The Laser- Rubiks Cube for adults

[Scott Shepherd]

I've had quite a few conversations with others on this forum about this topic,so I thought I'd bring it up for public discussion.

I believe the laser is nothing more than a Rubik's Cube for adults :) Anyone else feel this way? When you do one thing, it causes something else to happen somewhere else? Something that works for someone else, doesn't work for you?

I'm mainly talking about power/speed/material situations. I recall opening the book with our Epilog and going by the recommended speed/power settings for that material. I didn't know any better, so that's what I did. The book would say 80% power, 100% speed, and that's what I used.

Now, about 2 years after entering the laser world, I'm just as confused now as I ever was before. There are some jobs that I get to run every so often, so I get a little more time to tweak things. What I have found is that my major fault is that I am too prone to using WAYYYY too much power. In the case above, 80% power is the recommendation, however, I'm breaking the cap sheet with 20% power. WOW....that's a big difference (Also a HUGE difference in the clarity and sharpness of letters).

I don't recall the actual numbers, but I was doing some Alumamark about 6 weeks ago and I used the "Alumamark" setting from my materials database that's built into the ULS. Didn't come out worth a plugged nickle. I went to the manual side, entered the spec's directly from the people who make Alumamark. I think they had it at 80-90% power, 45-50 speed. Did that. More waste. Ended up with 100% speed, 18% power.

Huh? They recommend 80, I'm at 18? Do these people work for our goverment in the accounting department?

I've seen post after post where people say "oh, I cut that all the time on my 45W and I use 20% power, 100% speed", and I try it and end up at 40% power, 100% speed. There really seems to be no constant even on same wattage machines.

I really thought after doing this for almost 2 years, I'd have a pretty good handle on this thing, but to be honest, I feel like every day is the beginning of this process. Just one more turn of the Rubik's Cube, only to reveal that one cube on the top that just changed to the wrong color.

It really is far more perplexing than I thought it would be. Oh, and I didn't even mention the DPI, PPI, Image Density, Image Enhancement, PhotoGrav, thingy......

Anyone else experience this same sort of thing?

[Margaret Turco]

I go through the same thing! I usually have to use a lower power setting than recommended, and I save every setting that works, as well as write it down for a soon to be created xcel file. At this point, I've been cutting a little test square in an unused corner just to make sure everything is going to work as expected. I am still just experimenting with photos, and that's a whole new learning curve for me.

[Bruce Volden]

Scott,

Some of it's a Rubik's cube at times but I think after a little while longer it'll all
fit in place.

Where I really can get lost/imaginative is mixing and matching various media and designs for inlay work!!!! ie; Some woods work well together, MOP looks best in darker stuff, pauau (sp) looks best in lighter wood.
Turquoise I like in most things, malachite in lighter. Polished agate slabs engraved AND inlaid look great in most things.........

Don't worry, time will come when you can trust your instinct on S/P settings.
I just don't think there will ever be an end to all we can do and have yet to attempt??!!

If I remember correctly I even seen an Einstein on toilet paper, here, on this site......or am I wrong Joe;)

Bruce

[Doug Griffith]

Me too. I think it's more magic than science.

I have color mapping files set up with different speed/power/frequency settings. On any new materials I encounter, first I usually run the map with a high power and variable speed on the edge of the material. Then I take the setting with the best results and run a corresponding frequency map on another edge. This gets me close in 2 quick runs. Sometimes I have to do it the other way around.

Cheers,
Doug

[Mike Null]

Steve

Your observations are dead on. I can tell from comments here that too many are using way to much power to cut as well.

Anodized aluminum and laminated plastic are two materials where accurate settings produce superior results.

[Martin Boekers]

To me it's more like controlling a car on a sheet of ice, there's no way to keep it going in a straight line. They best you can hope for is to be somewhat in control of your slide.

The other thing to remember is that we buy things that are made for the awards industry not the space insdustry, the latitude is pretty wide on what a factory considers it's tolerences. I would wager to say if you asked a daily supplier what their tolerences on a piece of wood, acrylic or metal, they wouldn't have an answer.

Most of the product comes from overseas (could there be some lead in the painted coatings that interfere with the laser?)

A batch of products a supplier buys one week may not be made by the same factory for the next batch.

As for organic items such as wood, the only time I have seen a plaque made from one piece of wood and not pieced together is one that we made in our shop, Look at the differences in one plaque between the panels that are glued together and the difference in burn. Even in plaques I had "sand carved" there was a huge ridge left from the "glue line" I've had customers that want to go through my plaques to pick the "best ones" (that's opening a whole new can of worms!) If the environment that trees are grown in isn't consistent how can the final product be?

Do any of us calibrate the laser or clean lenses everyday? In the photo field I ran pre-exposed "control strips" multiple times a day to check the status of my chemicals. It would be interesting to see a printout of a laser's consistency and performance over a half hour burning session. The laser manufactures talk of speed and power has anyone ever seen a performance chart?

