Bad Cuts

[chebytrk]

Looking for some help in trying to figure out why my project is not cutting my project out correctly. Just got my machine back from the shop and ran a project which carved out fine. However, as you can see in the pic, it is not cutting out correctly (the cutout is off set and not centered in the cutout). This was actually happening before I put it in the shop, but I kinda hoped that with the stuff they did, MIGHT have fixed my bad cuts when my projects are in the final step and routing out. Please see how everything seems "off set" from top to bottom as well. This is our June POM and all I've done is change out the pattern on the sides and top. Nothing else was touched or changed. I did notice that my flex shaft was getting hotter than usual so I've taken out the core and lubed it (with Molly). Everything else seems to be running fine. I even use masking tape on the bottom (thanks to Al Who) and also ran a strip of tape down the center (lengthwise) on the board. I crank it down and I pulled out the red handle. Any suggestions are greatly appreciated so I'll know where to start. FYI.. board was flat and even, the O ring is fine also. I haven't checked the head pressure lately. Could that be it perhaps?
35928 Thanks Gents...........

[Kenm810]

I would Definitely check the Head Pressure, to little or to much can make the project slip,
also check the side guide bar for being to tight against the project board.

Ps. I use three strips of masking tape on the bottom of my projects,
one down the center plus one down the bottom of each edge for a little added traction.

[chebytrk]

So I take it that the side guide bar should not exactly be touching the board? Maybe 1/16 away from the board?

[AskBud]

So I take it that the side guide bar should not exactly be touching the board? Maybe 1/16 away from the board?
Quote from the manual:
WHEN LOADING THE WORKPIECE, there are several critical checks to
make before proceeding.
o Do not attempt to load a workpiece that has a significant taper to the
sides. A tapered workpiece will bind between the sliding plate and the
squaring plate and will damage the traction drive.
o Do not attempt to load a workpiece that varies in thickness by more than
1/16” along its entire length. Using a workpiece with a larger thickness
variation than 1/16” can lead to damage to the machine.
AskBud

[unitedcases]

Whenever I load a board or one of my sleds I take two playing cards and put them next to the board/sled. Then I slide the squaring plate up to it. I then pull the cards back out and that gives just a touch of clearance in the event the board isnt true. I know the sleds are square but I still take the extra couple of seconds to do it. Go through an "X" drive gear a couple of times and it is well worth that 2 seconds.

[chebytrk]

I just checked the head pressure and it's reading between 85-90lbs. Is that acceptable? If not, how can I reduce it? I just don't know what else can be causing this "off set" cutting out of any given project. I'm going to go out and try a small project and see what gives this time....

[Digitalwoodshop]

Thank you for using Tape. So it comes down to a error in the X or Sand Paper Belt Direction. Since you applied masking tape to the brass roller area, is the tape still on the bottom of the board? Look and see if the board lost contact to the brass roller, looking for missing brass roller prints in the tape.

Did you design this project to keep wood always under the rollers? I am betting not. When the board is held by one roller it can tip up and loose contact with the brass roller. I used a method of taping a chunk of the same size board on the head end and leave the board longer. Some use Hot Melt Glue too.... This would help....

Next I would check is head tilt... The art looks level but a quick check would be to use Sensor Data to touch the left and right side of the board with the bit and look at the Z Data. Think it is a board tipping up more....

Also check the level of the in feed and out feed height.

Here are some carrier boards I use....

Picture 1 shows how I place a line in my Designer Art showing the 4 inch block of stay under the roller wood.

Picture 2 shows how I cut 2 clocks using the taped on scrap block for the forward stay under the rollers and the wood was left long for the rear. I cut the rear off.

Picture 3 and 4 shows a carrier board with 2 scrap blocks and then placement of pre cut blocks in the middle.

Picture 5 shows the carrier board.

Good Luck,

AL

[mtylerfl]

I just checked the head pressure and it's reading between 85-90lbs. Is that acceptable? If not, how can I reduce it? I just don't know what else can be causing this "off set" cutting out of any given project. I'm going to go out and try a small project and see what gives this time....

Sorry you're having trouble.

Yes, the project is designed to stay under the rollers at all times, and it looks like you used the proper length boards as I specified in the instructions. Still, the symptoms point to an x-axis tracking issue.

