laser pausing in the middle of a straight vector cut

[Gene Jameson]

hi all. been a lurker for a while but now have a question maybe some one can answer. I have 2 Epilog machines, and i notice that they both pause when cutting some straight vector segments. there is no node at that point in the geometry of the shape and the epilog tech offered a less than satisfying reason. This particular cut is a straight line and it is exactly on the x axis. The tech at epilog said that the shape "must" be slightly angled and the encoders cannot respond in such minute increments so it pauses for a pulse then resumes. Picture your laser cutting out a square in 3mm baltic birch. In the middle of one side it pauses for a beat, then continues. This happens even when there are no nodes in that span. I also don't remember this always happening but can't say for sure that it is related to a firmware update. Has anyone else experienced this. Thanks.
Gene

[Kim Vellore]

Gene, if the line is angled it will not have a node. for the square you are having a problem with did you draw it with 4 lines or did you use a square. This could be a quick check to test, just draw a couple of lines with slightly varying angles and see if it pauses when printed.

Kim

[Gene Jameson]

Kim, the square i described was just an example by me to try to over-simplify what i was trying to describe. The shapes i am cutting have a radius at the top two angled sides and a straight line at the bottom connecting the two sides. It looks like a trapezoid but it has a concave top. I cut the same shape on 3 lasers (2 epilog, and 1 universal). one epilog laser stops in the middle of the straight cut at the bottom. the other 2 lasers do not hesitate. The outline i am cutting is one component (a complete closed shape) not 4 lines. I am can use the Pen tool in corel and "trace" the shape, and the new shape cuts without pausing. So i guess i can just re-create the shapes that have this problem, but it will be a hassle because we cut a lot of different shapes, and i don't want to have to redo them all when this problem seems to have started recently. I really just want to understand what in the geometry is causing this and how i can fix it without redrawing the piece.

[John Blazy]

I have had this mysterious "skip" in the cut twice already since buying my rabbit in January. It has only happened at the beginning of a job, in the morning when I started the laser up, and started cutting about ten minutes later. Once warmed up, I have never had a skip in the cut.

So I just assume that maybe the laser has to warm up a bit more, and has to do with expansion/contraction somewhere in the wiring or the tube itself.

I even wondered if it might be a dust particle shaking loose and falling across the beam path somehow.

[Kev Williams]

the shape "must" be slightly angled and the encoders cannot respond in such minute increments so it pauses for a pulse

Epilog told you that? Sheesh....

All any X-Y axis machine does is follow a dot-to-dot plot file, moving from one XY coordinate to the next. The only pause comes from 'tool start' and 'tool stop' commands so the tool can move to the next shape. When coordinates are close together, the machine will run slower than when they're farther apart. But no 'pausing'...

When drawing a square, there's only 5 X-Y coordinates- corner 1 laser-ON, corner 2, corner 3, corner 4, corner 1 laser-OFF. The steppers have absolutely NO reason to stop between ANY 2 coordinates.

I just cleaned the encoder disk on the back of the X stepper on my Explorer, as I had engraving that was shifting. It was plenty dirty, and works fantastic now. Your pause may be a speck of dirt or whatever on your X encoder....

[Gene Jameson]

With regard to the epilog not lasering immediately when the "go" button is pressed, that has always been an annoyance for both of my Epilog lasers but I live with it. If you don't know what I am talking about, on both of my epilogs, when I press "go" first thing in the morning the carriage starts to move as it should but the laser does not fire immediately. It varies how long this period is. I simply stop the job when the laser starts then restart the job. After that it starts fine. But this is not the issue I am talking about. I have uploaded a short video to youtube (this was done spur of the moment and I normally cut on the vector cutting grid) . During the last segment at the botom, right in the center, the laser head stops for a split second then continues. This stop/start shows up in the cut edge of the piece. it is not terrible but it did not used to happen. When I first got my laser and was still learning, I realized some of my parts had way too many nodes and this was causing a slower than necessary cut, but there is no node along this span. Thanks!
Oh, and the laser (and encoder strip) is clean. I just cleaned it top to bottom while waiting on my laser tube replacement. I do not believe this is related to the new tube, either. it seems to me something in the print instructions sent to the machine

gene

[Kev Williams]

The laser not immediately firing, my LS900 has been doing this for about 3 years now (and it's 11 years old)- in my case it's because the tube is getting old and I have my 'tickle' turned off. Tickle fires the laser occasionally to keep it warmed up.

Mine only does it when first started up in the morning. I always run a quick dry-run just to get the tube firing nicely.