Yes some material are more forgiving than others (Alumark I have never been able to get to work at a level I feel I could routinely achieve)

On the other end with more lasers in the marketplace newer products are coming up and maybe being test marketed in our world. Ex. recently I received an email from a company that makes a solution that with a laser you can mark ceramic and metals. Over a year ago I bought some and couldn't get it to work. I did all the rep asked and still it didn't work (consistenly)
Here comes the email a few weeks ago expressing they made some mistakes and now their new formula and instructions are so much better and to prove it the will give me 20% off!
I emailed them back saying I still had a jar that didn't work and maybe they ought to send a sample. Never heard back from them, shows their sincerity.

I'm sure in the last few years companies like Rowmark and IPI have changed their formulations for their plastic and here some of us still rely on manufacture's guide sheet that maybe 10 years old! I use Epilog (I love them and their support) but even they for my 75 watt say to vector cut .125 alder at freq 1000 does anyone here cut at that many pulses?

Maybe we could have a reference graph posted here for say 20, 40 75, 120 watt lasers for basic materials it may help.

I don't sweat it anymore when a client brings in a bottle of fine scotch to be engraved, because :rolleyes:well, sometimes mistakes happen!:o

Whew!! I haven't typed this much since college.

Just some thoughts,

Marty

[Dave Johnson29]

I'm mainly talking about power/speed/material situations.


Scott,
Having been involved with lasers for many years, just not in the engraving industry, part of the problem may come from the power-rating of the laser.

Most if not all companies test each tube to see that it meets the minimum rated output. I have made many laser output testing devices over the years and it is not uncommon to find a rated 25W actually putting out as much 40+W. Usually they are much closer than that at around 30W-35W but but rarely (never?) have I ever sen a laser rated at 25W measure at 25 or 26. Usually they are well past the rated output.

Also the laser will lose efficiency with use so a 35W today may in 2 years time, output say 28W. Hypothetical figures of course.

It is a little like, "how long is a piece of string."

This however would not explain a 80% versus 18% difference with the Alumimark issue. Maybe the Alumimark people tested on a 10W and then just calculated using ratio and proportion to give figures for other Wattage lasers.

[Scott Shepherd]

Another example would be the settings Photograv kicks out. On simple acrylic, it would commonly say that you need 70-80% power. On clear acrylic? Huh?

I do that and I'm blowing out the image so bad it's not funny. I'm less than 20" on extruded acrylic with Photograv. That's their setting, so I begin to wonder where they come up with things and I'd like to see examples.

[Richard Rumancik]

Unfortunately the settings for rastering and laser cutting are not very transferrable between different lasers. Some of the reasons have been mentioned in this thread as well as a similar one running concurrently.

I did quite a bit of analysis of my LaserPro some time back and found that even though the spec'd speed of the laser was 42 ips, this only applied to rastering. The manufacturer does not state maximum vector speed. I believe for my machine it is around 10 to 15 ips. (But it also depends on the shape of the object.) When I set the vector speed to anything above 30%, it made no difference to the actual travel speed of the carriage. So if I set the vector parameters to Speed 30% Power 100% or Speed 100% Power 100% I would get the same results, because it won't vector at 42ips. This would be very misleading information. The point is, when you have algorithms like this built into the driver/firmware that aren't obvious, there is just no way to compare settings between two brands of lasers.

The best you can do is set parameters for one model of laser, but as had been mentioned, with laser power varying considerably over the life of a tube, even this is prone to error.

Another factor is that the lens and focus will have a very significant effect on the engraving and rastering result. If you are achieving a .004" beam diameter vs a .003" beam diameter, you will experience a power density of about 56% of the .003" beam. The total energy delivered will be the same, but the watts per square meter will be quite different. Clearly this will have a great effect on (a) the speed that you can cut and (b) the appearance of a raster engraving.

The beam diameter will change if you are engraving out-of-focus (whether on purpose, or by accident) and it will change depending on which lens you are using. If you use a 2" or 2.5" focal-length lens you will have a larger beam diameter than for a 1.5" lens. Yet hardly anyone (including the manufacturers) specifies which lens they are using when specifying parameters. And as Scott mentioned in the original post, there are even more parameters (eg. ppi aka frequency) that are also a factor.

I extracted some info from the Rowmark site. If you compare lasers in this table you will see the fallacy in attempting to transfer settings between two different machines. Rowmark at least made a good attempt at quantifying the settings for a model. Notice for example that vector speed on an Epilog means something much different than on a Universal or GCC.

For a single model of a machine, much more information than just %Speed and %Power is needed to be useful. Some of which, like true wattage, is often not available at all to the laser operator.

