[binfordw]

So, 15+ years ago I bought a G0619, ballscrews and the electronics to cnc it. It was fun, its seen its uses, but its never been great. Mainly the Z column. I've worked on it numerous times over the years trying to improve it, but it is what it is.

I'm at a point where again, I want a good, useable cnc mill. I don't need speed or deep cuts, but, acceptable quality, and the spindle speed/ability to run smaller endmills (.0625")


Should I buy a new manual mill platform and cnc it? What models are the best bets to surpass the mediocre G0619 results I saw?

Should I further consider fixing the G0619? New XY ballscrews, Hiwin rails/trucks to replace the dovetail on the Z column, and a better spindle?

Any thoughts are appreciated!

[cncmakers001]

So, 15+ years ago I bought a G0619, ballscrews and the electronics to cnc it. It was fun, its seen its uses, but its never been great. Mainly the Z column. I've worked on it numerous times over the years trying to improve it, but it is what it is.

I'm at a point where again, I want a good, useable cnc mill. I don't need speed or deep cuts, but, acceptable quality, and the spindle speed/ability to run smaller endmills (.0625")


Should I buy a new manual mill platform and cnc it? What models are the best bets to surpass the mediocre G0619 results I saw?

Should I further consider fixing the G0619? New XY ballscrews, Hiwin rails/trucks to replace the dovetail on the Z column, and a better spindle?

Any thoughts are appreciated!

Better buy a new machine than fixing the G0619.

[joeavaerage]

Hi,
the problem is always rigidity or rather the lack thereof.

I would be extremely reluctant to sink more money/time/energy into a project that will at the end still not be rigid enough. Any donor machine that is rigid enough is worthwhile.

I make a lot of small parts and PCBs for automotive instruments and use small tools like 1.5mm four flute endmills, and down to 0.5mm two flute endmills, 0.2mm drills etc.
You want a 24000rpm spindle, but it needs to be low runout otherwise you'l break tools left right and centre. I bought an 800W air cooled 24000 rpm spindle from Mechatron GmBH.
It seemed expensive by comparison to the Chinese equivalent, but then I've used it daily for ten years. Whatever money I spent on it its paid me back dozens and dozens
of times over.

Craig

https://www.mechatron-gmbh.de/en/?gc...caAq2SEALw_wcB

[joeavaerage]

Hi,
whoops, just noticed I attached the wrong spec sheet, this one is for a water cooled model whereas mine is air cooled. Same core motor however.
I'd really like an ATC model next time...but they are expensive. The one I want (ATC-8022-42-HSK25) is 2.2kW, 42000rpm with an HSK25 tool interface, was,
the last time I priced it, 5700Euro....so I'll be wanting it a bit longer!!!

In the mean time I have a 3kW servo motor that I want to drive a belt driven ATC spindle, probably BT30 at up to 6000rpm. This is for high torque, low speed
especially for steel and stainless. Much of this I'll have to make myself.

Craig

[binfordw]

Better buy a new machine than fixing the G0619.

What new machine would you suggest?

I've looked around a bit and it seems like the Precision Matthews mills are a popular mill to convert now. But, they all have low spindle speeds.

[binfordw]

Hi,
whoops, just noticed I attached the wrong spec sheet, this one is for a water cooled model whereas mine is air cooled. Same core motor however.
I'd really like an ATC model next time...but they are expensive. The one I want (ATC-8022-42-HSK25) is 2.2kW, 42000rpm with an HSK25 tool interface, was,
the last time I priced it, 5700Euro....so I'll be wanting it a bit longer!!!

In the mean time I have a 3kW servo motor that I want to drive a belt driven ATC spindle, probably BT30 at up to 6000rpm. This is for high torque, low speed
especially for steel and stainless. Much of this I'll have to make myself.

Craig

I don't know much about spindles really, but those seem like router spindles to me?

I would just want a quality set of bearings that would run 6-8k rpm, preferably R8 as thats what all my tooling is.

[ardenum2]

I don't know much about spindles really, but those seem like router spindles to me?

I would just want a quality set of bearings that would run 6-8k rpm, preferably R8 as thats what all my tooling is.

its much easier to achieve required machine stiffness for a high speed spindle and low OD endmills, take a look at datron neo, it kinda has a router spindle but it achieves magical tolerances, you can take a facemilled plate and it will float on a granite plate right off the machine.

0. what OD endmills do you have?
1. budget?
2. willing to build it yourself?
3. required to machine steel?

[binfordw]

its much easier to achieve required machine stiffness for a high speed spindle and low OD endmills

0. what OD endmills do you have?
1. budget?
2. willing to build it yourself?
3. required to machine steel?

EDIT*

0. Most of my endmills are .0625 to .375". .375 would be the largest I'd ever expect to run.

3. Yes, I would want to machine CRS and 6061 mainly.



I'd just want the best bang for buck- balanced a bit with the amount of time invested. I'd say 4-5k would be a rough estimate. If 7K would end me up with a far superior machine for my uses, I'd be interested.


