at all sure that grey cast iron (E=110GPa) would be enough......
I'd be using a good grade of SG iron (E=170GPa) at least, or casting in genuine steel and be done with it. Steel is at least three times stiffer (E=205GPa) than granite and has tenfold-twentyfold the impact strength of granite.

No, shy away from really high speeds and accelerations, your budget just won't stand it and your choice of basic material is wrong for that sort of machine.

5mm pitch ballscrews and servos in the 200W to 750W class of 2000 rpm or more will give you plenty of speed and accelerations with thrust to accommodate cutting forces.

Craig

[jaguar36]

Question - Ballscrews, Linear Rails and Base Stiffness?
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...

I would certainly go thicker with the granite base. 10mm more on the thickness almost doubles your bending stiffness. Flat plates like that are very poor in bending so making it thicker is hugely beneficial I would imagine the cost is going to be pretty minimal to make it thicker. I would go even beyond 70mm. Nobody ever said their mill was too rigid. Take a look at the overall weight of the granite and make it as thick as you can before you start running into handling issues.

For the the gantry, consider putting some steel plates on the front and back side of the granite. Even some fairly thin steel on the outside of the beam would have a big impact on its stiffness.

[GeckoSub]

Hi,



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

Yes, there's a good video on home switches for hobby machines by JBWorxStudio on youtube and his findings are very much aligned with yours. He really liked the small Panasonic optical switch, too.
I am taking the liberty of attaching a screenshot here from his website that shows his results a bit more clearly than the video:


The reason I was thinking of the Metrols was because in the Philippines where I am now, we sometimes have power cuts so was thinking it would be great to have amazing home sensor accuracy to restart or continue a job, but I guess in reality I would just re-probe, re-zero from the vise corners or something like that. Though to square the gantry with the two Y-motors it would still be great to have some really precise sensors.

[GeckoSub]

I would certainly go thicker with the granite base. 10mm more on the thickness almost doubles your bending stiffness. Flat plates like that are very poor in bending so making it thicker is hugely beneficial I would imagine the cost is going to be pretty minimal to make it thicker. I would go even beyond 70mm. Nobody ever said their mill was too rigid. Take a look at the overall weight of the granite and make it as thick as you can before you start running into handling issues.

For the the gantry, consider putting some steel plates on the front and back side of the granite. Even some fairly thin steel on the outside of the beam would have a big impact on its stiffness.

Granite Thickness:
Yes, I think since granite isn't that stiff to begin with, you're right - it makes sense to go up in thickness. And I think that since the factory will mill this base from a solid chunk of granite, that doing less milling workand thus leaving thicker walls may not cost a whole lot more.

Right now, as pictured above with 60mm walls, we are at 90kg for the base. So, it's already so heavy I'd need help moving this thing around. If I bump that to 75mm, I am at 115kg. Not that bad a penalty for a fairly big increase in stifness (sorry, if I mix up stiffness, strength, rigidity, etc).

I did see one other build like this where the customer had asked the factory to mill a pocket out of the base all the way at the rear of the Y - to save weight as the spindle can't reach down there anyways.

Like this pretty much:

But that would save less than 10kg in my case, so probably not worth the increase in cost.

What could possibly make sense would be to mill some channels out of the gantry plate for weight saving...? At 60mm thick it's at 16kg and at 75mm it will be 20kg. On the other hand, since this is now a servo build, the two Y servos probably dont care about having to move 2kg more each.

As for the steel reinforcement, let me mull that over. I am not opposed to it, but I would likely need a bunch of inserts to screw it down well and/or to epoxy it to the granite plate. More inserts drive the price up though. Right now, I am inclined to perhaps forego steel plates. After all, a new gantry plate is pretty much the simplest part of the machine in case I need to have a new one made down the line.

As a benchmark of sorts, Fox's granite machine has a way longer gantry span than mine and I think his plate is 50mm thick and he does great work in alu on that machine. Then again, there is that titanium dream lingering...

[jaguar36]

You could have them drill a bunch of large holes in the base to lighten it without effecting the cost much. They're probably very used to drilling holes for faucets and such and thus wouldn't charge a ton for it. Regardless though the difference between 90 and 115kg is pretty meaningless. 90kg is still going to be a struggle to move even with two people, you really need some mechanical help.

[GeckoSub]

Hi,
there are a number of Chinese suppliers claiming C5 and C3 but at prices little more than rolled C7's. Total BS. If you can find some suppliers that you can trust to supply
genuine C5's or C3's, but quite frankly China is not the best place for high end ballscrews.


