PDA

View Full Version : First problem with my laser



Walt Langhans
12-02-2013, 1:29 PM
Hi all,

Working on some new stuff, I cut out some test pieces and found that things are not lining up properly. The notches are 1/8" by 1/8".

276187

It's 2 different materials: MDF on the bottom, and heavy cardboard on the top but I've been doing stuff like this all along so I know it's nothing to do with the material.

I've double checked the corell file so that's not the issue.

I cleaned the mirrors and optics, and realigned everything. Now here's where things start to get fun... When I initially aligned machine the spots were never dead on, but everything cut fine and I did NOT have this problem. When I aligned the machine this time (the first time I've done this since the initial time) it was a bit out of alignment and it's now closer to dead on then it was from the initial alignment but it's still not perfect. However since it's aligned better than it was the first time when I didn't have this issue I'm thinking that the alignment is not issue, but I'm open to the possibility.

So to try to figure out where the problem was I cut 1" squares across the entire size of my bed, about 1' by 2'. The sheet is orientated just as it was in the laser bed, note the mark in the lower right corner for orientation.

276189

Measuring the horizontal lines on the right side, the distance remains constant

276190

However, as we stay measuring the horizontal lines and move from the left to the right (towards the center), the squares at the top (@ 9", 10" and 11") of the sheet it's about by a little less than a 1/32". As I continued to move to the left the problem stayed confined to the 9", 10" and 11" marks and the distance increases ever so slightly to just a hair over 1/32"

When I measure the vertical lines, starting in from the top right side it starts to shift around the 9" line. Then when I move down to the bottom of the sheet the problem starts at the 8" line.

276193

Then continuing to measure across the vertical lines moving towards the right side problem continues by around 1/32" and stays pretty consistent weather I measure from the top of the sheet or the bottom of the sheet.

276194

So now I'm thinking it's the belts since I went through a re-aligned everything. I inspected them and they don't look worn, however the tension across them is not consistent. But before I started messing with, I wanted to check to see if I was even going down the right path. I'd hate to start messing with something and it be the wrong thing and I cause more problems.

Regarding the belts:
1/ how much tension should be on them?
2/ does / should the tension be the same on the vertical belts as well as the horizontal belts?
3/ can I over over tighten them?
4/ How do I tell if it's a matter of correct tension or if the belts have stretched and need to be replaced?

If needed I can provide higher quality of the pictures above, run any other tests that might shed some light on the issue or even take video if it's needed.

Thanks!
Walt

Gary Hair
12-02-2013, 2:08 PM
Regarding the belts:
1/ how much tension should be on them?
2/ does / should the tension be the same on the vertical belts as well as the horizontal belts?
3/ can I over over tighten them?
4/ How do I tell if it's a matter of correct tension or if the belts have stretched and need to be replaced?

Walt - I can't shed much light on your issue but I can tell you that tension does make a difference and it is possible to over tighten the belts. I have recently noticed a misalignment between raster and vector and one of the things I did to try and troubleshoot it was to replace the X belts. Being a bit overzealous I tightened them up pretty good - too good! The alignment was way off, much worse than before. I loosened up the belts and the alignment was better, about the same as it was before I replaced the belts. I had lots of work to catch up on so I didn't get to experiment with tension to see if I could determine the best tension but if I get some time I'll see what happens. My guess is that you'll go from misalignment on the too-tight end to backlash problems on the too loose end. Good luck and keep us posted.

Khalid Nazim
12-02-2013, 4:50 PM
Walt,

I have had this problem for quite sometime and I haven't found a way to resolve this. There has been a discussion in the past on the accuracy of the dimension of cut parts across the entire table on a Shenhui and in my experience parts cut a different locations on the bed have slightly different dimensions.

We should all report this to Shenhui to see if they can provide and answer/solution to this problem.

Regards
KN

Walt Langhans
12-02-2013, 4:59 PM
Walt,

I have had this problem for quite sometime and I haven't found a way to resolve this.

Well that's a little disheartening... but it has to be something that is fixable since I use not to have it, hopefully someone has some ideas.

What have you tried that did work? Perhaps that could help narrow things down?

Dave Sheldrake
12-02-2013, 5:28 PM
Incremental errors can be the motors losing steps or belt tension. In this case as the error isn't progressive I'd say belt tension / worn belts Walt. Replacing them isn't expensive though.

Tension wise, hard call, a longer belt with the same tension as a shorter belt will have the same net effect, longer belts have less "apparent" tension due to the load being distributed over a longer distance.

cheers

Dave

walter hofmann
12-03-2013, 5:19 AM
hi all
my experience with belt tension it should be not as tight as a guitar string. but if the belt is worn you will overstretch them with tis.
mostly if you see lots of dust around the pulley its a sign of worn belt.
another reason could be the pulley itself I they are worn.
reetng
waltfl

Rich Harman
12-03-2013, 2:52 PM
I have also run into this problem. Again. The first time it was mostly solved by replacing the belt and not putting too much tension on it. Several months later it is back to the way it was before. There can be as much as a half mm difference is the X dimension of parts cut at different points on the table.

I have ordered some parts to try a different modification, will share the results if it works.

I don't agree that the same tension on different length belts will have the same net effect. Stretch is measured as some amount per unit of length. The longer the belt the larger the error will be due to stretch.

Dave Sheldrake
12-03-2013, 3:21 PM
I don't agree that the same tension on different length belts will have the same net effect. Stretch is measured as some amount per unit of length. The longer the belt the larger the error will be due to stretch.

Most people will check belt tension by deflection when the belt is pressed on or pulled upwards (pinged like a rubber band) as they have no means to measure it with a load meter at the bearings. With the same pull force (tension)on two belts of different lengths the spring in the middle of the longer belt (deflection) will be more noticeable hence even though the load is the same the way commonly used to measure it will give different results.


longer belts have less "apparent" tension due to the load being distributed over a longer distance.

cheers

Dave

Ray Scott
12-03-2013, 6:32 PM
to all, I realize that this is an older thread, but I still wanted to jump in with my solutions. The belt needs to be installed and tensioned. The belt should be snug, not loose, firm fitting... not so tight that it starts to curl. If you are replacing a belt, make sure to get the right thickness, width, length, and profile edge(mxl, xl, 3m)... if you can't find the 15mm wide belts, it is NOT ok to use two smaller belts. Most timing belts are a nice nylon fiber with urethane rubber coatings. I would love to use a Kevlar belts, but it might be impossible to find the right specification belt. After the right tension is found, the laser software needs to be calibrated to ensure a 100mm square design is actually cut at 100mm. That adjustment is normally a tweak in the "Measurement per pulse" or "Pulse Size". Ray Scott

Walt Langhans
12-03-2013, 8:00 PM
Yeah this problem escalated to a bit of panic and I started a new post here:

http://www.sawmillcreek.org/showthread.php?210855-Well-now-I-ve-really-made-a-mess-of-things-and-need-some-help

And in short Ray came in a saved the day and the problem is now solved!

Rich Harman
12-03-2013, 8:44 PM
to all, I realize that this is an older thread,

??? First post was yesterday.

Walt Langhans
12-03-2013, 9:49 PM
??? First post was yesterday.

Yup this here creek flows quick, got to pay attention or get left behind :D