I'm well set up and skilled to build something, I just have little direction for a plan. I do CAD design/ assembly management for an OEM equipment manufacturer, and am well versed with electronics, steppers, etc. If I had a solid, bare machine with XY and Z travel, I could source, mount and wire up the electronics to get the movement easy enough.


I have considered Hiwin rails/ a new headstock mount to correct the Z on my existing G0619, but, I'd still be left with a poor spindle and XY screws. I could replace all this stuff, but not knowing how good of an outcome I'd be left with is a deterrent.


If I would end up with a machine that could mill CRS with a .250 4 flute carbide endmill, at 15-20 ipm @ .100 depth, without drama, I'd be pretty happy.

[ardenum2]

If I would end up with a machine that could mill CRS with a .250 4 flute carbide endmill, at 15-20 ipm @ .100 depth, without drama, I'd be pretty happy.

FSWizard(https://app.fswizard.com/) says that's a 300N cutting load, definitely achievable, stiffness wise.

one good way would be to weld the machine and then stress relief it at a local company. Whether your budget allows for it is up to you to find out.

another way would be to use a ready-mix uhpc or mineral cast, one is a special concrete, the other is an epoxy based mix used for machine tools. an alternative to this would be a class c non shrink grout that's double compensated. this one is more of an experimental approach we're trying out in this forum, benefit is it's cheap but the results can vary from great to disastrous.(it hasn't been tested yet)

a lot of dental companies build their machines on and from a granite surface plate. that's also a possibility. this requires the use of a high speed spindle and a different machining approach. there are spindles that directly grab onto the endmill and they are available for 6.35mm endmills. the price for such a spindle would be high even from china, starting at about $2000.

material prices start at $3/kg for uhpc, $5/kg for mineral cast, $1.5/kg for grout.

for that budget you'd need to form all precision surfaces with a special epoxy and a straight edge.

If I were you, I'd start by digitizing my machine in cad and running fem on it, to be sure where you stand in terms of stiffness, then design something at least 2x or 3x stiffer. think it's a good place to start. you only need to translate the column and the spindle headstock with the spindle. grabcad might even have a model available for that purpose.

[binfordw]

FSWizard(https://app.fswizard.com/) says that's a 300N load, definitely achievable, stiffness wise.

one good way would be to weld the machine and then stress relief it at a local company. Whether your budget allows for it is up to you to find out. since you already have a machine, machining the surfaces could be done by you.

another way would be to use a ready-mix uhpc or mineral cast, one is a special concrete, the other is an epoxy based mix used for machine tools. an alternative to this would be a class c non shrink grout that's double compensated. this one is more of an experimental approach we're trying out in this forum, benefit is it's cheap but the results can vary from great to disastrous.

a lot of dental companies build their machines on and from a granite surface plate. that's also a possibility. this requires the use of a high speed spindle and a different machining approach. there are spindles that directly grab onto the endmill and they are available for 6.35mm endmills. the price for such a spindle would be high even from china, starting at about $2000.

material prices start at $3/kg for uhpc, $5/kg for mineral cast, $1.5/kg for grout.

If I were you, I'd start by digitizing my machine in cad and running fem on it, to be sure where you stand in terms of stiffness, then design something at least 2x or 3x stiffer. think it's a good place to start. you only need to translate the column and the spindle headstock with the spindle. grabcad might even have a model available for that purpose.

Appreciate the info-

I had added a 1/2 thick support structure to the column years ago, but my main issue is in the Z dovetail. I've removed and hand worked the gib, blued, and dial indicated play to get as much out of it as possible, but its still too much/not ridgid enough.


As far as just flat out making a machine- It would need to be cast or a material as you mentioned to alleviate vibrations? So even if the machine was made from 3,000'lbs of CRS, it would have bad harmonics that would probably lead to damaged tools or even spindle bearings?

[handlewanker]

Hi.....my two pennoth worth from experience.......if it's possible you would get the most bang for the buck if you removed the Z axis dovetails, ,as in machine them off completely and fit linear rails.......the Z axis dovetails on these mills were never designed to mill on the move and mostly they rely on the quill to do a down feed.

[binfordw]

Hi.....my two pennoth worth from experience.......if it's possible you would get the most bang for the buck if you removed the Z axis dovetails, ,as in machine them off completely and fit linear rails.......the Z axis dovetails on these mills were never designed to mill on the move and mostly they rely on the quill to do a down feed.

Appreciate the input,

I am leaning towards this option, atleast to try. I know I can modify the column for linear rails ( I have 4 huge HG35 blocks here at work I could get for next to nothing- they were mis-ordered/non returnable)

I could source some good/or atleast decent ballscrews to replace the surely worn, probably not that great screws to begin with as well.



