Laser power output

I’m doing some research with my Glowforge and need to determine exact speed (in mm/sec) and laser output. I’ve been able to convert the “Speed” values to mm/sec, but I’m not able to convert any percentage of Full Power to watts. For example, if I run the Glowforge at 20% Full Power, how many watts is that? Has someone done this and/or are these values available online somewhere? Thanks!

Per conversations way back at the beginning it’s not a straight scale. AFAIK there’s never been a successful grid made up.


I don’t know of anyone who has measured the beam power at all - that said, it’s likely to be a touch different between machines depending on the initial factors calibration and age of the tube.

Really, what you’d need to do is get a laser power meter. I haven’t used one, and don’t have one - @jamesdhatch may have a recommendation on a decent one (I think he has one).

Since you can’t test pulse, you’d need to set up a job that fires the laser long enough and in a small enough location for the tool to measure the output.

That power scale has never been released, so you’d have to do some testing…

At minimum, I would test at full power. Then at 100 power, then at 1 power. With 1 and 100 filled in, you could fill in the rest of the data. I’d probably spot check a few of your fields to make sure you’re on the right track.


That’s the problem. The meter I have uses a calibrated temperature gauge - you stick the probe in the beam while it’s firing, the probe heats up, that’s then translated to a power scale based on the temperature the probe reaches in 30 seconds.

The problem with trying to use it on the GF is the lack of a “test laser” button that’s common on other machines. Without that you need to create a design that will fire in one place and will take 30 seconds to fire. There’s some trial & error needed to do that. I’d try something with tiny circles but the GF ignores overlapping vectors so I’d need to do something with a bmp maybe?

Never seemed to be worth trying to work out. On other lasers where you can inadvertently overpower the tube and this drastically shorten its life it was useful to check on the tube health over time. With the GF’s power throttling I didn’t find a need for trying to monitor changes in power output over time.


I appreciate the feedback and thoughtful replies! My hope is that someone on the Glowforge staff will see this question and have or know of some “insiders” data that resulted from their calibrations using a high-quality power meter.

I think it would work just on overlapping circles set to cut - it’s only the engrave function that gets knocked out where vectors overlap. We’ve seen enough people asking, why is it making two passes when it’s only set to 1 pass? :slight_smile:

They don’t monitor anything regularly except problems and support (since posts there generate trouble tickets). Not saying they won’t see this, but a good chance not. And the odds of sharing that data, I don’t want to be a pessimist, but that also seems fairly low. They are generally pretty tight-lipped around things like that.

Just out of curiosity, can you tell us what you’re working on? I’d love to see someone get numbers like that - just because I like seeing numbers lol


Wondering whether this device would get the numbers I want???

1 Like

I guess I’ll hope that one of the Glowforge folks stumble onto this discussion. My application is pretty far removed from the usual work of the Glowforge. I’m working on developing a technique for improving the germination of seeds. The data I’ve collected so far is very encouraging and I’d like to publish the methods in terms/units that are more readily applicable to anyone using a laser cutter other than a Glowforge.

1 Like

This device seems to be a more cost-effective alternative:

The exposure time is set on each meter and is usually around 30 seconds. I plan on creating a square of a size just smaller than the measuring device and, at defined power percentages, setting the number of lines per cm so that the exposure time is exactly met. That should result in several readings as the power percentage is varied between 0 and 100. Unless I’ve overlooked something, I should be able to construct a response curve relating Glowforge’s power percentage with actual watts.

Keep in mind the machine’s speed and power is not consistent over small areas, by design, to reduce the effect of burning corners as the head has to change direction. A lot of effort was put into that and the reduction in power during those transitions made a big difference in print quality when it was released. Any measurement over a small space will not be representative of the power output as the head is moving across the bed.

You can try to engage support by posting in P&S but it’s mostly likely they’ll just move that discussion back here and/or come back with this - it’s on their canned answer sheet for questions and all they’ve ever said with regards to power. You can use the search function to see how those discussions have gone in the past.

