steppers for cnc fusion x2 conversion

[Numb Nut]

I'm starting to see the appeal in a CNC milling machine...

Hoping to draw on other people's experiences on what might be a well trodden path - what stepper motors are a good choice to go with the full ballscrew cnc fusion kit #4 conversion of the X2 mini mill?

http://www.cncfusion.com/minimill1.html
http://www.cncfusion.com/images/X2mo...0pics/kit4.jpg

The cnc fusion ballscrews are 5 tpi aren't they?

If I went ahead with the plan, I'd want to install a counterweight for the z-axis, removing the spring loaded arm from the left side of the X2.

I think the Keling KL23H2100-35-4B steppers rated at 381 oz-in might be a good choice for the X & Y axis. Question is, would the same also suffice for the Z axis, bearing in mind that the cnc fusion kit is direct drive (1:1) on all the axis, including the Z axis ballscrew?

http://www.kelinginc.net/NEMA23Motor.html
http://www.kelinginc.net/KL23H2100-35-4B.pdf

Also curious to hear what speed of rapids I could expect from this (on all axis)? I mean fully reliable rapids off course, no missed steps!

If so then should be able to make use of a Gecko 4-axis G540 driver and keep things simple.

http://www.geckodrive.com/product.aspx?c=3&i=14469

I've been reading alot of threads about X2 conversions and there seem to be many variations for driving the various axis - leadscrews, ballscrews, direct drive, pulleys, etc. With the linear motion mechanics nailed down I'd just like some advice on the remaining variables...

Thanks for any advice!

[hoss2006]

With the G540, the keling 381 oz/in motors (KL23H2100-35-4B) would be best for the X and Y.
It used to be listed as a 318 oz/in, same part #, but they've upgrade it to 381 now?
I have the old pdf which lists it as 318.
They match the Amps (3.5) and Voltage (48) of the G540 almost perfectly so you will get the best results with them.
Would have recommended the 387 for the Z but if the 381's are really 381 oz/in they would be better.
The 387 has the same 3.5 amps but needs a higher voltage (64) to get peak performance.
You should expect 100 IPM for the X and Y and 60 for the Z.
I can nearly achieve that with only 40V and 3 Amps motors, and I have a little more mass moving around than most too. ( alright no fat jokes )
Hoss

[Numb Nut]

[Have corrected a typo in the above post - got the wrong motor part number and torque rating, at least the URL was correct!].

Thanks alot Hoss, that's actually what I was hoping to hear!

Rapids at 1 inch per sec sounds good to me. With the G540's 10 microsteps and the CNC fusion 5 tpi ballscrews (correct?) that gives a theoretical resolution of 0.0001" and requires a pulse rate of 10 KHz for 60 ipm which is well within the Mach3 spec.

I'll drop Keling a quick email and see if they can confirm the change in specification for that motor. If confirmed I think many folks may find this combination of kits to be a very convenient and minimal DIY build for their first CNC mini-mill.

Would like to hear from anyone using CNC fusion kit #4 on an X2, their choice of motors and their evaluation of the result.

[Dougal]

I'm using 280 oz-in steppers with my CNC Fusion kit, they're good for around 250kg of linear force.

I'm using G250 driver cards (similar thing found in the G540), my X seems to rapid around 2600mm/min (just over 100 in/min), my Z not far off (still using the torsion spring on the Z axis) but my Y has issues and only runs at 1/3 the speed of the X axis.

[Numb Nut]

Thanks for the input Dougal, sounds like you're getting good performance with those motors.

On your X2 build thread I note that sansbury mentions that Xylotex 270 oz-in motors are popular with the CNC fusion kit.

On another thread I saw that Jay C is using some Keling motors rated at 200/282 oz-in motors - depending on how they are wired up - with his cnc fusion ballscrew kit.

Sounds like I will be ordering the 3 KL23H2100-35-4B steppers, G540 driver panel and PSU soon...

[Dougal]

Thanks for the input Dougal, sounds like you're getting good performance with those motors.

On your X2 build thread I note that sansbury mentions that Xylotex 270 oz-in motors are popular with the CNC fusion kit.

On another thread I saw that Jay C is using some Keling motors rated at 200/282 oz-in motors - depending on how they are wired up - with his cnc fusion ballscrew kit.