Al mentioned checking under the board for "skipped" teeth marks where the brass roller rides along the underside of the board. If you find a "blank" spot in the teeth marks, that's where the x-axis data became garbled and caused the offset cut out. In cases like this, folks will often discover the board was "bad" - may have had a divot, depression, was cupped/warped - anyway, something along those lines.

Please let us know what you discover and/or if you need more help. Also, the recent Tips & Tricks about maintaining accurate tracking will have a few other suggestions on causes of tracking errors of this kind.

[chebytrk]

Well appears that I'm still having problems and I hope that the team members here can help me find the problem. Here's what I've done so far so please help me in telling me what I'm doing wrong as I'm sure that I am some where.
1. I'm using the Jun POM as a pattern and modified a few things, but only the designs and nothing else.
2. The board I used this time is 23 3/4" x 11 3/4 x 3/4
3. The "add on" board that I taped on one side was 6 3/4".
After I placed the board and it measured everything (stay under rollers, use original size) I had it place project on corner. The "add on" board was placed on right side (pad side). The last time I did this I didn't check the tape for brass roller marks and thought that perhaps this time I should place a second strip of masking tape on top of the first one just to make sure the brass teeth would grab on to something (maybe that was wrong?).
So I decided that this time I would not have it cut out each pattern, but only outline it and if it worked out I could just cut it with my jig saw. Well, as you can see in the pictures, it still didn't line up correctly when doing the routing/outlining. As I mentioned before, the carving is fine. It's just when it comes time to route the carve out that everything goes "off" from what it should be.
Should I have maybe used 2 "add on" boards (1 at each end) and told it to Center the project? I've actually done this both ways before and it usually works fine when I just use one "add on" board and have it Corner on project.
Any and all help is greatly appreciated cause this is really getting depressing and costing me a little on these mistakes I keep making. However, even before when I used a long enough board (adding 8 inches to the overall board project), it still came out just like this. Maybe I'm just losing the "hang of it" on how to do this. Thank you all for any help or advice.

[dcalvin4]

I had this problem an corrected it by reattaching the rubber damper that sits on top of a screw that holds the brass gear in place..read comments from me and others about this problem..http://forum.carvewright.com/showthread.php?14404-inaccurate-drilling/page7 replies number 34 thru 39 The rubber damper is an aid to the spring for the brass roller...and is simply a rubber tube 1/8 o.d. and 1/4 long that can be purchased at any parts store

denny

[PCW]

Jerry,

My POM came out the same as yours and I normally don't run into too many problems. I haven't tried to carve the project again because I have seven started projects going and I'm just trying to get them cleaned up first.

My thinking is that some of us may have gotten a corrupt file when we download the project. Mine came out just like yours did.

[mtylerfl]

Well appears that I'm still having problems and I hope that the team members here can help me find the problem. Here's what I've done so far so please help me in telling me what I'm doing wrong as I'm sure that I am some where.
1. I'm using the Jun POM as a pattern and modified a few things, but only the designs and nothing else.
2. The board I used this time is 23 3/4" x 11 3/4 x 3/4
3. The "add on" board that I taped on one side was 6 3/4".
After I placed the board and it measured everything (stay under rollers, use original size) I had it place project on corner. The "add on" board was placed on right side (pad side). The last time I did this I didn't check the tape for brass roller marks and thought that perhaps this time I should place a second strip of masking tape on top of the first one just to make sure the brass teeth would grab on to something (maybe that was wrong?).
So I decided that this time I would not have it cut out each pattern, but only outline it and if it worked out I could just cut it with my jig saw. Well, as you can see in the pictures, it still didn't line up correctly when doing the routing/outlining. As I mentioned before, the carving is fine. It's just when it comes time to route the carve out that everything goes "off" from what it should be.
Should I have maybe used 2 "add on" boards (1 at each end) and told it to Center the project? I've actually done this both ways before and it usually works fine when I just use one "add on" board and have it Corner on project.
Any and all help is greatly appreciated cause this is really getting depressing and costing me a little on these mistakes I keep making. However, even before when I used a long enough board (adding 8 inches to the overall board project), it still came out just like this. Maybe I'm just losing the "hang of it" on how to do this. Thank you all for any help or advice.

Hi Jerry,

Boy, it still looks like an x-axis error to me. You never told us if you noticed any "blanks" in the teeth marks from the brass roller on this time around.

Here's what I would do if I were you... I would follow the project instructions to the letter and use a solid board - in other words, no "add-on" boards taped on the end(s). Make SURE your board is flat/straight - I see knots on the pine board which tells me it is not kiln-dried Select Pine, but a grade or two below that. That's perfectly alright, IF the lower-grade board is flat and straight, and not too "wet" with pitch/moisture.

