[twocik]

I'm experiencing some serious play in my CNC X1. I can't seem to locate the problem and need some help. All of my aluminum parts are missing steps and look pretty bad.

I've tried to adjust the shaft collar and work the play out in the gibs (100%
perfect on the gibs). I can still push and pull on the Y table and fell the it move about 2mm - 4mm. I'm pushing in a forward to backward motion (toward and away from the column).


I've uploaded some pictures of my parts and if you look closely you'll see the offset and a really bad looking layers surface. I've lower my feed to 3 ipm and reduced my depthInc to .04 per pass, so the only thing I can come up with is the play in the Y axis, but can't figure out how to correct it.


Any advice would be greatly appreciated.

Chris

[cncbasher]

you may be running the controller at too high a step rate , or even the torque rating of the steppers may not be correct , so more information on your setup should help to narrow down the problem , you may also have resonance or vibration problem , excessive play etc in your z axis , can all give this sort of result

which motor and controller are you using etc , software ?

[j.guenther]

I had a similar problem with my CNC X1 on the X axis. The problem was that the X axis lead screw nut was loose. Since you say you can actually move the table by applying force either toward the column or away from the column there is something loose some where. I would check the nut to make sure it is securely fastened to the base. You should also check the preload on the bearings, if you have bearings on the lead screw. Check the backlash adjustment on the nut as that might be a source of the problem, but I doubt that 3 to 4 mm of movement would be caused by wear in the nut.

John Guenther

[LeeWay]

It could be power, rigidity or backlash in the nuts or any combination of these.
More info about the type of motors, drivers, power supply, screws and nut type would also help.
Sometimes slowing things down is not the way to go either. Especially with different grades of aluminum. If everything is running too slow and there is slop somewhere, you will get a lot of catching and jerking with vibration.

[philbur]

The offset doesn't look like backlash. Backlash will give you a step at the change of direction. Are the inner and outer diameters the correct diameter and are they circular. Could be stepper calibration. when you say that you can move the y-axis 2 to 4 mm is with the machine powered up. With ballscrews and without power-up it is quite possible you can move the table by hand. How did you produce the G code, have you looked at it to see that it is correct. The rough surface finish on the walls of the component looks like cutter deflection to me. Did you do a finishing pass or was it cut as a slot.

Just some thoughts
Phil

[Lexx0001]

I'm probably missing something, but 2 to 4 mm play seems rather excessive. I would think with that much give, the Y axis is being pushed and pulled by the end mill instead of the other way around. The motor mount to the slide may not be tight, coupler could be slipping, or the nut could be loose inside the saddle.

I guess to help narrow it down, when you can move it, does the screw move with the saddle, or stay with the machine?

Just my 2 ¢

[Rally]

I would start by looking at the Nut and where it is mounted to the Machine.
If my Eyes are not cheating me you should also change the e-mill it looks like it is smearing instead of cutting.
The table should not move no 4mm no matter what!!
I do however have to agree that it looks kinda smooth for a backlash problem.

[philbur]

A "normal" ballscrew is not self-holding. For example a mill head can fall under its own weight, with the power off the stepper. Ask Tormach PCNC1100 owners.

So with the power off the y-axis motor it is quite possible you could move the table by hand and turn the ball-screw. Depends on how hard you push.

Phil

The table should not move no 4mm no matter what!!
.

[omkargupta]

PLEASE TIGHTEN THE CHECKNUT OF BALL SCREW. Either Change Ball screw

[LeeWay]

If the machine is powered up and you can still move the table, then it could be a loose drive nut, screw mount, coupling, or weak motors.
Table movement with no power is inconclusive because is possible to move them unless they are something like threaded rod and have a very fine thread. More input is needed.

[twocik]

Wow you guys work fast!! Sorry I wish I had Ball screws, but I don't (stock lead screws).

My machine is bolted down in my water tight enclosure, so I don't believe I'm getting that much vibration. As for the Z axis, adjusted this before anything (works great).


