[joeavaerage]

Hi peteeng,

My Epoch project has gone full circle and now I can see it easily being 3,4 or 5 axis within the same basic structure.

Easy to say, lets see how it actually works out in practice.

I can tell you from personal experience the advantages and disadvantages of a column design, and I don't have to explain it so much as walk into the workshop and take a photo to show you.

Craig

[peteeng]

Hi all - Heres a nice table of small mill specs.

https://www.ajax-mach.co.uk/mini-cnc...lling-machine/

If your going column like these go 2:1 on the envelope plus you need room for clamping. If you go gantry go square... Peter

[G-Spot]

Hi peteng,
just scanning through the Hass PDF and it looks like a standard (BT40) VF1 is 3200kg while the VF6 breaks 9500kg. Both are still very heavy but not quite 11.5T either.

Craig

Yes most of the Haas machines are under 5000 kg and these figures are the shipping crate figures so the actual weight of the machine must be a few hundred Kg less.

CM-1 950 kg
DT-1 2336 kg
DT-2 2427 kg
SMM 1820 kg
TM-1P 2050 kg
TM-2P 2360 kg
TM-3P 2815 kg
VF-1 3550 kg
VF-2SS 3540 kg
VF-3/4 SS 5316 kg
VF-5SS 5760 kg
VF-6SS 10100 kg
VF-7 11600 kg

https://www.haascnc.com/content/dam/...428_rev_dn.pdf

[peteeng]

Hi PIO - enable your PM's Ta Peter

[piotr07da]

Hey,

Thanks for the suggestions. I tested my initial design in many different configurations, so for now, I will stick with my initial travels of X: 400 mm, Y: 300 mm, Z: 380 mm, and I will model steel-C and steel-gantry with those values. By the way, although the Z travel is 380 mm, the distance from the spindle nose to the table ranges from 100 mm to 480 mm. I did not mention that at the beginning.

peteeng
Okay, I have just enabled it.

[peteeng]

Morning Pio - The chase for machining steel using a medium size mill in a 3rd floor apartment, what a set of parameters! So interesting!
Datron is worth a visit. I think that style of machine is where your headed. Are you sure the neighbors will be OK with the noise? Anyhow keep at it... The Datron NEO is their smallest. Fits thru a doorway weighs 700kg. Same envelope as your aiming at but shorter Z.

https://www.datron.com/hard-milling-on-cnc-machines/
https://www.datron.com/resources/blo...urface-finish/
https://www.datron.com/resources/white-papers/

A typical apartment max floor loading here would be 2kPa. That's 200kg/m2. The neo has 1m2 footprint and weighs 700kg. So its 7000/1=7kPa Hmmm
Maybe the Neo is a good muse- https://www.datron.com/cnc-machines/datron-neo/

https://www.youtube.com/watch?v=ZgqCY3gUHcM

edit - lightweight machines https://www.datron.com/resources/blo...ctor-trailers/

Peter

[mach_crafter]

Going for those ambitious cutting parameters on a small concrete CNC might be pushing it a bit, especially with tough D2 steel. Maybe ease into it with slower rates and see how the machine holds up?

Your FEM results are showing really minor deflections, which is great. Just keep an eye on those during your rough cuts to make sure they don't start affecting your tool life or the accuracy of your parts.

With a 1.5 kW spindle and strong stepper motors, you're well-equipped to handle different jobs. That setup should give you some leeway to experiment safely without straining your machine too much.

And hey, making sure your machine stays sturdy under all that stress is key.

[joeavaerage]

Hi,
if milling steel is your goal then your choice of spindle is not easy.

Most hobbyists go for high speed/low torque asynchronous spindles driven by a VFD, typically Chinese made. To be fair to them they are very good, low to very low runout and good power delivery and
remarkably cheap. The problem is that most are 24,000 rpm, with some 18,000 rpm and fewer at 12,000 rpm. All of these are rather too fast for steel while they are good for aluminum, brass, plastics and wood.

As a benchmark: an uncoated carbide tool in mild steel requires a surface speed of 100m/min or less, and that assumes effective cooling. Tough steels like D2 that will drop to 70m/min or less.
Good coatings (AlTiN, TiN etc) will lift that surface speed by about 25%. Thus a coated carbide tool in mild steel the max surface speed is 125m/min. I run1/8th Raptor Destiny Tools, just a good brand
and well coated at 148m/min, but it reduces the tool life to about 4 hours in mild steel. Its acceptable, but at $25USD per tool its only just acceptable. I'd need to slow down to say 100m/min
but then I'd get 8 hours or so.

Lets take a 12mm end mill, not small but certainly not large either.

100m/min: RPM=2653rpm
125m/min:RPM=3316rpm
148m/min:RPM=3922rpm.