What would be my options for replacing the spindle/headstock? Anyone have a link to a US website that sell an R8 spindle unit capable of 6k+ rpm? I could rather easily add a AC motor/VFD and a belt drive for the new spindle, I would probably have a pretty capable machine at that point I would think.

[joeavaerage]

Hi,
if you want to run 1/16th tools then 8k rpm is going to be interminably slow. Ideally runout should be <5um and anything over 10um will cause small tools to break.
Are R8 spindles capable of that?

Craig

[binfordw]

Hi,
if you want to run 1/16th tools then 8k rpm is going to be interminably slow. Ideally runout should be <5um and anything over 10um will cause small tools to break.
Are R8 spindles capable of that?

Craig

Good question, not sure. I assumed a quality R8 spindle was/could be just as low runout as any size toolholder.


I definitely need to be able to use toolholders and have the ability to change tools rather easily, I'm not after a cnc router.

[joeavaerage]

Hi,

How are you measuring a sub 10 micron runout?

Usual way, with a micron reading dial gauge, not that I bother measuring it.....you soon know if its off. Small diameter carbide tools are very fragile. Any dust
or swarf inside the collet housing and suddenly you start pinging tools. I use Rego-Fix collets, I find it just plain required when you are using tools in the sub 0.5mm range.
For tools in the 3mm plus range then good but otherwise cheap Chinese collets are OK, but they are rubbish below 0.5mm.

Craig

[binfordw]

I'm not finding any real solutions for a spindle. Maybe I should aim at rebuilding the spindle I have with good bearings and see where that gets me.

[joeavaerage]

Hi,
versatile spindles of good quality are indeed a bottleneck for hobbyists. If you are prepared to pays tens of thousands for an industrial spindle then OK, but otherwise there is
little choice.

As I posted earlier I use an 800W 24000rpm spindle, a router spindle by your description, and I use it A LOT, maybe 90% of the time. When I need to spin bigger tools
and especially in steel then I use my second spindle, an ER25 Rego-Fix tool holder running in matched angular contact bearings direct coupled to a 1.8kW, 6.1Nm (cont), 18Nm (overload),
3500rpm Allen Bradley servo, all pretty much handmade.

You may be advised to go the same way, ie have one spindle, say your rebuilt R8, for larger tools and a smaller and cheaper asynchronous (18000-24000rpm) spindle for small diameter tools.

Either way rebuilding your R8 is likely to be useful....but be prepared to weep when you go to buy P4 matched angular contact bearings, they can be hundreds of dollars a pair!!!

Craig

[binfordw]

Hi,
versatile spindles of good quality are indeed a bottleneck for hobbyists. If you are prepared to pays tens of thousands for an industrial spindle then OK, but otherwise there is
little choice.

As I posted earlier I use an 800W 24000rpm spindle, a router spindle by your description, and I use it A LOT, maybe 90% of the time. When I need to spin bigger tools
and especially in steel then I use my second spindle, an ER25 Rego-Fix tool holder running in matched angular contact bearings direct coupled to a 1.8kW, 6.1Nm (cont), 18Nm (overload),
3500rpm Allen Bradley servo, all pretty much handmade.

You may be advised to go the same way, ie have one spindle, say your rebuilt R8, for larger tools and a smaller and cheaper asynchronous (18000-24000rpm) spindle for small diameter tools.

Either way rebuilding your R8 is likely to be useful....but be prepared to weep when you go to buy P4 matched angular contact bearings, they can be hundreds of dollars a pair!!!

Craig

Interesting. Yea, most of my use would require tool changes, center drilling, drilling, and profiling with a roughly a .250" endmill. I just couldn't see a high speed "router spindle" as I call it, to work for that well.

Maybe best would be accommodating a spindle holder on the side of the R8 spindle headstock- for high speed, small tool use like you mentioned.


I'd be fine with a few hundred bucks for bearings- IF, I knew it would get me reasonable results.

[joeavaerage]

Hi,
a four pole spindle (12,000rpm@400Hz) 2.2kW would be fine in Aluminum with a 6mm tool.
Where the highspeed spindles are not good is steel. You would be ill advised to run a 12000rpm (rated) spindle at less than half speed, and that
would limit you to about 5.3mm diameter (100m/min surface speed) in steel and about 2.6mm diameter in stainless (50m/min surface speed).

Your description of such spindles is not inaccurate, but neither does it do justice to what they are capable of. The limitation is high torque/low speed for which they are poor.
Big industrial mills have the same setup, just much bigger, and you wouldn't call them router spindles....but that exactly what they are, just huge, and they can accommodate high torques at slow speeds
just by virtue of their size. Prohibitively expensive and heavy for you or I though.

Craig

[binfordw]

So-

I started tearing into the old mill, removed the Z ballscrew. I was wanting to find replacement screws, but I noticed the fixed end of the Z screw had a loose nut.

It has a pair of thrust bearings, and a single spanner nut with nothing to lock it.. Looks like I'm going to have to figure out the best way to replace/correct the fixed end bearings before going much further.