Craig

Still reading up on your previous posts and reading spec sheets of servos, etc but just wanted to address this one

Thing is, I am indeed looking at used Japanese hardware just like you. At the very least, Taiwanese.
China is - for better or worse - the factory of the world so there is litterally tons of used gear around and a lot of that is proper Japanese hardware as a lot of the Chinese factories or their overseas owners had the money to buy German, Japanese or TW machines. So, the question is not one of whether it will be Chinese or Japanese or real C3/C5 or "Chinese C5", the real question is whether I can get some that have not been run into the ground.

Honestly, some years back, buying anyting used was not easy. I think there was a bit of modern, cultural bias towards always wanting the newest bling so what was on the used market seemed to pretty much have been "run until destruction".
But that has changed, there's no shame in buying secondhand now, it seems whether it's consumer eletronics, clothes or machine parts. And for the latter, certain industries are already moving to even cheaper places and with the slowdown in the economy leading to closures, I feel like that could be the reason I start seeing what appears to be more good quality used, hardly used gear and even a bit of NOS.

E.g. for the servos, I can get a set of B2 400w with driver and motor for about 100 bucks. The ones where the people disassembling didn't just snip the cables but actually kept them are a tad more but still nothing bad. This price makes sense as they have soooo many of them around. I even see some OK priced Panasonic and Yaskawa stuff, but no need for that for me. And then there's Estun which is 100% Chinese but seem slightly more expensive than Delta.
But I think, due to you, and the sheer numbers of them, I have settled on Deltas

[GeckoSub]

You could have them drill a bunch of large holes in the base to lighten it without effecting the cost much. They're probably very used to drilling holes for faucets and such and thus wouldn't charge a ton for it. Regardless though the difference between 90 and 115kg is pretty meaningless. 90kg is still going to be a struggle to move even with two people, you really need some mechanical help.

Faucets? As the things water come out of? I am lucky - I am not getting this from a place that makes kitchen counters but rather a place that makes surface plates, perhaps even CMM machines but definitely makes machine bases.
I could just make two sets of drawings and let the difference in price decide for me.

But yes, I need to think ahead on the handling once it gets here... Once it's on the stand/table it wont be moved easily.
I think the allure of aluminium extractions (yes, I went through that phase, too) was that I could disassemble it and pack it up and more easily ship it somewhere else. I am a bit of a nomad and not sure how many years I will be on this island. Worst come to worse, I will have to build another machine if it's not feasible to ship this one to where I might go next in a few years...

[joeavaerage]

Hi,
I think you are misunderstanding how Home switches work.

When homing the machine will move in the programmed direction at the programmed speed UNTIL it encounters the Home switch, whereon it decelerates to a stop and then backs up UNTIL
the switch deactivates. The axis is Homed when the switch deactivates not when it activates. So the accuracy or repeatability is not how much the switch travels before activation, but how
repeatable the hysteresis is, ie the travel until the switch deactivates.

So your Metrol switches might be best if you need to precisely define an edge or similar, and that I would call absolute positioning, but for repeatable homing then a switch with defined and repeatable
hysteresis is what you want. Such a switch might cause the axis to define Home as 0.435mm from the switch activation point but because it has defined hysteresis it will be 0.435mm plus or minus 0.01mm every time.
the absolute position is not that critical, but having a repeatable switch deactivation point is critical.

Craig

[peteeng]

Hi - Here's an interesting German company that makes granite machines and they provide lots of info. Peter

https://www.planolith.de/en/production/material/

Machine weight for a home machine is always an issue. Earlier you say that granite is not stiff. Not quite true. Its as good as aluminium and can be as stiff as cast iron. Unfortunately they do not test granite for stiffness often. I've tried to talk a couple of electronics guys into making an acoustic modulus tester so this would be easy to do. I'll have to start reading up on ardeono stuff to make one myself. Now before you get too deep into granite and if your going to do some finish machining in your country consider the same design in aluminium. Would look the same but billet Al can be machined accurately, can be made modular and is just as stiff as the granite. If your going to set up your granite for final machining then it may as well be AL and be machined. The set up costs will be the same and once in a machine the run costs are not much. Peter

granite is damper than aluminium and has a lower thermal expansion but for your purposes as a billet design it would be fine.

[pippin88]

Hi,
I think you are misunderstanding how Home switches work.

