V carve issues with Colorcore

[letterman7]

Hi gang,
New to the forum but not to routing. Cut my teeth on a 6 x 12 Vutek years ago until the company folded, so I went and bought a used PR48 (at least I think it's a PR... beige computer box with steppers..). Anyway, over the past several years I've been doing pretty good with FCO letters and v carving HDU and pvc. I run VCarve Pro with everything and I like the way it integrates the SB program with it. The issue I'm having is with a repeat customer who needs these ColorCore polyethylene signs. The material is 1/2" thick, the bit is a 120° 1" v-bit running at about 1/2" per second at 21000 rpms. The issues are seen in the photos below. I have the machine as square and level as I can possibly get it without using milspec laser levels, and the spoilboard is flat.
I don't think it's a program error since it's on every letter or number cut. I've tried varying the depth of pass and it simply changes where the error occurs. What bothers me is that not only is it not cutting straight, the corners aren't even square and sharp as they should be. The router is literally a brand new Porter Cable with new collets, so I can't think that there is any runout on the bit itself (though I haven't rolled or mic'd it to see if the shaft is dead straight... no reason to think it wouldn't be). I have the same issue to a lesser degree on the HDU and pvc, and those can be hid with some sanding - on ColorCore you can't do that.

Any ideas of even where to start? I've been able to slip these by for a little while, but it's really starting to bug me and I want to be able to give these folks an A1 job on these... and these signs aren't it!

Thanks,
Rick

13663

13664

[chiloquinruss]

At first glance it appears your material is moving and not the bit. How are you holding down the material, tape, vacuum? It looks like the material might be fluttering while making the cuts. I cut some small badges from thinner stock than what you are using and the issues were all about making sure the material never moved. BTW I was using two sided tape. Russ

P.S. Oh Yeah, I turned off my dust collector so it wasn't trying to pick up the material along with the chips! :) :) :) Russ

[tmerrill]

Hi Rick,

You have a mismatch between the v-bit angle the toolpath is calculated for and the v-bit angle you are making the cut with. In this case it looks like the toolpath was calculated for a 90 degree bit and you are cutting it with your 120 degree bit.

Double check the toolpath, including you have the same angle selected as in the bit's name.

If all this checks, your 120 degree bit may not be a true 120 degrees. There is a great technique for checking this:

http://www.vectric.com/forum/viewtopic.php?f=39&t=7239

If all this checks out fine, can you post the file?

Tim

[tmerrill]

Just did a quick example to show you and the difference between bits is not as much as a 90 verses 120, more like a 110 verses a 120.

The left rectangle was toolpathed and previewed using a 120 degree bit.

The right rectangle was toolpathed with a 120 degree vbit but previewed with a 110 degree bit. It looks very similar to the square edges of the number 1.

Let us know.

Tim

[letterman7]

Hi Russ,
The material is screwed to the table. In this particular case, there are drilled holes through the sign anyway, so I have additional screws holding through there as well.
Tim, the toolpath was created using the 120 degree part. I've checked the angle via another 120 degree bit I have that's 2" and it matches. This particular bit is one that has the replaceable carbide tips, but it appears to be straight and true. On a hunch I ran the larger 2" "solid" carbide bit and it did the same thing. What part of the file would you like to see, assuming I can copy it off the router computer?

[tmerrill]

Not sure what program you are using, but it would be the design file. For example, if you are using PartWorks it would be the .crv file.

Tim

[letterman7]

It's Vetric's VCarve Pro6, their latest version. The forum won't accept the Vectric extension (.crv) and the .sb is too large a file. I can always send it to you in an email if you wish.

[tmerrill]

You can zip it and the forum should accept the .zip file extension. It would be best to post it here if possible so other eyes can look at it also.

If it still doesn't work, let us know.

[letterman7]

:D Ah.. didn't think of that. Let's see if this works...

13666

I had to append the file to just a couple sample cuts due to the overall size of the project. But what is contained in the file would pertain to all the other cuts within the project.

Thanks for the help so far, Tim!

[tmerrill]

Rick,

I only found one issue with your setup in VCP 6. You have the feedrate set to 150 inches per second and plunge rate set to 100 ips. I think you probably want units of inches per minute, plus this doesn't agree with what you posted in your initial post.

I ran your .sbp file through the ShopBot control software in preview mode and everything looked fine, so with the exception of the excessive speeds everything looks fine with the file. (In case you are wondering, I don't think you have the Previewer in the older control software that your machine would use).

You seem comfortable with the bits being 120 degrees so the only other thought I have is the accuracy of your X, Y and Z axes. In other words, for Z, after you zero it to the machine bed or material surface, and manually move it upwards 1, 2, 4 inches and measure, has it actually moved that distance? For X and Y, after setting them to zero and giving them a manual move command, do they move the exact distance?

Hopefully others will take a look and have other suggestions, but I believe we are beyond VCP as being the issue.

Tim

[letterman7]

That's my next step in investigation Tim. The feed rates I somewhat ignore since they are older stepper motors and will only move so fast. I also contracted with Brady Watson to stop out one day and troubleshoot - his initial thoughts were simply wear in the pinions or something like that. But it'll be good to go through the machine and get it right once and for all!

[letterman7]

Ok... spent about three hours this morning troubleshooting without any forward movement. I ran some test cuts on a new file on the old version of VCP and the same error showed. Tried it with a known 90 degree bit and the same thing. Disconnected one stepper motor at a time and tried some test moves to see if one motor was slacking and each motor on each axis moved exactly where it was told via the 'Bot programming. On thing I did notice was that one of the X pinions wasn't quite in line with the other.. it was off by about a tooth and a half. A quick re-set and another test and still no love.

I'm out of ideas. I've ordered new pinions from SB, though I can't see the current ones being worn to the point that they are causing issues. They were all replaced in late 2008, and the router only runs maybe four days out of the month. We'll see. I also have a 'Bot expert scheduled to come and troubleshoot at the end of the month. Hopefully I can get this resolved and post the results!

[robtown]

Have you tried making the same cut on different parts of the table?

[letterman7]

Yep, in fact I resurfaced the table and tried again with different materials. Same error, and I'm believing it's a Z axis error as pointed out by a couple people. Whether it's the pinion or motor wearing I'm not sure, or a board circuit wearing out. More testing to be done...

[blackhawk]

Rick - Have you checked the mesh of the pinion to the rack on the z-axis. The z-axis is not spring loaded like the x and y. I had this problem once on my PRT.

[jerry_stanek]

If it is a PR check the coupling to the Z see if the set screws are tight. also the ball screw may be getting worn out

[letterman7]

Solved - Brady came out today and we went through the machine piece by piece. Brad had it on the money - the Z mesh was really out and it wasn't really noticeable unless the head was under load. Brady squared everything up and re-set the lash on the rollers and voila! Seems to be right on the money with whatever bit angle I put in it. Thanks again to Brady and the folks on the board for troubleshooting!

R

[Brady Watson]

Rick,
It was a pleasure working with you today. Your PRT has plenty of life left in it!

-B