[Rickyg]

hey guys.
Need some urgent help. I need to cut some carbon fiber this weekend, it being expensive I thought I would cut some fiberglass parts first just to make sure everything was in order.
Well its not, it seems that when I try to cut what should be a half circle or a rounded end to a rectangle I get this out of round situation. I am going to try to attach some pics to help illustrate the issue.you will see the small holes are not very round but thats always been a problem for me and frankly not something I was to concerned with. Thats what drill presses are for. But the corner, not so good.
Also if you look at the strait cutouts in the images, they are funky as well.
I am not sure the reason for it.
Any assistance is appreciated.
I am using Mach3 with CAMBAM if that helps.

[LeeWay]

I think what you are seeing is in the machine. Seriously doubt it is a Mach or software related issue.
What does your machine look like?

[Rickyg]

I will upload a pic as soon as I can get it off my phone.
I am wondering if my steps per inch is gone out of whack and will check that.
While I struggle with my phone and its camera any thoughts on what could be the cause would be appreciated.

[Rickyg]

Picture attached.
It s desktop home built and it works great till I started seeing this happen

[LeeWay]

My guess is that you are seeing flex in the system. If you have unsupported rods on the Y axis or gantry, then that could very well be a major culprit. The tougher the material you cut, the tougher or more rigid a machine has to be.

[Rickyg]

The rods for y are supported.
I did just check my steps per inch and they may have been off a bit. I am wondering if the machine is simply loosening up from use and the steps have gone out of whack.
I re calibrated the steps and saw that x was off by nearly .25 of an inch over 4 inches. At least I think that's what I saw.lol. I did the cal so many times I am not sure anymore what it was to start.
I am also going to check to see if I have microstepping on as I think that its not turned on as the level of accuracy I was getting in the past was find for plywood parts, but now I am getting into more detailed stuff and need better fit and finish.

[LeeWay]

The next thing to test is backlash. Since the rails are supported and all the rails including Z are tight and the bearings aren't worn, then the screw or nut would logically come next. 1/4" in 4 is a lot.

[Rickyg]

hmm not sure how to test backlash. I have anti backlash for each axis in the design of the unit but not sure how to test it.

[Rickyg]

I also checked I am at 1/8th microstepping, perhaps its time to go 1/16th. Not sure what that will get me, but I will set it and recalibrate.

[LeeWay]

If the screws and nuts are all tight and in good shape (including the end mounting points and motor couplers), then you may have a CV issue with Mach depending on the speed of the axes. I always just use the standard default CV settings, so someone else may have to help you with that.

[Rickyg]

Thanks Lee for your getting back to me
I will check all the set screws tomorrow. Boss says I am taking her out tonight, so it has to wait.
I did one last calibration on steps and I get a consistent 4 inch move now. So this is good tomorrow I will check the screws and cut a part to see if there is improvement
If not then I need to look some more.

[Rickyg]

went over all the mechanical s this morning..
I could tighten the shaft couplers a little, but they didn't seem real loose.
Nothing else was off. I will lube the rails and shafts before my make a test cut later today.
I will also again double check the steps per inch.
I have no other clues as to what could be the cause but will post back the results of the test later today

[LeeWay]

If you are running fast cuts, then the CV setting may be what is needed. Try running a small part using exact stop and see if it comes out alright. If it comes out better cut, then you may need to increase or decrease the CV angle and switch back to constant velocity mode. In exact stop mode, the machine will operate rougher. More jerky. Slower too. CV mode tries to keep it smooth when switching axes direction and tries to maintain speed around corners. The amount of corner roundness will vary with different CV settings.

[Rickyg]

It appears that I don' t have CV angle enabled on my machine.
I have printed out the details on the setting and will give it a try.
also will try exact stop with some hole tests later today.

[mactec54]

Rickyg

Do you have V-rollers running on aluminum for the Z axes rails, if this is what you have, then you will find it to be, where your biggest problems are coming from

[Rickyg]

no V rollers
CNC Bushing Linear Bearing Block:vxb:Ball Bearing
I read to much negative around v bearings.

[mactec54]

no V rollers
CNC Bushing Linear Bearing Block:vxb:Ball Bearing
I read to much negative around v bearings.

That's a good step in the right direction, not the best, but a lot better than the v-rollers

What kind of screws are you using, do you have any backlash with the screws or the thrust bearings,for the screw mounting

Are the couplings aluminum or steel

Have you set the step/per with an indicator, .0005" reading or better

For a test try running at a slow speed to cut a hole

[LeeWay]

Hi Ricky.
I looked at the picture again. Have you changed the rails since the picture was taken? Those appear to be unsupported rails. Supported rods don't typically mount with end collars like the image shows. Supported type basically sit on aluminum extrusions that are then bolted to the surface from 2 sides along it's length. They require split type bearings, where a solid rod can use closed bearings.
If your rods are only supported on the ends, then moving your parts very close to one side or the other will yield better results.

[Rickyg]

ah interesting idea
No I haven't changed the rails. I never thought of them when I was asked the question originally.
I hope to test cut later today and will move the parts to the side as much as possible.
I will also test exact stop and set the CV angle since I didn't have it set.

[hjl4]

Exact stop will give jerky motion on radius. I strongly suggest you rethink that strategy as it is hard on the machine, even if you have a rigid machine, it just doesn't sound too healthy, especially if running high IPM.
Also the Absolute and Incremental settings have alot to do with it.