I know you are trying to avoid wasting wood, but you have wasted more than 7" inches of wood already trying to do the taped on board ends and the place on corner method (which one would think should work just fine, but...?) Please give it another try, but this time, do it exactly as the step-by-step instructions specify and see if it makes any difference in the outcome.

Again, be sure your wood is "good wood" - no warps, cups, dents, knots, or divots along anywhere the brass tracking roller rides and use a solid board cut to the length specified in the instructions.

I assume you already checked your head pressure, lubed the posts, made sure the rubber O-ring is in place and undamaged - you know, the normal things to check when having tracking errors.

[mtylerfl]

Jerry,

My POM came out the same as yours and I normally don't run into too many problems. I haven't tried to carve the project again because I have seven started projects going and I'm just trying to get them cleaned up first.

My thinking is that some of us may have gotten a corrupt file when we download the project. Mine came out just like yours did.

Hi PCW,

A corrupt file is extremely unlikely. If the preview in your Designer software looks alright, then I think the file should perform fine. Are you doing anything at all that deviates from the written instructions?...such as taped on board ends?...or anything else? Look at the bottom of your board...any blank spots in the teeth marks? Let us know.

[mtylerfl]

I had this problem an corrected it by reattaching the rubber damper that sits on top of a screw that holds the brass gear in place..read comments from me and others about this problem..http://forum.carvewright.com/showthread.php?14404-inaccurate-drilling/page7 replies number 34 thru 39 The rubber damper is an aid to the spring for the brass roller...and is simply a rubber tube 1/8 o.d. and 1/4 long that can be purchased at any parts store

denny

Thanks for mentioning that, Denny. Yes, that is something that PCW and Jerry could check as well. Also, check to see that the rubber O-ring is NOT touching the traction belt when board is under pressure - that fouls up tracking data too.

[PCW]

Hi PCW,

A corrupt file is extremely unlikely. If the preview in your Designer software looks alright, then I think the file should perform fine. Are you doing anything at all that deviates from the written instructions?...such as taped on board ends?...or anything else? Look at the bottom of your board...any blank spots in the teeth marks? Let us know.

Michael,

I have seen stranger things happen with software. Yes I did tape 4" tails on the board just like I do all my projects and Al would be proud of me too. Three strip over the seams (not overlapped) and three strips on the bottom and top of the board. It was rigid.

I did not change the project at all and ran it using Ver.1.164 Windows 7 64 bit. It came out just like Jerry's photo. I would be more than happy to forward the file to either you or LHR to verify the outcome but I just don't have the time to test it at the moment.

I'm just trying to rule out possibility of software error just in case it's not a mechanical or user error.

Jerry, have you ran any projects other than this one since you started having problems? If so what was the outcome? I have ran several since and that was the only one that I had any problems with.

[mtylerfl]

Michael,

...I have seen stranger things happen with software. Yes I did tape 4" tails on the board just like I do all my projects and Al would be proud of me too. Three strip over the seams (not overlapped) and three strips on the bottom and top of the board. It was rigid...

Thanks for the additional info, PCW

Ok, we have a common denominator! Here's what I THINK might be a possible cause for the tracking error. The spot(s) where the end boards are taped on the main board might be acting like a little speed bump (or some type of momentary tracking interruption), which may be messing up the tracking data sent between the brass roller and the x-motor.

The bad data is just enough to throw off the x-tracking so that the carve and perimeter cutouts don't line up properly. Notice the offset is ONLY in the x-direction...the Y-axis is still aligned properly. I'm more convinced than ever that it's a simple matter of x-axis tracking being thrown off "somehow".

When you have a chance, please try the project again with a solid one-piece board (and, of course, make sure the board is flat, etc., etc.)

Thanks again for the feedback. I think we may have nailed the problem.