When homing the machine will move in the programmed direction at the programmed speed UNTIL it encounters the Home switch, whereon it decelerates to a stop and then backs up UNTIL
the switch deactivates. The axis is Homed when the switch deactivates not when it activates. So the accuracy or repeatability is not how much the switch travels before activation, but how
repeatable the hysteresis is, ie the travel until the switch deactivates.

So your Metrol switches might be best if you need to precisely define an edge or similar, and that I would call absolute positioning, but for repeatable homing then a switch with defined and repeatable
hysteresis is what you want. Such a switch might cause the axis to define Home as 0.435mm from the switch activation point but because it has defined hysteresis it will be 0.435mm plus or minus 0.01mm every time.
the absolute position is not that critical, but having a repeatable switch deactivation point is critical.

Craig

No, the controller dictates what it calls home

Some controllers (e.g. LinuxCNC) have different options for homing and whether it is switch activation or release etc...

[pippin88]

After you add gantry, spindle etc etc you machine will be well over 100kg even if the base is only 60kg.

Don't skimp on the base. Use a lifter / engine crane to move. 200kg is nearly as easy to move as 100kg if you are using a crane.

[joeavaerage]

Hi,

No, the controller dictates what it calls home

The explanation I've given is how Mach and UCCNC do it, and OP is leaning to UCCNC. Note also that both Mach and UCCNC are run on Windows platforms which
are not realtime. Homing requires realtime response to home switch events and consequently it is the motion controller board (which is realtime) and handles homing autonomously.
That is baked into the hardware and firmware of the respective motion controllers. While some of the homing behaviour, offsets, back-off distance etc are programmable both Mach and UCCNC
are both homed on switch deactivation, and that is not alterable by the user.

Craig

[GeckoSub]

Hi,
I think you are misunderstanding how Home switches work.

When homing the machine will move in the programmed direction at the programmed speed UNTIL it encounters the Home switch, whereon it decelerates to a stop and then backs up UNTIL
the switch deactivates. The axis is Homed when the switch deactivates not when it activates. So the accuracy or repeatability is not how much the switch travels before activation, but how
repeatable the hysteresis is, ie the travel until the switch deactivates.

So your Metrol switches might be best if you need to precisely define an edge or similar, and that I would call absolute positioning, but for repeatable homing then a switch with defined and repeatable
hysteresis is what you want. Such a switch might cause the axis to define Home as 0.435mm from the switch activation point but because it has defined hysteresis it will be 0.435mm plus or minus 0.01mm every time.
the absolute position is not that critical, but having a repeatable switch deactivation point is critical.

Craig

Still, a Metrol does that same job but with a repeatability that is even better. But this is probably a case of me wanting more than I need in real life and not yet having realized that +/-0.01mm is enough.
I mentioned I was hoping to be able to home the machine after a power cut and pick up the code where it left. But yes, +/-0.01mm is likely more than OK for that.

[edit] Ah, I re-read it all and I think you are saying that even though a switch has good repeatabilty on "switching on" it may not be the case for when it switches off. Got it. I would assume the Metrol would be great in both instances but would need to check if I go that way. After all, we know what they say about assumptions as a mother...

Not going the Metrol way would have the benefit of being able to more easily source NC switches and those Panasonic optical ones look small and come in many different configs: Some are right angled, some left, some straight and so on making it easier to make a clean and neat installation.
I will get a used Metrol for a tool setter, I think. They have a lot of those at good prices.

Actually, I still don't know how squaring the gantry off of two switches would work. Would I have to physically move the position of the switches so they both trigger at the point where the gantry is perfectly perpendicular? Or do you set them up such that they are close and then you can add a small offset in the software to get it "perfect"?

[GeckoSub]

After you add gantry, spindle etc etc you machine will be well over 100kg even if the base is only 60kg.

Don't skimp on the base. Use a lifter / engine crane to move. 200kg is nearly as easy to move as 100kg if you are using a crane.

Thanks for the reality check and that's a good point. Last night I did think about having some holes made high up in the side walls for a rod of metal to pass through to use to lift the thing. High up so the centre of gravity would be below the lifting point and reduce the risk of the thing keeling over when lifted.
I am on a provincial island in the Philippines and normally people solve the issue of lifting heavy stuff literally with man power, as in having the neighbors, friends, staff gather around and lift things. But I also have to admit, I haven't yet settled on where I will be having the workshop and need to figure that out as access should be thought through. But yes, I can borrow an engine crane fairly easily.

