Power and Speed equivalents - is there a formula or chart?

[Mayo Pardo]

In my experimenting on polystyrene I have my power and speed settings relatively close to producing an acceptable result. Not perfect but I don't think I can get it much better.

With my laser being a 30watt model, for this experiment I have the power at 35 and the velocity at 30. Given that the machine has the capacity to go to 100 for both, is there a formula or has anyone published a chart which would show (for example) that increasing both power and speed 2x will produce an equal effect on the material?

My goal is to increase speed but keep the same visual result I've achieved already, but without having to make another 40 tests.

For all I know the factor may be completely different - maybe the power should increase only 25% for every 50% increase in speed, or the opposite...

Does the power setting effect decrease by 1/2 if you increase the velocity by 2x?

Is this a manufacturer and machine specific ratio or is it somewhat universal?

[Dan Hintz]

Sorry, there is no such chart. Substrates react differently to varying amounts of heat... often you can double the power while doubling the speed and get similar results, but that's not guaranteed. Think of it like this... do you get the same well-cooked piece of chicken if you cook one at 250 degrees for an hour and one at 750 degrees for 20 minutes? Probably not. But you can boil water using that method, right? Water boils either way, so the end-result is the same.

[Glen Monaghan]

I don't recall any laser manufacturer ever characterizing their machines' complete speed and power curves but I've seen it repeated often enough that 100% power setting typically doesn't create ten times as much power as 10% power setting (but no idea whether it's more or less), and that 100% speed is not simply ten times 10% speed. Unless your machine has servos for the axis drives, it probably has stepper motors and a common "speed" setting method for steppers is to vary the time between step pulses, such as <step><delay 100 - speed setting><step>. At speed setting 100, there is 0 delay so the motor is stepped at its highest speed (for example, if the step time was 1ms, the step rate would be 1/1ms per step, or 1000 steps per second). At speed setting 90, the delay factor is 100-90=10, while at speed setting 10, it's 100-10=90. If, for example, the delay factor is in ms, 90 speed would produce 1/(1+10)ms per step = 90.9 steps per second, while 10 speed would be 1/(1+90)ms per step or 11 steps per second. With that simple algorithm, steps per second increases exponentially with linear speed setting increases. In this example, 100 speed steps about 11 times faster than 90 speed and 91 times faster than 10 speed. Your machine's actual algorithm will be different, but the idea will be similar.

You could simplistically characterize your machine's speed and power settings by measuring actual head speed at every speed setting, and using a laser power meter to measure incident power at every power setting. However, you still couldn't simply scale (up or down) both head speed and incident power by X and expect the same or even similar results because substrates don't all act the same (especially natural ones like wood). A given substrate will absorb laser power in a particular way, and react differently depending on the power level and, for lack of a better term, dwell time. As a more laser-appropriate substrate than Dan's chicken, consider a wood plaque. Laser it at some specific low power and low speed, and you might find that you achieve a particular depth of engraving and the engraved wood caramelizes or "toasts", producing a nice contrast with the unlasered wood. However, scale up to high power and high speed and you'll likely find that you get less depth of engraving and the wood simply vaporizes with little or no caramelizing or contrast. Even manufactured substrates won't react exactly the same at different points along the speed:/power curve. For example, you may find that you get a clean, nearly flame-polished edge when cutting acrylic at a certain speed:/power setting, but you likely will get a more ragged edge if you double the linear cut speed and increase power accordingly. At the lower speed, the laser power heats a localized area enough to vaporize the kerf, and does so long enough for to somewhat melt and smooth the newly created edges. At the higher speed, the vaporized acrylic is ejected more violently and the heat source doesn't stay around long enough to smooth the resulting ragged edges.

You're pretty much going to have to experiment to get the higher speed settings you want. As described above, you may or may not be able to get the results you want.

[Kev Williams]

My 2 Synrad powered ULS and Gravograph lasers do follow a somewhat equal power-to-speed pattern. Emphasis on 'somewhat'. Meaning, I get roughly the same results from 80% power and 80% speed or 40% power and 40% speed. Not exactly, but close.

My Reci powered Triumph however, does not. It seems to follows a path more closely prioritizing the power setting. Meaning, if I start something at 200mm/sec and 25% power, I will get roughly the same engraving results at 400mm/sec and 25% power, just twice as fast. The machine really does seem to adjust the power to meet the speed...

But even 'twice as fast' isn't quite true, because the faster you set a Chinese machine to run, the more end-to-end "overrun" is added, to compensate for the slowing and stopping for the turnaround. This adds time to the engraving process, sometimes quite a bit of time. Example, if you're engraving something 2" long, it will likely engrave faster at 100mm/second than at 500mm/second because the overrun is so much less at 100mm/sec. The overrun seems built-in, so the controller doesn't care that the lens never comes close to reaching full speed across 2" at 500mm/sec, it will still add in the extra overrun.

So to echo those above, there's no cheat sheet for lasering-- except those you make for yourself! Once you find a setting that works, keep a record of it. For my western machines I have large files of saved settings that work for certain jobs. For my Chinese machine, I have a large NOTEPAD file with my settings wrote down. When new projects come in, it's all trial and error to start with... :)

[Roy Sanders]

This thread was fascinating to read; my problem is I am still stuck of the word "polystyrene." It doesn't seem possible to engrave that stuff, it melts so easily

[Joel Ifill]

I'm a welding engineer by trade and spent quite a bit of time developing laser welding parameters.