Rowmark's data is based on three "25 watt" machines. Were the machines checked as to actual power output when the comparison was made? The point is that even the material manufacturer's recommended settings are subject to the same accuracy problem.

[Doug Griffith]

(But it also depends on the shape of the object.)

On my Epilog Mini, I question whether the shape has anything to do with speed. It shouldn't since the quality of the cut doesn't care what shape it is. I tested this by creating a complicated shape using all bezier curves and another with straight lines that was the exact same path length. The time to run each came out exactly the same.

Another issue that affects speed/power/frequency is the formulation of the substrate. Acrylic from one manufacturer may not react the same as acrylic from another manufacturer. Colors make a difference. extrusion or cast makes a difference. In a nutshell, it all makes a difference.

Cheers,
Doug

[Bruce Trethewey]

~With a photographic background, we try to get a test graph with power and speed graph boxes numbered. one corner will be untouched , the other burnt through but it is a great way to have your reference on the wall for future use as a starting point. Blindly repeating numbers without checking is a mistake.

[Martin Boekers]

Bruce, good thinking! I to have an extensive photographic background. this should be no different than the test strips I did in the "old days".

By doing that you can set the end point parameters and interpolate (guess) from there. Much better than as you say blindly repeating what used to work.

I do find myself reffering back (way back) to my photo experience and problem solving, that was before auto photo everything!


Marty

[Tim Bateson]

Yes Doug, Shape does matter. When first starting I just cut small squares to fine tune the settings. However when I applied those same settings to the same piece of wood only the straight lines cut through.
I now test with a complex graphic that contains various curves and lines.
This works well, but we all know the next sheet of wood may or may not cut cleanly with the previous settings.
I now usually test each sheet up in one or all of the corners.

[Richard Rumancik]

On my Epilog Mini, I question whether the shape has anything to do with speed. It shouldn't since the quality of the cut doesn't care what shape it is.

I agree, Doug, that the speed should not be significantly affected by the shape. But on my LaserPro it does. Now I admit it is not the latest model and I can only speak for my own laser. But cutting a circle vs a large rectangle of the same perimeter showed that the circle took longer to cut. You machine may show different results on this test.

For a shape with corners, it is clear that the laser motion must completely stop on the corner so it decelerates to zero, changes direction, and accelerates again to maximum velocity. If you have a nice smooth curve like a circle or similar smooth closed shape, you might expect that the velocity could be approximately constant. But a shape with lots of sharp corners (cusp nodes) means that the motion is continually going to zero velocity. In addition, if the length of the next line is insufficient to achive maximum velocity before having to put on the brakes for the subsequent corner, you might never achieve maximum velocity anywhere on the curve. So it is really not possible in this type of situation to have constant velocity. Ideally, ramping is supposed to take care of deceleration effects but it depends on how well the algorithm has been implemented. So this is yet another variable that comes into play.

[Doug Griffith]

I ran another test and came up with a negligible difference. The file with the straight lines and corners was actually faster.

I created 2 files:
file 1 consists of straight lines with corners
file 2 consists of only curves (random bezier)

Both files have a path length of 2188.40mm
Both files have the same start and end positions

Straight/corners @ 100s/100p/100f = 28 secs
Straight/corners @ 10s/100p/100f = 4.18 secs

curves @ 100s/100p/100f = 32 secs
curves @ 10s/100p/100f = 4.19 secs

My machine is an Epilog Mini 24.
Driver is vector only.

I attached the test files(eps) in case anyone else wants to see if there is a difference.
http://www.dogcollarlabor.com/smc/DOWNLOADS/speed_test.zip

Cheers,
Doug

[Barb Macdonald]

Yes, it is a rubik's. My machine has been working great. Went to do some reverse coloured/matte clear gravoply, used up a whole sheet to get 4 pieces of 6x12. Machine was freshly cleaned/aligned. Piece was flat. Started to not close contours, banding, I will never buy that material for slightly larger text again.
But the inlay work went great!
I may have gook in my spindle, I'm so looking forward to cleaning that again. The extra exhaust has really improved the build-up issue, but gook still will settle.
Technical, eh?
Happy Hallowe-en, everybody!!
Some morons in our school district have decided to start calling it Black and Orange Day, because the concept of the "holiday" might be offensive. For that, we pay them.

[Martin Reynolds]

I spent some time a while back figuring this one out on my Laserpro.

The laser only runs full speed on straight lines. As soon as it has to compute a curve, it slows waaaay down. Seems to be more to do with computing than motion.

Now, the Laserpro has a mode where it also reduces the laser power in line with the speed of the head. But the mode only seems to deliver about 70% maximum power on the straights, so it is quite slow.

I usually end up setting the power and speed for the right balance on curves and straights, at maximum speed.

Then, some materials vary in thickness. And laser power varies across the bed. But it is that straight/curved line thing that causes the most trouble.