[Redline]

Who here has used the CNC Fusion ball screw conversion kits for the Seig X2 mini mill. What do you think of them? Is it worth the money?

[MacCNC]

Hey Redline,

I have the X3 kit instead of the X2 so I can't comment on it but overall I am pleased with the X3 kit from cncfusion and especially the service from Michael. He was very helpful in the assembly and I had an odd machined X axis bearing block and I sent my original to him and he machined an exact match to fit no charge.

[Ron111]

MacCNC,
I'm following a thread that JL123 has started. What were your backlash numbers?

http://www.cnczone.com/forums/showthread.php?t=22211

If you could post some pic's and tell some of your details, it may help JL123 who is currently meeting some challenges.

Ron :cheers:

[MacCNC]

Redline,
Please ignore my "endorsement" of the cncfusion kit, you need a response from an actual X2 owner instead.

Ron111,
Thanks for the thread link. I'm just a newbie and feel kinda stupid after seeing some of the responses to the design of the cncfusion kit.

I too was a little supprised to see how the shoulder was turned on the end of the screws but I had a washer from the original parts that fit up against the threads to go between the thrust bearings and thought it would work just fine. That may not be true after it wears down when milling metal but I have just been working with wood so far trying to learn Mach3 and stuff.

The split ring locking collar is a pain to get on tight but by tapping it with a small mallet while tighting the screw and compressing by hand I can get all of the backlash out. I think it can even be too tight because the screw cannot be turned freely by hand, it took me a couple of tries to get it tight but not too tight.

I measured my backlash in the whole assembly last night using a dial indicator and jogging by 0.001 each axis. The backlash in the X axis is 0.008", which is what I understand the tolerance of the nook ball screws to be. The Y axis is a different story all together. It has 0.022" backlash because I shelled out the bearing balls while working on limit switches and had to replace them with a new set of balls that don't fit. Now if I could only get backlash compensation to work in Mach3...

After reading the link you sent, I will probably be ordering new thrust bearings and I know I will be ordering new balls to reload that Y ball nut. Again, thanks for the link.

Everything is assembled and there is not anything to get pictures of now, but I will if there is anything you really want to see.

[Ron111]

Maccnc,

What kind of problems are you having with the backlash with mach3. I have usually gotten excellent support from Art of Mach3. Tell me what not working right. And go ahead and go to the mach3 website and check with art and I sure we can get you a good answer. His link is http://www.artofcnc.com and just find the contact us link and let him know what is going on. He has always got back to me in a day or two.

Did you catch the technique that JL123 used to preload the bearings, using the pipe clamp. This might be helpful after you replace the bearings in the ball nut.

I think you will have a excellent machine once you get the bugs out. If you need help, shout out and there are plenty of folks to give suggestions.

Hope this helps,

Ron

[BobWarfield]

0.008" seems like a lot of backlash. I thought you had to get it down a lot closer to 0.001" before results started to be satisfactory on a mill. I'm getting 0.010" on my big old hoary ACME screws for the X-Axis on my IH Mill, and about 0.004" on the quill screw. Seems like a ballscrew could perform better.

Best,

BW

[S_J_H]

.008 backlash is quite a bit.
.022" would be very hard to work around I would think.
I see a lot of guys doing cheap ball screws and then tolerating the backlash thinking they made an improvement. I disagree.
The only advantage you gained from a ballscrew with that much backlash was less power to drive the screw.
A good acme with a good anti backlash nut will easily give you ~.001" backlash with little effort and just as good of accuracy as the rolled ball screws.
I have a mini mill cnc'd for example. Rolled Ballscrew on the z and acmes' on x&y with just a split nut.
Ball screw backlash is at .004" and the acmes at .0015- .003.

Yeah I can move the z axis around at up to 160+ IPM and the acme x & y at only 25ipm. That's just to showoff.
But actual cutting feed rate is less than 25 IPM so the accurate acmes out perform the rolled ball screw with single nut. Only the rapids suffer with the acme screws.
I'm doing my x3 right now. Just reusing the very good acme's with modified anti backlash nuts at less than .001" , mainly because I need to use it manually all the time as well as cnc. Yep it will be slow compared to the ball screws, but it will be very accurate, cheap to build and also a better manual mill than when new.
You might have to get those backlash numbers down. Even with .003" backlash in your screws and using the backlash comp in mach3 will result in circles not all that great. At least that is what I have seen.


