"Jitter" on Curves

[David Rust]

Hi All,

I have been experiencing a new problem with my machine. While vector cutting curves the cuts are not smooth, they appear to have a "jitter" or "ripple". I called Epilog and they have been great as always, they identified some exrtra slop in my x-axis linear guide and sent me a replacement. After installation the problem is still there. Epilog is still looking at the problem.

Has anyone experienced this problem and if so what was the solution?
250138


Thanks,
Dave Rust

[Joe Pelonio]

Has the X rail been cleaned and greased well?

[David Rust]

Hi Joe,

I replaced the whole x-axis rail and carriage with new bearings with a new unit supplied by Epilog. Here is another pic...
250142

Edit: The words were laser cut left to right along the x-axis of the laser (90deg to the shown picture)...

[Mike Lysov]

Do Epilog lasers have something called vector quality in their control panels? It may happen that these settings have been changed somehow and it can cause this jitter problem. I had exactly the same problem with my GCC LaserPro and it was caused by wrong setting in Vector Quality. All straight lines were cut smooth but anything curved had this jitter problem.

[Scott Shepherd]

Can you post your speeds, please. That might help in diagnosing.

[David Rust]

Hi Scott,

My settings are 100/18/750 (s/p/freq). I am cutting paper.
This is a new problem, I have cut many designs out of paper with these settings and have not had the degradation in quality that I now experience.
I will try reducing power and do a comparison for troubleshooting.

[Michael Hunter]

The only time I had anything like as bad as that on my machine, I found that the mirror/lens assembly was slightly loose on the carriage.

Other things to check -
If you took the belt off when you changed the rail and bearings, did you get the ends exactly straight when you put it back on?
Did you work as much grease as possible into the new carriage bearing? (AFAIK they are oiled but NOT greased when new).
Belt tension.
Turning mirror at the end of the beam.

Otherwise, is anything vibrating the laser or the floor it is standing on (compressor, exhaust fan etc.)?

[Glen Monaghan]

Be sure to clean the encoder strip and blow any dust out of the sensor on the gantry, if you haven't done that recently...

[David Rust]

Hi Mike,
I had the problem, then changed the rail and bearings with new, now the problem still exists.
I did not pack the new bearings with grease, thanks for the heads up, I will pack them. Right now they are very well oiled and they run very smoothly (because they are new), my next task will be to pack them with grease.
Belt tension appears to be good, I am letting the spring loaded tensioner do its job before tightening the screws, a finger push against the belt verifies the tension feels about right.
I have not checked the optics... I'll look at the mirrors and lens next.

Thanks!

[Scott Shepherd]

You might try reloading the driver as well. Sometimes that resolves those types of issues.

[Kim Vellore]

When I had this issue Epilog told me not to exceed 25% speed on vector cuts. I think this is a design issue I thought it was only with older machines, you just cant go higher speeds on vector cuts without this vibration. I see it more prominently when I vector cut diagonals, Two of my friends also on this forum have the same problem their only option was lower speeds.

Kim
Epilog 24TT

[George Brown]

I have the same problem on my ULS if I vector cut at too high a speed. Poorly written software/hardware design, but nothing I can do about that.

[David Fairfield]

Kim, that is not the normal sort of jitter I get vectoring curves and diagnoals at any speed on my Epilog. Even at 100 percent speed the jitters are just little tremors, acceptable for most crafts products, but not the sort of fine miniature work you and I do. The only time I experienced a prounounced jitter similar to David Rust's was when my belt needed re-tensioning. Followed the instructions the tech dept sent to me and the problem was solved. I wish Epilog would put all that good troubleshooting info into the manual inthe first place! My guess is something is loose somewhere.

Dave

[Greg Bednar]

David, can you post the .CDR file so I/we can take a look at it?
Hi All, I have been experiencing a new problem with my machine. While vector cutting curves the cuts are not smooth, they appear to have a "jitter" or "ripple". I called Epilog and they have been great as always, they identified some exrtra slop in my x-axis linear guide and sent me a replacement. After installation the problem is still there. Epilog is still looking at the problem. Has anyone experienced this problem and if so what was the solution? 250138 Thanks, Dave Rust

[David Rust]

Hi All,

Here is the .cdr file. It's pretty simple, all hairline curves.

250289

Dave

[Steven Cox]

Here is the .cdr file. It's pretty simple, all hairline curves.


Try saving the file less than ver x6 for us mere mortals who still use versions from the arc!.

Regards Steve

[Greg Bednar]

try this file on a piece of card stock and let me know if there is a difference.

[David Rust]

Greg,
Results are still degraded. Did you make any changes or did you just save as a v12?

[David Rust]

Ran the job at 50 speed, still jitter in the curves
Ran the job at 25 speed, there is still jitter however the "magnitude" of the ripples is much less...

The Helix sits on concrete, therefore there is no noticeable induced vibrations.
The exhaust fan is mounted outside, the exhaust hose at the Helix has little to no vibration from the fan.
The mirrors appear to mounted firm. I took a Q-tip wrapped in lens paper and tried to move them around, I also checked the adjustment screws and the mounting hardware. All appears to be tight and in good condition.
I increased to belt tension just a small amount above what the spring tension is, no difference in performance.

hmmm....