As you can see none of the high speed spindles are even remotely in the hunt. You can run asynchronous spindle at part speeds, but they overheat badly. Even a water cooled spindle its
recommended not to run it at less than 1/3 rated rpm. Thus with a four pole spindle rated at 12000 rpm, you can JUST go slow enough to run a 12mm tool at 148m/min,
but you have to dip below, even well below, the recommendation to get down to 100m/min.

The upshot is that if you want to cut steels with a high speed spindle you need small diameter, say 1/8th, tools.....and that in turn severely limits the amount of material removal.

I have an 800W 24,000rpm (rated), 0.3Nm (rated)spindle (Mechatron GmBh) and I've used it and used it over 10 years, and its been fantastic.......but its bloody hopeless in steel. I do use occasionally for steel,
but its very very slow and tedious and chews through tools. It was that poor in steel (perfect for aluminum, brass and engraving, which is mostly what I do) that I built a second
spindle especially for steel, ie low rpm high torque. I used a second hand 1.8kW Allen Bradley AC servo, 6.1Nm (rated), 18Nm (overload), 3500rpm (rated). Take about ten minutes to swap out.
It gets quite a bit of use. Now I can use 10mm, 12mm and 16mm tools and get 8 and 10 hours tool life in mild steel with flood coolant.

More recently I've bought a new Chinese ATC spindle 3.5kW (rated), 3.3Nm (rated), 10,000rpm (rated), 40,000rpm (max),380VAC, water cooled with an HSK32 tool interface. I'm still assembling the ancillary equipment
like seal air and taper blow solenoids and cooling components..... so have yet to use it 'in anger', but without cutting loads its beautiful, even at 40,000rpm its still quieter than the water pump!
Importantly I hope to be able to run this spindle at 1/4 rated rpm with good security, or 2500rpm. That will mean that I can use a 12mm tool in steel (with coolant) direct in this spindle. That is a significant increase
in (steel) capability over what I have had before, and yet still have a marked increase in top rpm's. Really looking forward to it.

You really need to give some thought to a spindle. If cutting steel is your aim then that dictates certain choices, but those choices are not the best for aluminum. If however you chose a solution that
works well with aluminum then its likely that it will do poorly or even very poorly in steel.

Using a spindle of similar type to my new spindle will get you a bit of both....but they cost....and anything much over 3kW is three phase, not, I imagine that you have in an apartment.

Craig

[piotr07da]

Hi,

peteeng

Regarding noise - surprisingly, this building has quite good acoustics. Once I asked my neighbors on the same floor if they heard any noise because at that time I was building a wooden workbench and using a lot of a thickness planer and a router. They said it was so silent that they sometimes wondered if I was still alive. I also wonder the same about all my neighbors. Nevertheless, I plan to build a sound-suppressing enclosure and put dampeners under the machine feet.

As for the weight of the machine, my understanding is that the maximum floor loading is specified as an average value per square meter, not per individual square meter, because otherwise every floor in every flat would collapse. I also plan to place the machine close to a load-bearing wall.

Datron is interesting, and thank you for the article explaining their thinking behind that machine.

mach_crafter

Thank you. I am considering making the machining parameters less aggressive, but for now, I treat those 300N of cutting force more like a benchmark and something I am aiming for.

joeavaerage

Thank you for your insight. I have been thinking about it for a long time. I plan to use a Chinese ATC spindle with a belt drive and a 2800 RPM AC induction motor. I am considering a way to easily change ratios, let's say between 1:2.85 and 1:1, to achieve a maximum of 8000 RPM at the cost of losing some torque, and a maximum of 2800 RPM. Or something like that

Something like this:
https://www.aliexpress.us/item/32568...2usa4itemAdapt
and this:
https://www.ebmia.pl/2800-obrotow/39...-120-19-g.html

They are also not very expensive.

What do you think?

Three-phase power is not a problem here. I have three phases with 16A per phase overcurrent protection. It is intended for a stove/cooktop.

[joeavaerage]

Hi,
yes that spindle looks to be the right thing for the job. I would suggest a PMSM (permanent magnet synchronous motor). They are very much like an AC servo with similarly high power density
but those optimised for spindle use have fairly low rated speed, typically 1500 to 2000rpm, with comensurately high torque but have a high max speed, say 8000rpm. Very versatile and may obviate the need
to change pulleys altogether.

https://www.alibaba.com/product-deta...381b6b04d3Lvvn

Craig

[peteeng]

Hi Pio - You could go the gantry design and cast the base yet use steel for the gantry, that's how Datron do it, except they use granite. Check out your floor loads. A sustained high load on a floor will make it sag. Floors have average loads (called live loads eg for when the room fills up with people and then they go away allowing the floor to recover and dead loads that stay there long term...) and local load specs... I once worked for a company and our office was on a first floor. We made towbars for Toyota and had to store test weld pieces and sample towbars for archival purposes. The floor was divided into offices and storage. When I was employed as Dept Manager I looked at the floor and said its sagged? No one seemed to worry about it! The floor was over a structures laboratory with people wondering about it most times. I organised to move the storage elsewhere, at some point it would have collapsed... Cutting steel is a big spindle discussion but you have that under control...Peter

Since you have decided on the basics for the spindle I suggest you design the spindle area first and work outward. It will surprise you how much real estate you will need. My guess is that if you start at the machine base you won't have enough real estate left when you get to the spindle area... Peter

[joeavaerage]

Hi,

They are also not very expensive.