I am probably going to make another thread about this, but being on an island in the tropics worries me, too in terms of rust on the screws and rails.... I am close to the ocean, it's hot and humid so salt is in the air and things just rust here. It's honestly a bit scary.
I guess if this was running flood coolant, which it wont be, that would help. But I wonder what people do in these cases.
So far, my thinking is to get a small space with nice walls and nice seals on doors and windows (not a wooden house) and then run aircon whilst I am there and a humidifier when I am out. And use an oil or lubricant with known good protection properties.

[GeckoSub]

Hi - Here's an interesting German company that makes granite machines and they provide lots of info. Peter

https://www.planolith.de/en/production/material/

Machine weight for a home machine is always an issue. Earlier you say that granite is not stiff. Not quite true. Its as good as aluminium and can be as stiff as cast iron. Unfortunately they do not test granite for stiffness often. I've tried to talk a couple of electronics guys into making an acoustic modulus tester so this would be easy to do. I'll have to start reading up on ardeono stuff to make one myself. Now before you get too deep into granite and if your going to do some finish machining in your country consider the same design in aluminium. Would look the same but billet Al can be machined accurately, can be made modular and is just as stiff as the granite. If your going to set up your granite for final machining then it may as well be AL and be machined. The set up costs will be the same and once in a machine the run costs are not much. Peter

granite is damper than aluminium and has a lower thermal expansion but for your purposes as a billet design it would be fine.

Hi Pete,
Yes, sorry, I did read that it is very stimilar to alu in stiffness (actually in specific gravity, too) but I meant compared to steel as I was expecting a few people to say just have a steel frame made

But the basic idea is to not have to do any machining at all on my end. Basically, Lego it all together once I have all the parts assembled.
The factory that makes this should be able to get the surfaces to 00 tolerance (not sure if that's AA) but it's very decent and then get the walls and rail surfaces parallel to each other, too. They make structures like this and surface plates for a living.
Also, I am almost dead certain I can't find a machine shop in this whole province that could machine parts to any satisfactory precision. That's why the granite approach is attractive.

As for the factory and their capabilities, the screw inserts will be glued in by them and don't have to be super, super precise as the screw bores in the rails and such have a few 1/10mm of clearance. But I am not sure how I could make the dowel pins work. I mean, in the alu plates being bolted to other alu plates, sure that should be doable. But a dowel pin into the granite and then into an alu plate - Not sure how to go about that. Perhaps, the factory can glue in solid insert and then drill and ream it afterwards in a known, precise location. I will ask them about that.

[peteeng]

Hi Gecko - Dowelling granite to alum is probably not the go. For parts that are not to come apart they can be bolted and epoxied (if you wax the surfaces they can be separated later). If they are to come apart use a wicking loctite like 290. This can be applied post tightening and works really well on lots of materials. Peter

[joeavaerage]

Hi,
no, optical ones are not the correct choice. Firstly they are optical, and when you get working there'll be chips and coolant all over the place an optical is bad in those circumstances.
Second issue is hysteresis. See if you can find a specification about how much an optical switch has to back off before you get reliable switch deactivation. The same applies with
Hall switches and proximity switches, you cant really tell when the switch is going to deactivate as the magnetic/ferrous target withdraws.

I use roller plunger mechanical switches. You see the the switch is mounted on slots and allows the switch to be slid back and forth a little. The roller is activated by a shallow ramp
that is attached to the axis. It too is slot mounted, but especially because of the shallow ramp angle the adjustment is quite fine. Another important feature is that if the axis over travels for whatever reason
the switch will not be wiped out or damaged as the ramp alone passes by without ever impacting the switch nor does the ramp cause the switch to overtravel and get damaged.

Actually, I still don't know how squaring the gantry off of two switches would work. Would I have to physically move the position of the switches so they both trigger at the point where the gantry is perfectly perpendicular? Or do you set them up such that they are close and then you can add a small offset in the software to get it "perfect"?

Yes, that is about it, adjust the switches to be as square as you can and use software for the last bit. Not sure how UCCNC handles that but I do know how Mach4 and the Ethernet SmoothStepper handle
it.

Mach4 has two programmable features to do with Homing, namely Offsets and Back-off. They are applied to each home switch. So with a gantry you'd have a master motor with its home switch, and a slave
motor with its own home switch. Both motors are home simultaneously but each will have (in addition to whatever slight mounting discrepancy between master and slave switches) but will have individual Offset and/or
Back-off applied for that last few um.

These switches are about $20USD depending on where you buy from and with care can get repeatability of 0.01mm, I regularly get 0.02mm. The Metrol switches you have proposed may result in better absolute accuracy,
but really who cares....what you want is repeatability and that can be had for less money.

Craig