- Harbor Freight X1
- Hobby CNC Board & Steppers 200 oz Kit
- Steppers running at low speed () for more torque
- Running Turbo CNC for my controller software
- Extended X & Y axis tables
- Fusion Motor mount kit
- Using 6061 aluminum
- 3/16 2 flute carbide end mill
-




j.guenther

I'm going to check again. The only thing I wish I would have thought of before I made this enclosure is, how am I going to get this in & out for maintenance.





"You should also check the preload on the bearings, if you have bearings on the lead screw. Check the backlash adjustment on the nut as that might be a source of the problem, but I doubt that 3 to 4 mm of movement would be caused by wear in the nut."




I'm using a trust bearing setup, but never thought about the backlash nut. The last time I played around with the nut I broke it and had to buy a new one. I'm going to take the mill out of the enclosure again and try and find this problem.






"Especially with different grades of aluminum. If everything is running too slow and there is slop somewhere, you will get a lot of catching and jerking with vibration."




6061 aluminum, adjusted the tables with an indicator (.02mm off ).






"Are the inner and outer diameters the correct diameter and are they circular. Could be stepper calibration. when you say that you can move the y-axis 2 to 4 mm is with the machine powered up. With ball screws and without power-up it is quite possible you can move the table by hand. How did you produce the G code, have you looked at it to see that it is correct. The rough surface finish on the walls of the component looks like cutter deflection to me. Did you do a finishing pass or was it cut as a slot."




Yes they're circular, but the center is offset. I've cut the part out on a soft piece of wood and came out perfect looking. So I'm guessing with harder material, this is what's pushing my part around. No ball screws. I produce my G code thru CamBam, but this isn't the problem. Well I didn't get around to making a finish pass, only because the part was offset pretty bad and couldn't be saved.

[twocik]

Ok so just before I started to take apart my mill to locate this problem, I took a closer look at the 6061 aluminum plate that I cut it out of. If you take a close look, you'll that the 3/4 of the parts walls look really really smooth. I'm kind of stumped now. I've uploaded some pictures to see what I'm seeing, any ideas at all pleas feel free. ???



My guess -

Since there is a play in the X axis (2mm - 4mm), when the end mill was cutting along its path, the end mill stayed snug against the side that was pushing on it.

What gave me this idea is on the part that was cut out in my first post, you'll see the walls on the right side are not bad looking at all, but on the original raw material you'll see the layered surface were that part was cut from. It's weird I know, the side that came out nice on my part is the side that looks like crap on the left over piece of raw 6061 material clamped to the table and vise versa for the other side.


So it's definitely from that damn Y axis slipping, I really need to find what is going on under my mill. I'm just about to take it out of the enclosure and try my best to find the problem.

[twocik]

Ok I've narrowed two of the now three problems down. I played around with the backlash screws on the lead screw nuts, took most of the play out. If I pull on the tables in a forward to backward motion I can feel the Y axis move just a hair, not sure if it's enough to really make any problems, but think it's a hell of alot better.

My second problem is with the tightening the backlash screws just enough to get rid of the play. If I tighten them too much the lead screw makes a really high pitched noise (very loud), so when I loosen them I can feel some play, but I think I've got the Y axis the best I'm going to get it.


Now on to my third problem, the Y axis.

After tightening the backlash screws and the gibs screws to where they're just about perfect, I'm now having trouble when (only) jogging right on the last couple of inches on the X axis table. I would say it's about 2 inches of table right before the motor stalls. I continued to jog at full speed (alt + right arrow key) after stalling the first time and it stalled again, but this time it won't move at all until I jog left.

Now I know you're going to say loosen your gib screws just a bit, if I loosen them just a hair and bring the X table back to center, the whole table is now very lose. So I can't figure out if my table has a slight bend in it or_____________ ????

[Pres]

What kind of lube are you using in the slides and leadscrew?