[chebytrk]

Thanks for all the input guys. Mike, I'm going to try this again tomorrow and make sure that the board I have is flat. I'm still going to leave the one portion of not performing a complete cutout as just outlining should be enough for us to see the results if it's straight or not. As for the "O" ring, it's fine and always one of the first things I check. As for teeth marks from the brass ring on the tape... there were no teeth marks anywhere on the tape telling me that no contact was being made. At least logically that's what it's telling me. I've checked and will check again, but I don't see any teeth marks whatsoever.... anywhere on the tape. That's why I made the comment if perhaps that might have happened because I placed one strip of masking tape on top of the other thinking that it might help give better traction for the brass roller. Also, I have run other projects (house sign and a small sign) and I could tell that it too was off center, but I was able to fix it by using my hand sander to shape it up. Here's a pics of them and you'll notice that along the "rim" of the oval that one side is actually just a little wider than on the other side. The other sign I did, the same thing happened, but that too I was able to adjust with my hand sander. Again, I'll double check everything tomorrow and report all of my findings as well as pics. I thank everyone for helping me out on this it really means alot that you're there to help this ol' guy out.

[dcalvin4]

I had this problem an corrected it by reattaching the rubber damper that sits on top of a screw that holds the brass gear in place..read comments from me and others about this problem..http://forum.carvewright.com/showthr...drilling/page7 (http://forum.carvewright.showthread.php?14404-inaccurate-drilling/page7) replies number 34 thru 39 The rubber damper is an aid to the spring for the brass roller...and is simply a rubber tube 1/8 o.d. and 1/4 long that can be purchased at any parts store
this is a repeat!
denny

[PCW]

Thanks Michael,

You are probably correct and I will try it once I get a chance.

I know it has been talked about lately here and it would be nice to see LHR make some improvement on the sensors.

We learn to deal with problems that arise from time to time due from the sensors and the more we use our machines but I feel really bad for the newer users that get frustrated and either returns their machine or just ends up putting it in the corner when they fail.;)

Your projects are always something I look forward to the first of the month.:mrgreen:

[liquidguitars]

I will take a stab- This looks almost procedural and that would be software however the raster carving of the letters look fine, so i would look at table pressure as a starting point. If it is software some tests will need to be looked into.

Table pressure- How do i know if the table pressure is fine before i commit to running a carve??, testing the table pressure before hand with a bathroom scale is not going to help as each time you load a board you could have different pressure.

How do i do it? first I use a full sleds no taping blocks little risky for my work at 60.00 or more for each top.
second, I never center my board in the CW I not talking about Keypad options but the actual placement of the sled in the unit, I "offset" the board so the board sensor is directly over the 4" tail, one of the advantages is I can lift the long end of the board like a fulcrum to check the rollers for spring compression like in a dirtbike you need some spring. To tight and the board will not lift up at all, to loose and the board will compress the roller springs over .25", as i lift up on the board i want the board to move up about a 1/16" of compression. Boards loaded in the center of the unit make this test almost imposable.

[chebytrk]

Good idea on checking the roller spring pressure. Question.... is it a good idea to pull out the red handle? I've heard different opinions on this. Sometimes I do and sometimes I don't pull it out.

[mtylerfl]

Good idea on checking the roller spring pressure. Question.... is it a good idea to pull out the red handle? I've heard different opinions on this. Sometimes I do and sometimes I don't pull it out.

Up to you...I never use the Red Handle. Seems to me somebody mentioned it helps in preventing "lifting" of the head during a project run, but I have not had any issues with the head lifting at all on either of my machines.

[Jeff_Birt]

I agree with the consensus, this definatly looks like a tracking error to me.

A few other things to check:

1) Make sure the O-ring is still on the tracking sensor.
2) Make sure that the rubber spacer/spring is still under the tracking sensor. (This presses it up to the wood.)
3) Make sure that the tracking sensor is actually touching the board and not getting hung up on the right side belt. You can do this by raising the head up and using a small strip of wood, say 1/4" thick by 3/4" wide, up against the squaring fence. The board tracking sensor should firmly press against the wood, (pressing down on the wood as the pressure rollers do should cause the tracking sensor to compress a bit.) Also make sure that the O-ring and/or brass roller is not touching the feed belt.

[Digitalwoodshop]

Well after reading this whole thread I believe I came upon the real reason for the errors.....

It came to me after a second person said they got similar results and gave up....

What does that tell you..... It's a .MPC problem.... Or not so much a problem but a result of the .mpc.....

I believe that some of the elements in the .mpc have YELLOW RESTRAINTS on the position. This SLAVES the position of the Element to "THE EDGE OF THE BOARD" or "CENTER in LENGTH.....".

I believe if everyone removes the RESTRAINTS from the .mpc then the project will cut properly...... Right Click on each yellow and remove the restraint.....

