[ecurb5]

Just started cutting some 3D stuff. The first piece came off with some weird lines in the cut where I expect smoothness. In vcarve all looks as expected, but the final cut has a sort of wave pattern. I'm guessing it is something in the hardware, but have no idea as to where to start looking for problems. I am also running Mach3.

Bit is a 1/8" ball nose carving bit.
Stepover of 5%
The lines that are visible in this cut run north/south, the toolpaths runs east/west. I tried rotating the toolpath 90deg, and the artifact lines also rotated, so they are always 90deg off from what the toolpath is.

Any ideas?

Attachment 283090

[AORD]

Perhaps try cutting a flat surface and see if you also get similar results.
Can you step through the code (in mach3) and see the tool path dipping and rising at the waves? Then you will know it is the code.
Can you change the decimal rounding in vcarve? (larger decimal places would be better).

You probably want to ask this question in the http://www.cnczone.com/forums/vectric/ section.

[ger21]

You probably want to ask this question in the http://www.cnczone.com/forums/vectric/ section.

No, it's not a software issue.

My guess is something mechanical causing a vibration, either the linear bearings or possibly the rack and pinion. Although it's strange that it rotated 90°, because typically it would be a single axis having the vibration.

Are you running in exact stop mode? What are all your CV settings?

Does the machine appear to run smoothly through the smooth parts of the dish, or is it stuttering?

[ecurb5]

Yeah, not a software problem.

I did find some unwanted vibration, cleaned the rack n pinion and that greatly reduced that vibration, but there still seems to be some left that I have not been able to track down. After cleaning rnp, I ran another test cut that looks a bit better. Ran it at 80ipm, and half way through dropped it to 50ipm, running at 14k rpm.
Attachment 283496

Sent from my SM-G925V using Tapatalk

[kaetamer]

I have the same problem with my CNCRP 4824. Like you I cleaned, tightened, squared, lubed, still the problem persists. Not sure what to do next. It is interesting that it occurs 90* to the direction of travel of spindle. I'm just guessing here but doesn't that imply it shouldn't be a R&P problem because it should be eliminated when the spindle sweeps along the Y axis and the X axis is immobile. Similarly, it shouldn't be a linear bearing problem for just the opposite reason. As the spindle sweeps along the X axis, the Y axis (and linear bearings) are immobile. It seems as though it ought to be related to something common to both axes - the Z axis itself. What problem with the Z axis would cause this? Don't know. I have noticed this though. I've cut a lot of v-inlay projects and have not noticed this pattern in the "male" components carved with a V-bit; the carving pattern in that case is serpentine and irregular, not rastering as in the problem carvings. Bit deflection? BTW, I also had the problem (so far) cutting walnut.

I have deliberately discounted the possibility that there could be completely different problems occurring in both the R&P and linear bearings which cause the exact same cutting error. Seems unlikely, if not very unlikely.

Perhaps Ahren or Cory could provide some insight. At this point, 3D carvings are an unwanted exercise in sanding around delicate features...

Thanks. Scott

[ecurb5]

Agreed, the z-axis was my next target to shoot at (probably should have been the first, but the pinion gears needed to be cleaned anyway).
Beginning to think that R&P is not all that desirable for doing 3D carvings.

-Bruce

[hub]

I don't know what kind of machine you have, but I'm guessing it's not rigid enough. ?
(Vibrations etc)

[kaetamer]

Beginning to think that R&P is not all that desirable for doing 3D carvings.

-Bruce

I'm not convinced the R&P is the problem. Many machine manufacturers use it with excellent results. We're probably overthinking this. It'll turn out to be related to sun spot activity or the Dow Jones.

[ecurb5]

I don't know what kind of machine you have, but I'm guessing it's not rigid enough. ?
(Vibrations etc)

Sorry for not being clear. The reason I posted this question in the cncrp area is because I have a cncrp 96x48 machine. So yeah, it's fairly rigid...

[hub]

Sorry for not being clear. The reason I posted this question in the cncrp area is because I have a cncrp 96x48 machine. So yeah, it's fairly rigid...

No problem, my fault, didn't check the area...
The Z axis is an AL profile.. I had that design on a previous design, wasn't rigid enough.. It was 2 x 90x45mm heavy duty profiles.
What feed & RPM did you use?

[ecurb5]

What feed & RPM did you use?

See my post, #4.

[hub]

Ok, sorry
Try maybe 30ipm. If it's smoother then you might have some vibrations..
Also try pushing/pulling on the spindle. If it flexes, you need to go slower. If you have a dial indicator, you can measure how much it flexes.