Perhaps some good/different lube will minimize the sticking?

Pres

[twocik]

I'm using white lithium grease

[LeeWay]

What kind of lube are you using in the slides and leadscrew?

Perhaps some good/different lube will minimize the sticking?

Pres

That was going to be my next suggestion.

On such a small machine, you really need to have all the travel you can. The troubles you are having is precisely why I chose not to cnc my X2. Not near true enough motion to suit my needs. I built my own instead using ball screws and linear bearing blocks. Pretty pricey, but I needed something I could count on to run one or two days a week with minimal upkeep.

With just the standard screws and ways on your machine, you are going to have to compromise. You will have to find the sweet spot that has the least play, but will allow you to still move. All three axes will need to be done this way until you throw some money at the problem. You could also spend some time lapping the ways. Stiffen up the column. Design and install some antibacklash nuts. Design most of your parts so that you are conventional cutting more than climb cutting. On more critical parts, take extremely light passes, just more of them. Also allowing for a finish pass will help. You don't have a Beaver. You have more of a wood pecker. You will have to make considerable compromises with the machine to get any kind of acceptable results.

Do so searches on here. I have seen some threads that show how to get better preformance out of an X1 without breaking the bank. Post the things you do to help with your problems as well. I would like to see you get some better results with it. It is possible I think.

[thackman]

twocik

That's an interesting part your cutting, mind if I ask what it's for.

Could you post a few pic's of your clamping setup? I have some parts that need to be drilled, pocketed, surfaced, and profiled and I've been having trouble finding a good fixturing setup. That's a fairly complex part, how many setups does in take to produce? I've tried drilling the bolt patterns first but bolting it down through those holes doesn't locate the part accurately enough and interferes with surfacing.

FYI, I have an CNC'd X3 that I'm just learing to use and my surface finishes are much worse than yours. Guess I need to get that flood coolant setup.

[twocik]

Hi guys and thank you all for you advice & help!!!



"With just the standard screws and ways on your machine, you are going to have to compromise. You will have to find the sweet spot that has the least play, but will allow you to still move. All three axes will need to be done this way until you throw some money at the problem. You could also spend some time lapping the ways. Stiffen up the column. Design and install some antibacklash nuts. Design most of your parts so that you are conventional cutting more than climb cutting. On more critical parts, take extremely light passes, just more of them. Also allowing for a finish pass will help. You don't have a Beaver. You have more of a wood pecker. You will have to make considerable compromises with the machine to get any kind of acceptable results."




Leeway I believer you're right, this is why I'm going to either keep the machine for plastic, light metal work and engraving or I might sell it. The part in the picture is .38 thick and was taking passes of .04, but I think it will perform much much better now that I've located most of the problem (i've found the sweet spot last night. My travel is limited, but hey it's only a few inches.

I really can't spend anymore time on fixing up the little warrior at the moment, I really need to finish my parts. However if I could find a company that makes a ball screw & nut kit, I'd buy it. So for right now I think I'm going with the Tormach system, waiting on my loan rep.


Anyone here located in the Naples, Florida area ?







"That's an interesting part your cutting, mind if I ask what it's for."


This is an adjustable locking support bracket, it belongs to another part of my invention. I'll post some pics of it when finished, along with the other part.

[jhar123]

based on the information presented; i believe your machine is ok your problem might lie in the application arena, usually if there is lost motion in your machine you will see it especially will circle milling. the diameter will be egg shaped. the way to check backlash i'm sure you already know, just make a linear move usually the smallest incriment available both + & - and see if you machine repeats, is you machine gibed or does it have linear guides if it is gibed sometimes gibes not being set right will cause lost motion either from slack or dragging, slack tighten dragging loosen you'll see with a.0001 indicator set up parallel to the way surface skip if it is dragging or if it is linear guides as opposed to gibed with boxed ways fix indicator to spindle and checke motion of table. Hope this helps some.