[Greg Bednar]

I reduced the amount of nodes in the file.
Greg, Results are still degraded. Did you make any changes or did you just save as a v12?

[David Fairfield]

Its a long shot, I'm sure you would have noticed, but is your material secure? Because if you are cutting something lightweight like paper, you can get some rippled cuts that look like this when your exhaust blower develops enough suction through the cut line to vibrate the paper like the reed in a kazoo. It can actually make a kazoo-like noise, and you will see a wavy pattern in the material. I reckon the air assist can potentially cause a similar problem. Dave

[Scott Shepherd]

Ran the job at 50 speed, still jitter in the curves
Ran the job at 25 speed, there is still jitter however the "magnitude" of the ripples is much less...



Just for the record, I have never seen a laser that cuts clean, smooth lines at 25% speed. Our Epilog wouldn't do it, our Universal won't do it, and our Trotec won't do it. I think you need to get a lot more realistic with your speed. It's been so long since I have run an Epilog, I don't recall good settings for it, but try a speed of 10-15 and see what you get.

I have run a lot of tests over the years on this very thing and it's always been there, in all 3 major manufacturers, for years and years. High speed vector cutting only works great on straight lines.

Try it at 10-15 and see if that solves the problem. I don't think you have any mechanical issues, I think you're just trying to cut way faster than the machine is made to cut.

[Greg Bednar]

FWIW - when cutting vectors, make sure the resolution is set to 1200 DPI. Do a search here and you will find that setting the resolution at this level yields superior results than at say, 300 and lower.

[Mark Ross]

We had this problem on our Epilog 36EXT. Your pulleys have multiple set screws that keep them tight against the rods. The rods have a flat spot that interface with the pulleys. When we had this problem the set screws were loose. Not enough that we could feel slop when we moved the gantry by hand but it was enough that when the machine ran we would see it. Our solution was to re-tighten all the set screws and loctite them. Use a serviceable loctite though in case you ever have to service the machine.

[David Fairfield]

David- were you able to determine a cause and get this resolved?

Dave

[Jeff Greer]

I have had my machine do it at times too. I believe it is more to do with the file then your machine. While toolmaking, we would have certain programs that would be generated with a massive amount of points vs. a program with curves in it and the CNC 's would do the same thing - jitter on a radius cut. I think it has to do with the way the program is made from a design to moves (posting).

[David Rust]

David- were you able to determine a cause and get this resolved?

Dave


Hi Dave, I was not able to get to the bottom of the the problem. I sent Epilog an example file with settings and they said that they also experienced "ripples". My confusion lies in the fact that I have two cut outs, on recent with ripples and one from a few months back without ripples. It is possible that I had different settings however I saved the setting configuration settings in the epilog print interface. I haven't dug back into it again since talking with them, once I get some time I'll do some more tests and get back to everyone.

[David Rust]

I think I got to the bottom of the Jitter problem...

Problem Recap:
- Jitter on curved cut edge while vector cutting paper, new problem, I have previously cut the same Corel file without the jitter. No issues on straight edges parallel or diagonal to the x-axis
Troubleshooting:
- re-aligned
- re-leveled table
- cleaned all lenses/mirrors
- cleaned x-axis encoder strip
- replaced x-axis motor (for different issue)
- verified file had round hairline circles in tst file
- cut circles at different speeds
- cut circles in different areas of the table
- sent files and pics to Epilog
- conversed with them over the course of a month
- Epilog said they ran the same settings and got jitter also, so it must be a limitation of the machine
- replaced the x-axis bearings (under warranty)
- varied tension on the x-axis belt
- cut circles at 25 speed and still had jitter issues
None of this fixed the problem...

Discussion:
I decided to do more tests and change other settings... I discovered that the settings for "vector only"cutting are not just, Speed(0-100), Power(0-100), and Frequency(0-5000 pulses per inch).
DPI(75 - 1200 dots per inch) also plays into the equation... Once I discovered this I searched the manual and could not find where it mentioned that DPI was a factor in vector cutting. However when choosing a "vector only" cut, DPI is not grayed out. When I increase DPI over 600 all Jitter goes away....

This is what I think is going on (may not be the case, just my observations):
The DPI is a setting that affects x/y-axis movement and the Frequency is a tube setting that affects the firing of the laser. Two separate settings that both are used for vector only cutting. The only confusion I have now is that the Epilog manual states that Frequency is "pulses per inch"... this is why I originally didn't think that DPI was a factor. Is Frequency really supposed to be"pulses per second" as opposed to "pulses per inch"? Another thought... maybe frequency for the tube is consistent with the freq settings and the laser output is gated by the DPI setting....

If frequency is really "pulses per inch" as the manual indicates then speed should have an effect on frequency...

Resolution:
- Set DPI to >=600
- 100s/18p/500freq/DPI600 cuts paper at high speed without jitter

[Henri Sallinen]

I know this is a OLD thread, but I'd like to say THANK YOU to David Rust! This really worked charms to our cut quality!

I can't believe that this worked! 1000x Thank you!