Problem with Z being perpendicular to work piece

[drodda]

I am surfacing an octagon surface on the bot. When the bot goes around the surface using a 1" bit it seems to leave lines in two places. 1st place is around x-5,y-30 as it turns the corner of the octagon. It also leaves lines at x-65, y-30. I have checked all the usual things. The weird thing that is stunping me is that the lines look like the Z axis is tilted in the x direction. at the 5,30 location it looks to be tilted left and at the 65,30 loaction it looks to be tilted right. The slight marks are large enough to catch your fingernail on and this is in oposite directions from one side to the other. These two spot are the only area that the bot travel in an almost straight y direction.

Any ideas would be greatly appreciated. I have spent all day calibrating the Z axis and can't get a readable difference on my digital guage in any direction that I rotate my router. It's hard to adjust if you can't find any error to adjust out.

My machine is a PRT alpha 120 x 60.

[knight_toolworks]

I had this same issue cutting a bevel on a picture frame using a 1" bit. I run the path a few times and it cleaned some of it up. but I wonder if it was some error with vcarve and the depth of the bit. larger bit would fix it of course.

[drodda]

I am using a 1" bottom cleaning straight bit. I have used it many times with success. I now have raised my table height to reduce my Z axis from 12" to 6" and all of a sudden this problem started. When you plane the table surface it is dead flat no marks just cutting this surface in an octagon file.

[drodda]

I am getting this Error while shaving .02" off the top of a solid cherry surface. So It is not a very deep cut to make.

[jim_ludi]

Since the error is in the Y direction: If you own a PRT I'd look to make sure that the wheels on the Y car are always in "centered" contact with the V on the rails as they travel its length. If the wheels are forced to ride up the shoulder of the V on the rails (out of parallel rail symptom) the Z will be moved out of plumb (same for gantry wheels and depth). I lay a straight edge on the table and run a dial indicator (mag mounted to the car or gantry - usually at the corners) along its length to check for this condition; looking for vertical deflection.

If you own a PRS the problem could be a a twisted gantry beam.

I'm guessing that you use a trammel when you're checking your Z do you check locations across the width of the table?

[Brady Watson]

My guess would be that your rails and/or v-roller bearings need adjustment and maintenance. It helps to dangle the X motors and glide the X car back and forth looking carefully at the v-roller wheels and how they contact the rails. It is very possible that one side could be riding up a bit, causing the entire end of the car to lift up. It's worth a look.

-B

[drodda]

If the car was riding up on the edge of the rail would this not show the same error or tilt condition at the same point on the Y carrige? The tilt seems to go one direction when it is at 5,30 and the opposite at 60,30. I have taken it all apart and filed the burrs off the rails and cleaned and removed all debris in the V roller. When Put back together It is still showing the same error. I am going to diagnose more this weekend but had to get the current project completed so I could deliver it this weekend. SO I just spent many hours sanding the Cherry to remove the lines left by the bot. Not a fun time.

Thanks for the help. If you can think of any other suggestions I am all ears.

-D

[Gary Campbell]

Dave..
We had the same problem while doing an area clear on a cherry plaque. It only made the marks during the profile path when all the raster clearance moves were done. Our problem, as we found it, was the bit running into angled grain and just "digging" in .0015. We were not able to determine if the Z was actually flexing, or the piece may have lifted. Either way, there was a small amout of additional sanding. Next time I will try modifying the parts file to run that inside profile pass in climb mill direction to see if that helps.
Gary

[Gary Campbell]

Dave..
I forgot to mention that it did not happen on a test piece of mdf with the same bit.
Gary