Sounds like I will be ordering the 3 KL23H2100-35-4B steppers, G540 driver panel and PSU soon...

I think the motors I have are identical to the keling 270's. Maybe a bit of give or take on the torque rating depending on who sells them.
I'm still working through the finer setup points on my machine so haven't cut enough to know if it's a success or not yet. The true test for me will be cutting 304 stainless. I've tried once already and without a stiffened column it was chatter city.

[Numb Nut]

Had a reply from Keling, the KL23H2100-35-4B steppers were 381 oz-in all along, the initial documentation was in error.

That's good news from my point of view, confirms the plan to use these with a G540.

Any idea how many amps the 48V power supply should be rated for?

I should probably make an allowance for a future 4th axis.

I see in the following G540 heatsink experiment using 4 steppers (KL23H284-35-4B) that the PSU was rated at 9.3A. http://www.geckodrive.com/upload/G54...Experiment.pdf

Dougal - look forward to hearing about your column support mod and how well it works for cutting stainless.

[Numb Nut]

Dougal - forgot to mention that I checked your 250 Kg figure for linear force and got the same result! The holding torque for your steppers through 5 tpi screws translates into a linear holding force of 250 Kg - that's ignoring the extra resistance that would be provided by less than 100% screw efficiency.

That doesn't mean to say that the dyanmic linear force generated on the axis will be 250 Kg (screw efficiency will incur loss and microstep torque is bound to be less than the holding torque, right?) but does hint at why those motors seem to work well enough.

[Numb Nut]

On the subject of PSU specification - have just noticed that the G540 manual says:

The power supply current will depend on the number of motors used and the motors’ rated phase currents. A rule of thumb estimate is add up all the motor phase currents and multiply the result by 0.6 to get the required power supply current.

So for a 3 axis setup that's would be 6.3 A supply, perhaps the KL-4875 rated at 48V/7.5A. Another axis would bump that up to 8.4 A, so probably the KL-4813 PSU rated for 48V/13A. Any comments on that choice? Could probably get away with the 7.5 if I knew the 4 axis weren't going to spin at the same time.

[escott76]

On the subject of PSU specification - have just noticed that the G540 manual says:

The power supply current will depend on the number of motors used and the motors’ rated phase currents. A rule of thumb estimate is add up all the motor phase currents and multiply the result by 0.6 to get the required power supply current.

So for a 3 axis setup that's would be 6.3 A supply, perhaps the KL-4875 rated at 48V/7.5A. Another axis would bump that up to 8.4 A, so probably the KL-4813 PSU rated for 48V/13A. Any comments on that choice? Could probably get away with the 7.5 if I knew the 4 axis weren't going to spin at the same time.

If you can afford the CAM to make 4 axis simultaneous moves, a couple extra bucks on a slightly larger power supply should make little difference.
I have the larger one and it works great, driving 3 387's at 3.5A. Soon I'll add a 4th axis and know it will be fine.

[Dougal]

Dougal - forgot to mention that I checked your 250 Kg figure for linear force and got the same result! The holding torque for your steppers through 5 tpi screws translates into a linear holding force of 250 Kg - that's ignoring the extra resistance that would be provided by less than 100% screw efficiency.

That doesn't mean to say that the dyanmic linear force generated on the axis will be 250 Kg (screw efficiency will incur loss and microstep torque is bound to be less than the holding torque, right?) but does hint at why those motors seem to work well enough.

Well that's good news.

Yes, 250kg is the theoretical frictionless force, it'd be interesting to stick a load-cell on the table and see what it can actually do including the friction from everything from the ballscrew to the Z axis slides.
But what I'd really like to know, is what sort of force I put on a typical drill-press when drilling steel. See how that compares.

Shame I pulled my mill-drill apart to make a CNC mill.

[Numb Nut]

Dougal - measuring the actual force at the table sounds like a neat experiment - let me know how you get on.

Escott76 - went with the 13A PSU too. It comes up at $129 when "add to cart" is clicked, not $149 as listed at http://www.kelinginc.net/SwitchingPowerSupply.html The payment system doesn't seem to generate an international shipping charge so have emailed them.