Your “project” sounds far better suited to a cheap Chinese laser over which you have complete control, rather than the GF - which abstracts much of the functionality and control away from the user, by design, to make it virtually foolproof. That’s a good thing for the target market, frustrating to tinkerers.


Maybe creating a spiral cut vector file will be better than a raster engraving for this. Ironic that I replaced my K40 laser with the Glowforge.

Personal opinion - I think at speed 100, which is what, something like .06-.07” inches per second, you could do a very, very small circle that’s stacked on top of one another to make it multiple passes and it wouldn’t slow down any further.

Sounds like an interesting project!

1 Like

Unfortunately, we have many years of evidence that Glowforge folks do not disclose anything relating to the technical specifications of the machine outside what is already published. There’s a 0% chance of getting an official answer to your question. I’d use it as an excuse to buy that meter.


I should probably work on setting realistic expectations. I agree - the chances are realistically 0.

1 Like

That’s the one I have. You’d have to design something that had enough tiny circles stacked on each other for it to run 28 seconds. I think you’ll need to disable the front door interlock so you can slide the probe in but keep the dial outside (I think the dial is more than 2" so you can’t do it with the tray taken out and the probe set on a stack of material to put it within a half-inch of the laser head). If you do that then you don’t need to get the time as precise - just so long as the laser will burn at least 28 seconds you’ll be able to yank it out when your timer hits 28 seconds.

Doing a square or pattern may not get the right answer. These are designed to measure the heat transferred to a spot on the probe end. Making a square or pattern the size of the probe end would likely diffuse the heat provided too much - you’d be warming the full probe end faster than it is calibrated. I expect part of the calibration takes into account how long it takes to diffuse into the probe end. You’d need a design (overlapping circles) that mimic as close as possible a laser beam spot.

1 Like

Here is a circle that is duplicated 8 times. Diameter of 0.08". It should run for about 30 seconds at 100 speed. I think.

0.06" inches per second (it shouldn’t slow down at all at 100 speed)
.08" circle is a circumference of .2512"
.06 * 30 seconds is 1.8"
1.8 divided by .2512 is just over 7 copies of the circle.

You could probably just set it to 2 passes and kill the job when the proper time has elapsed. (431 Bytes)


Thanks! So the idea is to import this file into the Glowforge app and place it in the middle of the power meter that’s laying in the Glowforge??? Having never used a laser power meter before I’m trying to visualize how all this will work. I did order the one that’s just over $100 so I expect to be able to generate a curve correlating Glowforge’s power setting to actual watts.

Yup! That would be the idea. I’ve looked a little into how the different power meters work and I think what you’ll want is for it to be as dead center as possible on the power meter. The very small circle should do ok at emulating what would be a “test pulse” on other systems - the idea is just to concentrate the power into as small of an area as possible for the duration needed. @MyDogsThinkImCrazy has some pretty good methods for getting exact placement using the coordinates available in the UI, as they relate to the bed.

I think you should be able to at least come close enough for government work, as they say. The meter itself will have a certain margin of error. I might be wrong, but close should be good enough for what you’re looking at.

The thing is - most people don’t know what the laser is actually firing at from a watts perspective anyways. They might have a % setting available, but the tubes also have a minimum firing power as well. But either way - as long as you’re not looking for exact, I think you can get close.

I, for one, am just very curious to see the readings for the different precision power levels. Do you have a Pro, Plus, or a Basic? The precision power settings are supposed to be the same between all systems - so 100, 20, 50, 73, etc. precision power are the same on all systems. The Pro, with 45 watts, has a higher hit at full power. So presumably for a pro you have a different drop off between full power and 100 power, than you would with a basic or a plus (since it has to drop at least 5 watts in output power).


Thanks for the guidance! I have the Basic Glowforge and I look forward to gathering some numbers from the power meter.

1 Like

Now I’m going to have to find my meter in the garage and do the test on my Pro. I’m interested in what the machines calibrate out to as well.