As for your video and the pause in the middle of the line-- Have you checked to see if that line is dead-to-rights parallel to the X axis? It doesn't matter if there's a node in the line in Corel; if the far left Y coordinate is (f'rinstance) 1.2015" and the far right Y coordinate is 1.2010", then the Y stepper will adjust for that .0005", right in the middle (just about where yours did), and when it does, it will slow down briefly to make the change....

[Keith Winter]

So if this was a Trotec, it would appear like the job transmission from the computer isn't able to transfer fast enough to keep up with the laser. Is this Epilog computer controlled in the aspect that the computer always has to be on like a Trotec, or does it have the processor on the machine itself and can run without a computer?

With regard to the epilog not lasering immediately when the "go" button is pressed, that has always been an annoyance for both of my Epilog lasers but I live with it. If you don't know what I am talking about, on both of my epilogs, when I press "go" first thing in the morning the carriage starts to move as it should but the laser does not fire immediately. It varies how long this period is. I simply stop the job when the laser starts then restart the job. After that it starts fine. But this is not the issue I am talking about. I have uploaded a short video to youtube (this was done spur of the moment and I normally cut on the vector cutting grid) . During the last segment at the botom, right in the center, the laser head stops for a split second then continues. This stop/start shows up in the cut edge of the piece. it is not terrible but it did not used to happen. When I first got my laser and was still learning, I realized some of my parts had way too many nodes and this was causing a slower than necessary cut, but there is no node along this span. Thanks!
Oh, and the laser (and encoder strip) is clean. I just cleaned it top to bottom while waiting on my laser tube replacement. I do not believe this is related to the new tube, either. it seems to me something in the print instructions sent to the machine

gene

[Rich Harman]

... if the far left Y coordinate is (f'rinstance) 1.2015" and the far right Y coordinate is 1.2010", then the Y stepper will adjust for that .0005", right in the middle (just about where yours did), and when it does, it will slow down briefly to make the change....

I can't think of a reason why a slowdown would be necessary just because the change in Y axis is so slight that the Y stepper only needs to move one step during the line. What if two steps were required? would you expect a slowdown at 1/3 and 2/3 of the way through? How about three steps, or four etc...

[Lee DeRaud]

So if this was a Trotec, it would appear like the job transmission from the computer isn't able to transfer fast enough to keep up with the laser. Is this Epilog computer controlled in the aspect that the computer always has to be on like a Trotec, or does it have the processor on the machine itself and can run without a computer?

There's still a processor in the laser, doing the realtime control of the steppers/servos: there's no way Windows/USB can handle that, just too much latency. The only question is how much of the job gets sent to the onboard processor at a time.

[Kev Williams]

A lot of raster data can really eat up the megabytes, but vector data is minimal...

In Gravostyle I drew up the same basic shape as is in the video, at roughly the same size and placement
Note that my upper left corner coordinates are X=0" and Y=19.5", as my machine's zero is the bottom left corner...



--then I copied the actual plot file that would cut this, which is below. The first half of the file cuts the shape,
the second half cuts the circle. The shape starts its cut at the bottom left corner. Just read the coordinates in thousandths of an inch,
so the start point is .881" X, 18.468" Y, top left corner is 1.320-X / 19.453-Y, then the machine plays connect-a-dot, quite a bit for the curve. The last 4 coordinates
before the circle are the top right corner, 3.002-X / 19.453-Y, then to the bottom right corner, 3.409-X / 18.468-Y, then
back to the starting point, where the PU (pen up) command comes in, then it moves to the circle. You'll notice there's a lot of coordinates just to make that
small circle, which is only .061" diameter!

And all this data adds up to a whole 957 bytes. Not megabytes, or kilobytes, just BYTES! (and that includes the 'begin/end file' text!)
An Apple IIe can move this much data in 1/10th of a second, so transfer speeds aren't an issue here!

BUT-- if the bottom left Y coordinate was 18.469 instead of 18.468 like the bottom right, then the stepper would have to move one click up on the
way to the bottom left corner, and it WOULD pause briefly when it does! -- this is what I think is happening, the bottom line isn't totally level--


----------- Begin file ---------------
IN;PA;
!PZ 1,0,55,6;
VZ 94;
!DR 0;
PU881,18468;
PD;
PD1320,19453;
PD1397,19422;
PD1474,19394;
PD1553,19370;
PD1632,19348;
PD1712,19329;
PD1793,19313;
PD1874,19301;
PD1956,19291;
PD2038,19285;
PD2120,19282;
PD2202,19282;
PD2284,19285;
PD2366,19291;
PD2448,19301;
PD2529,19313;
PD2610,19329;
PD2690,19348;
PD2770,19370;
PD2848,19394;
PD2925,19422;
PD3002,19453;
PD3409,18468;
PD881,18468;
PU;
PU2136,18813;
PD;
PD2126,18814;
PD2118,18819;
PD2111,18826;
PD2107,18834;
PD2105,18844;
PD2107,18853;
PD2111,18862;
PD2118,18869;
PD2126,18873;
PD2136,18875;
PD2146,18873;
PD2154,18869;
PD2161,18862;
PD2165,18853;
PD2167,18844;
PD2165,18834;
PD2161,18826;
PD2154,18819;
PD2146,18814;
PD2136,18813;
PU;
SP;
----------- End file -------------------------