Just my guess.. Recently got similar results cutting mild steel with too high feed; my machine isn't rigid enough to cut steel well. Also problems with AL if going too aggressive.

[ger21]

Also try pushing/pulling on the spindle. If it flexes, you need to go slower.

Not necessarily. If I pull on my collet, I can move it a lot, and yet I get perfectly smooth 3D cuts. And my machine has a lot of flex.

This is more than a lack of rigidity. It's a mechanical issue that is actually causing the tool to move in the same place every time.

Run the code in the air, with the spindle off, and your hand on the spindle. You can probably feel the vibrations with your hand.

[ahren]

There are a couple of things I would check. First of all, I have seen machines vibrate quite a bit when geometry is made primarily of bezier splines -- these seem to result in a lot of little moves that Mach has trouble smoothing out even with CV. We have had customers with the issue you are showing (including differences in cut quality with different speeds) and geometry like this. Once they fit arcs to the geometry to replace the splines, the resulting cut quality was better and they were able to cut at higher speeds.

Secondly, I have seen this issue occur when a router / spindle was not trammed correctly. I'm not sure I understand exactly why, but after the customer squared up and trammed the spindle, his 3D sculpting artifacts went away.

The issue could potentially be deflection related as well, but generally only if things aren't adjusted correctly. There are eccentric bushings that can be adjusted on the z axis to set the pre-load on the rails. It's worth checking these, but generally they are set correctly by our technicians. You may also want to tighten the cam mechanisms on the bottom of the gantry axis, as this is the next most likely source of mechanical slop in the system. If you need help with any of this or would like to discuss further, feel free to contact us directly as well.

Best regards,

Ahren
CNCRouterParts

[ecurb5]

A little more poking around the machine I think I may have found the source of the vibration. Must be time to replace the pinion gears..... I disengaged the pinion from the rack and the vibration for that motor went away (checked all 3). I've already cleaned them a few days ago, but the pinions do look a bit worn. Ah the joys of fine machinery...

Will report back when new parts are installed.

-Bruce

[ecurb5]

A little more poking around the machine I think I may have found the source of the vibration. Must be time to replace the pinion gears..... I disengaged the pinion from the rack and the vibration for that motor went away (checked all 3). I've already cleaned them a few days ago, but the pinions do look a bit worn. Ah the joys of fine machinery...

Will report back when new parts are installed.

-Bruce

Replaced the pinion, and updated the rack mounting. Looks like it solved my problem. Thanks Ahren!

[dgage]

Replaced the pinion, and updated the rack mounting. Looks like it solved my problem. Thanks Ahren!

Glad you figured it out (with Ahren's help) and thanks for sharing. How long/much had you been using the machine with those racks so I get an idea of when I might need to change/check them? Thanks.

David

[ecurb5]

Glad you figured it out (with Ahren's help) and thanks for sharing. How long/much had you been using the machine with those racks so I get an idea of when I might need to change/check them? Thanks.

David

How long have you been running your machine? I'm just a hobbyist so my machine doesn't get fired up too often. I've had it for a couple of yrs now, but it only gets used for maybe an hr every week or three...

I just used the racks I have, spent maybe $20 on the bar steel from a local steel shop (box box store will cost 3-5 times as much). Bought a counter sink bit, tap, and a bunch of screws from ACE and went to town on drilling/tapping holes.

[dgage]

How long have you been running your machine? I'm just a hobbyist so my machine doesn't get fired up too often. I've had it for a couple of yrs now, but it only gets used for maybe an hr every week or three...

I just used the racks I have, spent maybe $20 on the bar steel from a local steel shop (box box store will cost 3-5 times as much). Bought a counter sink bit, tap, and a bunch of screws from ACE and went to town on drilling/tapping holes.

OK. So this rack wasn't supplied by CNCRP? I'm thinking the CNCRP gear rack might be harder and therefore last longer than the rack you may have purchased...at least that would be my expectation based on how many people have used the CNCRP Pro machine for business, which is many more hours per week than you use in a month or two. But thanks for answering so quickly.

[ecurb5]

OK. So this rack wasn't supplied by CNCRP? I'm thinking the CNCRP gear rack might be harder and therefore last longer than the rack you may have purchased...at least that would be my expectation based on how many people have used the CNCRP Pro machine for business, which is many more hours per week than you use in a month or two. But thanks for answering so quickly.

No, I didn't buy the rack cncrp, but it is the same hardened steel rack they use. That's not the problem. The problem was the mount method in the orginal standard version and how it meshes with the pinion gear.

Sent from my SM-G925V using Tapatalk