[dneisler]

I have a X3 with the Deluxe Kit. 425 Oz and a 640 oz on the Z. Running Geck0 203's at 54v. I am getting 120 IPM Rapid on all axis. I know I will not cut that fast.

My problem is on my X axis I have .030 backlash. It appears to be coming from the coupler between the stepper and ballscrew. As if the rubber insert in the couple is undersized. Any one have this problem or suggest a fix?

On my Y axis I have about .004 backlash. I was hoping for better, I am think it could be from the coupler too, but have not checked it.

I cut a 1" square and it came out .972 and .976 so sounds like we are off a little.

Advice?

[tauntdesigns]

I haven't tested mine yet.

Are your steps in the control software set to 10160?

Try cutting two opp. ends (x) of the block making sure to approach x dim from the same direction, moving the x-axis in same direction for your xdim on both sides of the block will let you know if your steps are set correctly.

Checking backlash: one way is to use edgefinder. touch off your square block and set x zero, now go to other end of block, move until you touch, take dro reading minus the diameter of the edgefinder should give you the length of the block, the deviation would be backlash + or - .001 (not best way but close)

[dneisler]

10160? Where did you get that?
200x10x5 = 10000

OR CNCfusion told me wrong on the screw being 5TPI

I can put my had on the lovejoy under the X axis, I can feel the gap between the inner rubber and the outer steel coupler changing, so I am think there is some play there.

[tauntdesigns]

The ball screw are metric 5mm per rev. 1 inch equals 25.4mm
10000 steps = 5 revolutions = 25mm
the extra 160 steps is the .4mm for the rest of the inch

I guess I should ask are your units set to inches?

[dneisler]

Well that makes sense, I wish he would have made it more clear, he told me 5TPI. I will try that tomorrow and see if that helps.

Thanks

Very please at the performance so far.

[project5k]

see my understanding is that my cnc fusion ballscrews are 5tpi as well.. hmmm makes me wonder... the gap in the lovejoy is definately an issue, but if you cant find a mechanical fix, you could allways dial that lash into the compensation in mach 3... i would think that if you were to go to mcmaster-carr you could find some new spiders for a buck or 2...
mine are nice and tight...

[tauntdesigns]

Howdy guys,

I found a post over on the cncfusion forum under the thread name " Sneak Peak X3 Deluxe kit"

In a post a question was asked about the TPI. He (CNC Jedi) answered, The ballscrews are 5mm pitch.
He should have said, They're metric ballscrews at 5mm per rev.

Later on,
Jack

[dneisler]

Been doing more testing and playing.
..cutting a 1" square in wood right now for testing.

.972 x .972 is what I got.

Changed(lots of playing) steps per inch to 10220
and I am getting now

.998 x 1.00

So that is better.

I can see the coupling on my X from my bench pretty good the way it is made. I attached a DTI to the table, and jogged in .001 increments.

Jogged .010 and it showed like .00992(Close enough)
Jogged back to 0, and it showed +.004
So that is .004 of backlash correct?

I did the same thing on the Y with similiar results.
I had a sherline before this, never tested its backlash, I was hoping for less, but I may be expecting to much. Does .004 sound acceptable for the deluxe setup?

Here is a few pics, not up to date but...

http://www.donald-neisler.com/index....x3-cnc-update/

[rexstep]

Try wrapping teflon tape around the flexible insert- probably a single turn- and then wedge it back in to see if it fills the gap thus removing the lash

Rick L.

[Stepper Monkey]

.004 is pretty big, you should be able to cut that in half at least.
If it really is just from radial play in the couplers, you might try something other than the Lovejoy type. They can really suck sometimes. I really prefer the one-piece slitted stainless couplings. They may be pricey compared to the lovejoys, but not a big cost at all compared to the money you just spent on accurate ballscrews whose accuracy you can't use because of shoddy couplings! You could just try shimming the Lovejoys, but they will just wear again, and besides that your backlash can and will change under load as it compresses or deflects the spider - a problem you won't have with better couplings. Cheap Lovejoys are not a benefit to anyone!

[dneisler]

Can someone give me the direct links to some better couplings? I think I will try to shim the lovejoys right now and make sure that fixes the problem then deal with gettting some better ones.