Steve

[MacCNC]

Ron111,
When I turn on backlash compensation and restart Mach3 the X axis starts running slowly by itself until I hit Estop. Do you know what causes this?

I am pretty unhappy with the backlash in the ball screws but I am ordering the replacement balls for all the nuts today and hopefully I can get it down.

[Ron111]

MacCNC,

That's a new one on me. Email Art and get his input. I would try reloading Mach3, but move your configuration file to another directory just incase it gets wiped out and you can copy it back. (Just to be safe)

It almost sounds like some how the system, thinks it is seeing a jog input (an arrow key pushed down) for the x axis.

Just my guess. Your backlash may take a preload set up to eliminate. In one on the last post that I made on JL123 thread, I included a thread concerning preloaded double nut setup. You might want to look it over.

But, like I mentioned eariler, I feel that you will end up with an excellent cnc machine, but sometimes it just takes a little extra effort.

Do let us know what you find out.

Ron

[MacCNC]

FYI, I have reloaded my Y axis ball nut which I spilled. I use the 0.125 balls from McMaster Carr, part 9528K11. The backlash is now 0.008 just like the X axis which still has a factory load on it.

Does anyone know where I can get _slightly_ oversized bearing balls to reduce the backlash more?

Thanks

[Greolt]

Does anyone know where I can get _slightly_ oversized bearing balls to reduce the backlash more?

Have a look at this thread
http://www.cnczone.com/forums/showthread.php?t=23191

[jl123]

Actually I did a similar thing yesterday max. I was jogging my Y to the point that the ballscrew came out of the nut. I said the hell with it and cleaned up a little.

The cause of it? My laptop has that menu button (to the right of the windows key) very close to the arrow keys. While I was pressing and holding the key my big finger also pressed the menu button. Since I wasn't looking at mach I didn't catch it in time and that was that.

Now I have a reason to machine me out some mounts and get my limit switches working. I spent about an hour today putting the bearings back in the ballnuts and got everything back together. I noticed that there's a gap between the bearings if you look for it. Looks like another bearing could fit in there but its not big enough. I counted them when putting them back in. Maybe this is the cause of the backlash.

[Ron111]

JL123,
Did you reload with the .125 balls and did it reduce your backlash.

Have a look at this thread
http://www.cnczone.com/forums/showthread.php?t=23191
Greolt added the above thread which you may or may not have seen, but is a good source for larger oversized balls.

Ron

[Ron111]

Maccnc,
Did you determine what was causing the "Run Away" with the backlash compensation turned on?

Ron

[jl123]

Turns out I goofed when measuring the stock bearings.

Stock bearings .1257
Mcmaster .1249

Looks like the nook ones are a little oversized to begin with.

[MacCNC]

Ron111,

No I have not. I have decided to see how low I can get the backlash by reload with oversized balls and hopefully a better thrust bearing solution first. I have at least got the Y backlash from 0.022 to 0.008 and I have oversized balls on the way.

[jl123]

My backlash using the same kit is .009". I originally stated .011" but after preloading the thrust bearings with a pipe clamp it got better.

[MacCNC]

jl123, that's the same thing that happened to me, I jogged right off the end while working on limit switches. :frown:

Here is a video of some rapids on my X3 and towards the end of the video you can see the X axis limit switch. It's an Automation Direct prox switch and only costs about $20. I used a peice of angle to shield the switch and for it to detect the edges. It works in Mach3 for both homing and travel limits. I have a similar solution on the Y axis.

I gave up on trying to get anything for the Z axis since the downward limit always changes. Anyone have any suggestions for the X3?

Mac

[jl123]

Nice rapids and switches. I wish my machine was that clean...lol

My rapids are pretty insane. I can easily push 300ipm on the z without a problem. Of course that's just playin around. I had a stepper off and was pushing 450ipm at 10000 steps per inch. Sounded dangerous though.

[MacCNC]

Wow, that's screaming fast! I can do about 270 ipm but usually keep things tuned to about 100.

How do you like the G-Rex?