When it comes to spindles suitable for steel, then you require high torque at low rpm. There are two basic choices. The cheapest and most cost effective is as you have already
discovered, a belt driven spindle and an electric motor. As I posted earlier there are much better motors (PMSM) than just a plain induction motor and are yet still very cost competitive.

The second approach is a high power high speed spindle. These still achieve good low down torque, but only by virtue of their great power, say 9kW or 11kW. That gives them the torque
to drive an endmill directly in steel. Problem is they cost so much.

If you like my new spindle is a very modest example of that style of thinking. It has 3.3Nm of torque, and that is hardly anything to write home about, but is still useful in hard and tough materials
with tools up to about 12mm, and yet the same spindle still has good rpm for aluminum, brass, plastics and engraving. I paid $2000USD for it, including ship[ping to New Zealand and that included ten
HSK32/ER20 tool holders......very good value for money.

I think your solution is better for steel, you can bolt on whatever motor you require to get the torque you need say for a face mill or whatever. Where you'll miss out a bit is aluminum and other high speed ops
like engraving.....but you can't have everything, at least on a hobby budget.

There are a great deal of Chinese made ATC spindles on the market at the moment. I think the real bad slowdown of the Chinese economy has prompted manufacturers to advertise and cut prices. I rather suspect that the US and Europe
are going to respond with high tariffs to limit the Chinese dumping. There is an opportunity here....how long it lasts is at best a guess, but a few months to a year is my guess.
I would suggest you look very closely at some of them in the 5.5kW to 9kW class. You are looking for something with a lowish (8000-9000-12000) rated speed, but maybe a top speed of 15000-18000-21000.
If you can find a spindle of 4.5Nm to maybe 6Nm then that would be sufficient for steel.

My servo driven spindle is rated at 6.1Nm, and I can't really extract all that because the machine starts to flex too much.....so there is such a thing as 'too much of a good thing'.

This sort of thing:

https://www.aliexpress.com/item/1005...Cquery_from%3A

That works out just over $2000USD, has 6Nm and 12000rpm rated. That would be very useful. I think with good water cooling you could run that down to 2000rpm or 2500rpm, which is about right for 10mm and 12mm
tools in tough steel. Has a BT40 interface which is appropriate for that extra torque. A spindle that sort of ticks all the boxes without having to build anything.


The point being that with the market loaded as it is you might find a useful direct drive spindle for say $2000USD and that would be cost comparable to a belt driven spindle/motor combination.

Craig

[peteeng]

Hi Pio - Another worthwhile configuration is a portal machine like attached. Peter

[peteeng]

Hi All - I have been wondering why the Datron NEO is a portal type machine when all their others are gantries. I think its because with a gantry there is always a dead space at the back due to the gantry and Z axis fwd stack. With a portal machine you can slide the table thru the portal and use that space on small machines anyway. So I stripped down the high rail mill I have built and rebuilt it as a portal machine ie the table and z axis have the same relationship. I snugged the dims down a bit then ran some FE to see what it does. Current stats are X 101N/um Y 83 and Z 220N/um. Its steel and weighs 1361kg. I think I can do much better with some time on it, but won't at the moment. The footprint is 700mm deep by 995mm wide.. So it will fit thru a door if you had to.. I think this has possibilities for a small foot print machine. Since the Datron Neo is 700kg there seems to be scope to reduce weight further. Depends on what stiffness target you set... If you built this in aluminium it would weigh 470kg but obviously would not be as stiff.. Maybe there's some middle ground as Datron has found......Peter

edit - I have toyed around with exoskeleton designs off and on. They get complicated as they are based on triangles or octagons. One config worth looking at is attached. Since you have to make an enclosure that space may as well be taken advantage of as structure. I haven't modelled this yet but I expect it to be quite stiff. The gantry can be huge as it does not move etc... Just some more thought bubbles. My mill is for machining composites which are very abrasive and make sludge that's why I need my mechanics up high. With steel chips maybe the portal is a good config...

[peteeng]

Hi Pio- Not sure if your following the Milli thread. I pursued the exo-portal design and got the weight down to 730kg with all axes 100N/um. It could come down more with some more time. The portal design has merit and on very large machines (and I have seen it on small routers) they use a W axis which is a rising gantry design. This means the Z axis height is always small so its very stiff. Here's a large mill with that feature. There are a couple of lifting gantry machines in the forum. I've run out of ideas now so good luck with the project. Peter

https://youtu.be/f8nTNY0FO6o