[tauntdesigns]

Donald,

www.mcmaster.com

Type 'coupler' in the search products box at the top left of the site page.
scroll down the list to, motion control couplings, misalignment couplings, metric couplings, etc..

I took a look at your pics, looks good!!!

Your power supply, is it a KL-6515 from Kelinginc? That's (KL-6515) the one I'm using and was going to hook mine up to my C11G bob like you did but, when I got my bob I noticed it needed 5v@2amps and the power supply said it was a 5v@1amp regulated or 5v@2amps unregulated. I didn't know if I could use it so, I bought the bob power supply from cnc4pc.

happy machining,
Jack

[dneisler]

This is the Power Supply I am using. No issues at all with it so far.
http://cgi.ebay.com/54V-12A-5V-12V-R...QQcmdZViewItem

I looked at couplers and go so confused, way to many to figure out. Which eact one do you recommed?

[tauntdesigns]

I started looking at couplers again after reading about your results with the lovejoys.

I got my pulley this morning, so now I can finish installing my kit. Next I'll hook up my computer and control box and do some testing. I'll check my machine for backlash and report my results. I also have some oldham couplers (slotted disc) similar to the lovejoys. I may bore them out to fit the mill and test them too.

Bellow couplings:
I read good things about the "heavy duty bellows shaft couplers". They claim zero backlash and are for precision motion-control application requiring high maximum speed. I'd like a set of those, maybe on payday I can spring for some.

Jack

[project5k]

i dunno if its just me, or i got lucky, but my spiders needed some serious persuasion to go into the lovejoys.. once i got them in, thiers absolutely no play that i can feel.. ofcourse i see the potential for some wear over time.. but once the machine is up and running in its current state, i can probably machine some new ones out of some better material..

hmm wonder how delrin would hold up....

[dneisler]

I was thinking of doing that too with Delron, here is some pics of what I think my issue is. Not sure if they are good enough to tell. You can see where the gap is and goes away in these. Those are from my X axis. I had to drill a hole in the center of the spider, wander if that has anything to do with them. My stepper motor shaft was to long. Taking suggestions on what to replace it with. Be nice and help a fellow CNC'er out and click on a google ad on my site. :-)

[dneisler]

Ok, so I took the stepper off my X axis and inspected it all. I can put the spider in the coupler on the stepper and move it back and forth just a little bit, so I am pretty sure that is where my trouble is. I am trying to find some tape(ugh kids) to put on it long enough to test and see if it helps. I saw somewhere on here where Joe2000che made some from I think delron. Curious to know how it work and if he has a drawing so I don't have to reinvent the wheel.

[tauntdesigns]

My x-axis coupler seems to fit ok but, it was a very tight fit length wise (gap between motor shaft and screw).

The gap on my y-axis between motor shaft and screw was too short. The rubber spider would not fit between them. I cut a small hole (smaller then the motor shaft) in the rubber spider and forced it on the shaft. The spider being forced on the shaft made it bulge a little making a tight, tight fit.

I know, I know, all I needed to do was put a washer in between the bearing and ballscrew on the ballnut side of the bearing block. Didn't thank of that until I'd already cut the hole in the spider. Oh well, next time!

If you have gap problem on x-axis, a spacer could be made to go in between motor mount bracket and the table.

Seems like after I cut the hole, the spider didn't seem as stiff as it was before I cut it. After I forced it on the shaft it seemed be a little stiffer.

Next time,
Jack

[dneisler]

I didn't stated it above, but when I took mine appart I found that I did not cut a hole in the spider, it was the Y axis I did that on. So the idea that weakened it doesn't work now. I can easily wiggle mine in the coupler. I can just drop it on the coupler, no wiggling or anything to make it go on. I am going to try some tape or soemthing to see if I can get it to tighten up and and test. If that works not sure if I will cut one from delron, or buy a different coupler or what yet. I am open to ideas and suggestions.

[project5k]

yea, i ran into this motor shaft length issue on my y axis as well, i ended up drilling a hole through the spider to make clearance for it, and then slid the lovejoy deeper onto the shaft.. i havent hooked up power to the y axis yet to see how it runs, but i cant see it causing a problem... and getting the 3 parts to mesh was a job, all of mine are really tight..

ofcourse drilling the spider was an easy decision for me as i have extras laying around from other projects, so if i kill one, its no biggie to slip another one in...

i think i remember mcmaster selling spiders of a different material, something to look into...