Raster engraving inconsistent

[Glenwood Morris]

Hi, I have a ULS X2-600, X-belt and bearings are relatively new, but broken in. I am having the following problem illustrated in the picture below. The petals on the flower I am engraving are all of similar line thickness, however the petals to the right (mostly horizontal in line orientation) are cutting noticeably thinner and cleaner than the petals to the bottom (mostly vertical line orientation). I have played with the tuning settings, but what else should I be looking at to improve the consistency here? The problem looks worse in person than it does in the pictures.

http://www.gemorris.com/eBayPics/engraverplroblem.jpg

[Richard Rumancik]

First of all, make sure that all the petals are drawn exactly the same - i.e. either they are filled shapes with constant thickness outlines or else filled shapes with no outline. (Or are they all just wide lines/no fill?)

Perhaps the problem is how the laser rasters lines that are at various inclines. What happens when you rotate the image 90 degrees and raster? If the effect rotates as well, the problem is in the image; otherwise it is another effect.

I would suggest that you draw a "test image" because it may be easier to see what is happening than with the petal shape.

You could draw several vectors say .050" wide (or whatever effective line width you are using for your petals) which run radially from a point like the rays of the sun. You could draw maybe 12 vectors 30 degrees apart like a clock. The whole image could be inside a 2" diameter circle or less for quick tests.

You can then raster this image at various settings and see what you get.

I am not positive, but I had the impression that the ULS lasers allow you to use unequal burns-per-inch in the x and y directions. My laser, and many others, use the term "dpi" for both the x and y axes and they must be equal. But I believe the ULS lets you have unequal settings in each axis (x=ppi, y=dpi). The ULS owners can correct me if I am wrong here.

What I am getting at is that perhaps the problem is caused by the ppi and dpi settings you have chosen. I assume that to avoid a "directional" effect you should keep the "x" and "y" burns-per-inch to be the same. At least, that would ensure that the appearance of a "3 o'clock" line will be the same as a "6 o'clock" line. (In between it approximates the best it can.) So as a guess maybe it has to do with the ppi/dpi settings you have been using.

[Mike Mackenzie]

Glenwood,

is this system a dual tube set-up?

If it is un plug the bottom tube and run a test. If this corrects the problem then the two tubes are not aligned together.

[Glenwood Morris]

Glenwood,

is this system a dual tube set-up?

If it is un plug the bottom tube and run a test. If this corrects the problem then the two tubes are not aligned together.

Not a dual tube setup.

I rotated it and cut again, same result, any lines (these are discrete objects that are filled, not strokes) that are vertical in orientation come out wider and sloppier than any lines that are horizontal in orientation.

[Scott Shepherd]

What settings do you have? What's the wattage, the speed/power/DPI/Image Denisty, etc.?

Looks like the horizontal is using a high DPI (500) and the vertical (Image Density) is using more like 2 or 3?

[Glenwood Morris]

50 Watt, 100% power, 75% speed.

1000dpi and 6 image density

[Richard Rumancik]

Previously I suugested that ULS allowed different settings in each axis.

Well, it seems they still do, but from the last few posts I was reminded that they changed the terminology over the years. It seems dpi has gone by the wayside and now the two controls are ppi for x and image density for y. I did a search to see if I could translate image density of 1-6 into dpi but have not found the translation table. But I did find something Scott Shepherd wrote in a different thread.


. . . . you have two settings for rastering, one is PPI and one is Image Density. The PPI controls the spacing between the dots it burns moving in the X-Axis. The Image Density controls the amount of dots in the Y-Axis. You can also select "Image Enhancement" which brings up 4 more fine tuning options. You can control "Density, Contrast, Definition and Tuning". There is a procedure for tuning each piece of material for maximum clarity of the image. It involves burning test patterns and text and then adjusting each of those 4 controls until the image is razor sharp. Once you set that, you can save it so you don’t have to do it again.

Not sure why ULS went to this system. But in any event I still think you need approx. equal burns-per-inch in each axis for consistency of line width in each axis, regardless of what you call it. 1000 and 6 is probably in the ballpark. Glenwood, have you tried tweaking the controls Scott refers to above?

BTW – Scott and Glenwood – I thought I just about had it figured out the ULS terminology, and now I see you are talking about dpi settings again . . . did you mean ppi?

[Roy Brewer]

however the petals to the right (mostly horizontal in line orientation) are cutting noticeably thinner and cleaner than the petals to the bottom (mostly vertical line orientation).Glenwood,
No laser tube puts out a perfectly round dot; that is, there will always be some elipticity(sp? or even a real word?). To my understanding, this is somewhat the "nature of the beast." Looks to me like your "ellipse" is a bit extreme is completely in the horizontal axis.

A close analogy would be a calligraphic pen.

[Allan Wright]

Could it be a wood grain issue? Instead of rotating the graphics, try a sample with the wood rotated. If it's the wood grain you may have to adjust your graphics accordingly or get a different wood.

[Scott Shepherd]

Richard, that is correct. I don't know why they use it, but I can tell you that without the ability to vary the two and tweak them, I'd be in deep trouble. For months Epilog told me my problems were material related. Didn't matter what material I used, I had the same problem, but they kept telling me it was a material problem. I took the files and material to a ULS rep, and with the ability to tune these things, he had me an example that was superior to anything I had ever been able to do with the Epilog. For me, in my business, if I didn't have the ability to tweak it like I do now, I'd have lost 90% of my business due to quality issues of the finished product.

I don't know why they do it, but I'm really glad they do.

[Richard Rumancik]

. . . I rotated it and cut again, same result, any lines (these are discrete objects that are filled, not strokes) that are vertical in orientation come out wider and sloppier than any lines that are horizontal in orientation.

Okay, it is not in the graphic itself. Since rotating the graphic and lasering does not rotate the "bad" lines it means the problem is not in the graphic.

You could try converting it to a bitmap at 1000 dpi, then plot as you did before and see if it has an effect. Just trying to find a possible work-around in case you can't isolate the problem.

Is it possible the the x-belt is too loose? If there is slop (overshoot) in the x axis it would be more noticable when rastering a vertical line and would tend to be be less obvious when rastering a horizontal line.

[Mike Mackenzie]

Glen wood,

How old is the system? Have you inspected the optics real good? Have you checked the alignment?

Where is your tuning set to in the driver?

Do you have any other Len's Hpdfo?

Try to move the PC from the 0,0 position to the 0,32 (right side of the table) and try it.

If this corrects the problem you may have a slightly worn X belt or some debris in the drive gear that is causing the belt to slip.

You may have to send some samples to the factory for evaluation contact customer service and send them this picture and also send them the file and pc of material that way they can test to see where the problem is coming from.

[Bill Cunningham]

1000 dpi sounds a little excessive for wood! Cut back to about 300-600 and if you need it deeper, then slow down the machine..

[Richard Rumancik]

Along with checking belt tension, also check that the toothed pulley on the x-motor is secure. I assume there is a setscrew holding it. If there was some slippage due to a loose screw there could be some overshoot in the x axis.