Yes, I could be wrong.... But I could be Right.... Let me know.... I have not seen the .mpc...

AL

[chebytrk]

Wow....... just came back in and checked the Forum. However, I've already started my project again. This time I cut a solid piece 28 1/2 inches. Ran three strips of masking tape (center and each edge) on the back and it's going........ Hope it works this time. I did notice what appeared to be one of my board rollers (on the flip out shelf) was a little high on the right side(pad side). I noticed this when the board was measuring and it kinda lifted up a little when it went on that roller. Not alot, but maybe enought to throw things off? Anyway, I re-leveled and am running the project thru once again. I don't have it cutting out the project sides, but only outlining where the cut would be. Hopefully that will be enought to tell me if it centered everything correctly. If it did, then I'll just finish cutting it out with a jig saw or band saw. Sure hope it works this time. I'll report back in a couple of hours and post my results........ Hope the Force is with me this time o' great Obie Wood Kanobie ! : )

[mtylerfl]

Well after reading this whole thread I believe I came upon the real reason for the errors.....It came to me after a second person said they got similar results and gave up....What does that tell you..... It's a .MPC problem.... Or not so much a problem but a result of the .mpc.....I believe that some of the elements in the .mpc have YELLOW RESTRAINTS on the position. This SLAVES the position of the Element to "THE EDGE OF THE BOARD" or "CENTER in LENGTH.....".I believe if everyone removes the RESTRAINTS from the .mpc then the project will cut properly...... Right Click on each yellow and remove the restraint.....Yes, I could be wrong.... But I could be Right.... Let me know.... I have not seen the .mpc...

AL

Hi AL,

Good guess, but not the situation in this particular case. I guarantee it is not an MPC problem. There are no restraint/attachments at all in the MPC (that is intentional, and by design!). Pretty clear it's an x-axis error - they just have to do a little investigative work to determine the cause.

[PCW]

Al

I thought you had it but when I looked at the mpc I don't see any attachments.

[mtylerfl]

...Hope it works this time. I did notice what appeared to be one of my board rollers (on the flip out shelf) was a little high on the right side(pad side). I noticed this when the board was measuring and it kinda lifted up a little when it went on that roller. Not alot, but maybe enought to throw things off? Anyway, I re-leveled and am running the project thru once again...

Hi Jerry,

Good thing you re-leveled the head. Yes indeedy, the head not being level can/will mess up a project! Maybe that was one of a couple other possible things that was causing the tracking roller to lose contact intermittently. Eliminate all the variables and you're on your way to solving the problem. Can't wait to hear how it comes out this time! Should be "spot-on" unless there is some remaining issue with the tracking you/we haven't caught yet.

[mtylerfl]

Al

I thought you had it but when I looked at the mpc I don't see any attachments.

Guaranteed, there is no problem with the MPC whatsoever.

[PCW]

Guaranteed, there is no problem with the MPC whatsoever.

I agree with you 100%.;)

[chebytrk]

I'm back to report on my latest...... drum roll............ SUCCESS! Everything seems to have centered and lined up perfectly. Well... except for 1 small thing. I have it marked on pic # 3. It wasn't bad, but it is something. However, what does still bother me is that I can't find ANY brass roller tracking teeth marks ANYWHERE along the back side of the board on the masking tape. I've checked and rechecked. Even used a magnifying glass and don't see ANY teeth marks. Now I know I've seen them before on previous projects, but since I haven't had any problems I haven't been looking for teeth maks. So... that part really has me stumped.
What I did this morning was I took the bottom plate of the CW and blew everything out. I then blew out everything on the inside of the CW (rollers and all). I then "re set" the board leveling rollers to make sure they were level (one was off a little). I didn't pull out the Red Handle and then just ran the project. I placed 3 strips of tape on the back (center and each side). Again, what bothers me now is not seeing ANY teeth marks along the length of the tape.

I tried to check the rubber damper that was mentioned earlier, but I couldn't find a way to see it. Anyone know how I can do that?

[mtylerfl]

I'm back to report on my latest...... drum roll............ SUCCESS! Everything seems to have centered and lined up perfectly. Well... except for 1 small thing. I have it marked on pic # 3. It wasn't bad, but it is something. However, what does still bother me is that I can't find ANY brass roller tracking teeth marks ANYWHERE along the back side of the board on the masking tape. I've checked and rechecked. Even used a magnifying glass and don't see ANY teeth marks. Now I know I've seen them before on previous projects, but since I haven't had any problems I haven't been looking for teeth maks. So... that part really has me stumped.
What I did this morning was I took the bottom plate of the CW and blew everything out. I then blew out everything on the inside of the CW (rollers and all). I then "re set" the board leveling rollers to make sure they were level (one was off a little). I didn't pull out the Red Handle and then just ran the project. I placed 3 strips of tape on the back (center and each side). Again, what bothers me now is not seeing ANY teeth marks along the length of the tape.