[Lee DeRaud]

I can't think of a reason why a slowdown would be necessary just because the change in Y axis is so slight that the Y stepper only needs to move one step during the line. What if two steps were required? would you expect a slowdown at 1/3 and 2/3 of the way through? How about three steps, or four etc...

What he said. The obvious worst case scenario is a 45-degree line, pausing at every step.

In any case, step pulses are going to the stepper motors (or a full two-way command/response cycle in the case of a servo) at thousands of times per second. So if one of them takes twice as long, so what? It's not like you can see the difference between 1ms and 2ms.

[Rich Harman]

if the bottom left Y coordinate was 18.469 instead of 18.468 like the bottom right, then the stepper would have to move one click up on the
way to the bottom left corner, and it WOULD pause briefly when it does!

Why do you think the x axis needs to come to a complete stop for the y axis to move one step??? These machines are quite capable of moving two axis simultaneously.

[Keith Winter]

There's still a processor in the laser, doing the realtime control of the steppers/servos: there's no way Windows/USB can handle that, just too much latency. The only question is how much of the job gets sent to the onboard processor at a time.

I believe you but Trotec has repeatedly told me there is no memory on Trotecs and the only processors are the circuits on the control board, aka no CPU. I've gone round and round with them over it trying to figure out issues. They all stay firm to that, tech support and sales.

[Keith Winter]

A lot of raster data can really eat up the megabytes, but vector data is minimal...

In Gravostyle I drew up the same basic shape as is in the video, at roughly the same size and placement
Note that my upper left corner coordinates are X=0" and Y=19.5", as my machine's zero is the bottom left corner...



--then I copied the actual plot file that would cut this, which is below. The first half of the file cuts the shape,
the second half cuts the circle. The shape starts its cut at the bottom left corner. Just read the coordinates in thousandths of an inch,
so the start point is .881" X, 18.468" Y, top left corner is 1.320-X / 19.453-Y, then the machine plays connect-a-dot, quite a bit for the curve. The last 4 coordinates
before the circle are the top right corner, 3.002-X / 19.453-Y, then to the bottom right corner, 3.409-X / 18.468-Y, then
back to the starting point, where the PU (pen up) command comes in, then it moves to the circle. You'll notice there's a lot of coordinates just to make that
small circle, which is only .061" diameter!

And all this data adds up to a whole 957 bytes. Not megabytes, or kilobytes, just BYTES! (and that includes the 'begin/end file' text!)
An Apple IIe can move this much data in 1/10th of a second, so transfer speeds aren't an issue here!

BUT-- if the bottom left Y coordinate was 18.469 instead of 18.468 like the bottom right, then the stepper would have to move one click up on the
way to the bottom left corner, and it WOULD pause briefly when it does! -- this is what I think is happening, the bottom line isn't totally level--


----------- Begin file ---------------
IN;PA;
!PZ 1,0,55,6;
VZ 94;
!DR 0;
PU881,18468;
PD;
PD1320,19453;
PD1397,19422;
PD1474,19394;
PD1553,19370;
PD1632,19348;
PD1712,19329;
PD1793,19313;
PD1874,19301;
PD1956,19291;
PD2038,19285;
PD2120,19282;
PD2202,19282;
PD2284,19285;
PD2366,19291;
PD2448,19301;
PD2529,19313;
PD2610,19329;
PD2690,19348;
PD2770,19370;
PD2848,19394;
PD2925,19422;
PD3002,19453;
PD3409,18468;
PD881,18468;
PU;
PU2136,18813;
PD;
PD2126,18814;
PD2118,18819;
PD2111,18826;
PD2107,18834;
PD2105,18844;
PD2107,18853;
PD2111,18862;
PD2118,18869;
PD2126,18873;
PD2136,18875;
PD2146,18873;
PD2154,18869;
PD2161,18862;
PD2165,18853;
PD2167,18844;
PD2165,18834;
PD2161,18826;
PD2154,18819;
PD2146,18814;
PD2136,18813;
PU;
SP;
----------- End file -------------------------

True normally but if something interrupted the file bad memory, processor error, computer bogged down etc even if it's a small file it can cause issues