Even if your laser power settings are linear (50% power is exactly 50% the watt energy of 100%), the amount of time to heat the surface to the same temperature is not, therefore the relationship to speed is not linear and the formula is highly dependent on the material, thickness etc. Enough so, that the only way to ever develop this is "guess and check" the vendor or whoever has to sit down and experiment, however as soon as you change the overall speed of the machine or the laser wattage you would have to build the charts again.

In case anyone is wondering, cutting and burning parameters are going to be dominated by energy density and the kinetic speed of heat. Kinetic speed of heat is how fast heat energy is sinked into the surrounding material, it's pretty intuitive that you can whip your hand through a lighter but can't leave it there, in this regards the faster you move you have to modify your input energy non-linearly to get the same amount of heat input. In laser cutting applications you need a very high energy density to cut instead of burn and melt, which is now a function of the kinetic speed of heat. It's also the same reason you can't cut 2" thick wood on a 40w by just going "very slow"

The short story: I spent months researching this and we decided it was way easier to guess and check with a basic formula than to try to model that behavior for laser welding and cutting. That was at a national lab with several PhD's and folks smarter than me researching the topic.

[Richard Rumancik]

Mayo, there are too many things going on to be able to do much math on the speed and power parameters. There are some old threads on this topic if you choose to read more. Although I assume that for small lasers power output is linear (eg on a 100 watt machine 25% = 25 watts, 50% = 50 watts, 75% = 75 watts etc) I can't say the same for speed. Epilog explains it this way:


RECOMMENDATIONS
SPEED AND POWER
The Power settings are linear(50% power is half as much as 100% power).
The Speed setting scale of 1% to 100% is not linear (100% speed will not be twice as fast as 50% speed). This non-linear scale is very useful in compensating for the different factors that affect engraving time, but using speed to predict a jobs engraving time is not practical.


Note the explanation: "This non-linear scale is very useful in compensating for the different factors . . . " Make sense? Not to me it doesn't. But the point is that for this laser speed is not designed to be linear in the firmware. Your machine? No idea.

In addition some lasers use the speed scale % based on maximum raster speed. I don't know what Trotec does. But GCC seems to set the vector speed % based on maximum rastering speed. So for vectoring if I specify anything faster than ~30% speed it flatlines. It will accept a higher vector speed setting, it just won't actually do it. So you can fool yourself by commanding a higher speed, thinking the laser is actually doing what you asked.

You could find out what your laser does by plotting a 24" /60 cm horizontal line and timing the travel as you change speed. (Actually it is easier to do several passes back and forth as it makes timing the event easier.) If you are interested try it at 25%, 50%, 75% , 100% speed and see what you observe. (Of course there is a bit of turn-around time involved here but it is the best you can do for an approximation.)

[Dan Hintz]

Mayo, there are too many things going on to be able to do much math on the speed and power parameters. There are some old threads on this topic if you choose to read more. Although I assume that for small lasers power output is linear (eg on a 100 watt machine 25% = 25 watts, 50% = 50 watts, 75% = 75 watts etc) I can't say the same for speed. Epilog explains it this way:


RECOMMENDATIONS
SPEED AND POWER
The Power settings are linear(50% power is half as much as 100% power).
The Speed setting scale of 1% to 100% is not linear (100% speed will not be twice as fast as 50% speed). This non-linear scale is very useful in compensating for the different factors that affect engraving time, but using speed to predict a jobs engraving time is not practical.


Note the explanation: "This non-linear scale is very useful in compensating for the different factors . . . " Make sense? Not to me it doesn't. But the point is that for this laser speed is not designed to be linear in the firmware. Your machine? No idea.

In addition some lasers use the speed scale % based on maximum raster speed. I don't know what Trotec does. But GCC seems to set the vector speed % based on maximum rastering speed. So for vectoring if I specify anything faster than ~30% speed it flatlines. It will accept a higher vector speed setting, it just won't actually do it. So you can fool yourself by commanding a higher speed, thinking the laser is actually doing what you asked.

You could find out what your laser does by plotting a 24" /60 cm horizontal line and timing the travel as you change speed. (Actually it is easier to do several passes back and forth as it makes timing the event easier.) If you are interested try it at 25%, 50%, 75% , 100% speed and see what you observe. (Of course there is a bit of turn-around time involved here but it is the best you can do for an approximation.)


I think Epilog's explanation may be a bit misleading, Richard. From a "human watching" standpoint, the speed is not linear, i.e., a project that takes 10 minutes at 50% speed will not take 5 minutes at 100% speed. This is because the human cannot take the ramp up/down time at the end of each line into account when engraving. From a substrate standpoint, however, the laser will fly over it at twice the speed for the main portion of each line. It's moving twice as fast over the substrate, but the ramp times adds an additional (for example) 20% to the time... that 5 minute job is actually 6 minutes.

For vectoring, the same issue comes into play... but instead of ramp up/down time at the ends of each engraving line, the slowdown comes from turning corners.