I tried to check the rubber damper that was mentioned earlier, but I couldn't find a way to see it. Anyone know how I can do that?

Well, Praise the Lord! Glad it finally worked for you. I'm puzzled how the little "glitch" occurred at the end of that one side (it's the first photo as posted), but again, it's somehow related to tracking. At least it's not a major problem you can't sand out. Ya got lucky!

I think I may know why the "teeth" marks aren't showing up - the masking tape may be acting as sort of a resilient pad - when the tracking teeth make their mark, the tape "bounces" back as if nothing happened, therefore "masking" the teeth marks! I have never used masking tape (ever) on any of my projects and all my boards show the teeth marks very clearly. (NOTE: I DO recommend the use of masking tape for anyone using "slick" materials! The folks at LHR make most of their sleds out of MDF and use masking tape under them, for example.)

As far as inspecting the damper under the Brass Roller assembly, download and read this PDF:
http://www.carvewright.com/2010CWweb/service/Service_instructions/CarveWright_service_replace_board_tracking_sensor. pdf
It describes how to get to the damper, along with illustrations.

[Kenm810]

I remembered taking a photo of the Brass Roller tracks,
on the bottom of one of my Sleds, the Pressure Rollers were at a perfect 85Lb.

Took me a while to find the photo --- 6-18-09

[mtylerfl]

I remembered taking a photo of the Brass Roller tracks,
on the bottom of one of my Sleds, the Pressure Rollers were at a perfect 85Lb.

Took me a while to find the photo --- 6-18-09

Thanks for the photo, Ken! I see the tooth marks clearly. I wonder if Jerry's head pressure is too low? Maybe multiple thicknesses of tape "pad" it too much to see the teeth marks? Don't know.

In the photo, I see a slight "wrinkle" in the tape - wondering if the tape is not pressed down enough if there is a potential for the Brass Roller to skip a beat if the tape isn't very securely pressed down. Again, I don't know.

[Digitalwoodshop]

WELL...... Looks like I was WRONG Again.... Well.. I gave it a try.... Sorry to step on some toes.... Ouch.... I was thinking customer added restraints.... so off that soap box....


So we are back to the Brass roller.... Could be a missing rubber bumper.... or a head that is not level on the brass roller side..... Is the carving level?


Getting no Roller Bite marks...... One more thing..... A ROLLED UNDER BELT.... I had to rip off a chunk today before I ran a project... I hear the "THUMP" in the machine when it slipped over.... The board must not have been perfect..... A Rolled Up belt will cause the board to be raised up and have less contact with the brass roller.....

The only one that rolls up for me is the forward one under the cut motor.... toward the Brass Roller... Is the belt missing from the far side....? It is almost impossible to see the roller over belt....


AL

[mtylerfl]

Getting no Roller Bite marks...... One more thing..... A ROLLED UNDER BELT.... I had to rip off a chunk today before I ran a project... I hear the "THUMP" in the machine when it slipped over.... The board must not have been perfect..... A Rolled Up belt will cause the board to be raised up and have less contact with the brass roller.....

The only one that rolls up for me is the forward one under the cut motor.... toward the Brass Roller... Is the belt missing from the far side....? It is almost impossible to see the roller over belt....

AL



Good suggestions, AL!

Yes, as AL mentioned, please check for belt roll up (actually the belt rolls UNDER just like AL said, if a severe tracking problem occurs - we just call it a "roll up"). There are a couple closeup photos of a "normal" belt edge and a "rolled" belt edge in the Jan-Feb 2010 Tips & Tricks article I wrote about Maintaining Accurate Tracking. That article also has numerous other items to check and be aware of when troubleshooting tracking problems (most of which have already been mentioned here in this thread, but is more organized in the article). Here is a link to that particular Tips issue:

ISSUE 27 January – February 2010 – Maintaining Accurate Board Tracking (http://www.carvewright.com/downloads/tips/CarveWrightTips&Tricks_Jan_Feb10.pdf)