Jtech lasers

[Echd]

Picked up a Jtech 14 watt laser today for a good price secondhand.

If you're in need of a laser to complement a cnc and need something bigger than one of the little Chinese gantry machines, it's a solid option. I'm not a huge fan of a $1k laser diode unit (I paid around $600, but msrp is way up there) coming with a 3d printed housing, but hey, if it works it works.

They have a higher power version- the 24 watt quad (which apparently is closer to 28 watt) but for the delta in price I'll settle for this.
Just a few test burns of random logos and things tonight to see what it can and can't do. Cutting is a secondary consideration for a diode IMO- they really are better suited to engraving- but this one gets through 3/8 walnut with relative ease. 5 passes at 10 IPM and the parts fell right out. Sure, I could do it with an endmill in a twentieth of the time, but that's not the point. Sometimes you need that Itty bitty kerf. With the way diode laser setups are improving though I won't be surprised to see 40 watt or greater true output before long. Although the safety aspect of that is kind of concerning in an unenclosed environment...

 

[Wilsoncb]

Interesting, do you think the laser you have is capable of cutting thin sheet metal?

 

[Echd]

Nope. You want to do metal, you'll need to spend a lot more. I think atomstack recently came our with a small diode in a wavelength appropriate for etching or thin metal cutting but I don't know how well it works.

If cutting in general is the goal, a considerably more powerful co2 laser would be desirable. You have to drive these slow for cutting.

That said diodes are getting better and better. I expect we'll see some punching in the weight class of 40w co2 lasers soon. They also have much longer lives than tube lasers and while no powerful laser is "safe" being in the visible spectrum at least might give you some warning prior to blinding yourself. It's also easy to fit a diode to a much larger cnc machine or even on a small unit you can just take to your part, whereas a co2 laser would need to have a part fit in it or the pass through port.

 

[riggsp]

I just got a Jtech 14 watt laser myself, and after playing around with it some on my CNC router, it does very well...I haven't tried to cut anything with it yet, but I've done several engravings and even a photo (same pic that's on my avatar) just for grins and to see how it works...when I did the photo I left all the software setting to default (no changes)...the outcome is kind of rough but I think a little tinkering with the settings will change it...file attached.

 

[mkepke]

Cool thread. I’ve been eyeing a laser upgrade as well.

What’s everyone using for CAM - Lightburn? Vectric?

-Mark

 

[Echd]

I use vectric and fusion for most things but lightburn for lasers.

I'm a little concerned about doing projects that are joint cnc and laser work with differing programs so I might get the vectric module at some point. I know you can use a postprocessor jtech lists and code in a "laser endmill" without buying their module but I haven't tried yet.

 

[riggsp]

I use Vectric Aspire with the Vectric laser module...I was going try Lightburn, but it doesn't work directly with my brand of controller...I wrote my own post processor to use for my laser because neither Jtech or Vectric that would work with my machine.

 

[Echd]

Ah, looks like integration with Vectric is quite simple even without paying for their laser module. The Onefinity module I was using already came with a couple of lower power laser presets.

Here are their instructions as well as the post processors that you will need.

I suppose I'll just make multiple "tool entries" for different applications as opposed to just varying the power for every cut.

I spoke with Jtech as well (very responsive to email) about best practices for integrating cuts and burns where focal issue would need to change for ideal performance. They recommended doing a burn pass where the Z drops each pass to where the laser shroud is just above the workpiece surface for the final pass. Logical. Lightburn has some Z axis control and integration but I've never used it, as my old laser was a manual focus piece that was directly driven over USB by lightburn feeding it Gcode. I'm much more comfortable with vectric for that sort of work, especially if I'm going to be cutting on the same piece.

My hope now is that I can group multiple uses of the laser under different parameters to run consecutively instead of requiring my permission to proceed to each tool. I've found that to be an occasional hangup with my machine, where I want it to sequentially run toolpaths using the same endmill, but because the specifications changed based on the toolpath type (like when drilling and also doing normal side cutting) it will sometimes want to prompt for a tool change for each program. I'm sure that won't be too hard to surmount, though.

Eta:

Notes after setting up a burn based on jtech recommended specs-

A similar burn through lightburn is faster. I settled on that "planet express" logo for the past couple of days because I thought it was funny. A 2 inchish in size logo with a cutout pass was 19 minutes in lightburn. It shows 39 on my onefinity using a similar program through vectric. That means to me that it's likely lightburn was using a default kerf width that was wider than jtech prescribes. I'm sure that's a setting that could be changed, I just didn't give it any thought. Probably just need to play with it some, but it's been running a few minutes and the burn quality looks good. It is a bit hacky to make it work, but as the laser is held on with a magnetic mount, even if I'm dumb hopefully the worst that happens is it falls off.

 

[riggsp]

Ah, looks like integration with Vectric is quite simple even without paying for their laser module. The Onefinity module I was using already came with a couple of lower power laser presets.

Here are their instructions as well as the post processors that you will need.

I suppose I'll just make multiple "tool entries" for different applications as opposed to just varying the power for every cut.

I spoke with Jtech as well (very responsive to email) about best practices for integrating cuts and burns where focal issue would need to change for ideal performance. They recommended doing a burn pass where the Z drops each pass to where the laser shroud is just above the workpiece surface for the final pass. Logical. Lightburn has some Z axis control and integration but I've never used it, as my old laser was a manual focus piece that was directly driven over USB by lightburn feeding it Gcode. I'm much more comfortable with vectric for that sort of work, especially if I'm going to be cutting on the same piece.

My hope now is that I can group multiple uses of the laser under different parameters to run consecutively instead of requiring my permission to proceed to each tool. I've found that to be an occasional hangup with my machine, where I want it to sequentially run toolpaths using the same endmill, but because the specifications changed based on the toolpath type (like when drilling and also doing normal side cutting) it will sometimes want to prompt for a tool change for each program. I'm sure that won't be too hard to surmount, though.

I saw that list of post processors when I bought my laser, but none would work with my controller...so I modified a Vectric post processor that was listed for my controller, and I use it to control my laser.
If you use the Vectric laser module, you set up your laser in the tool database, then when you selact a laser toolpath all you have to do is enter your speed (intensity) and feed rate...if you have multiple toolpaths that need different settings, all you have to change are those two settings because the tool remains the same and there shouldn't be a tool change prompt...at least that's the way mine works.

 

[greenhornWoodworker]

Im going to be starting a laser engraving business soon (hopefully), if my research proves right. Would people be interested in seeing content geared toward lasers?

 

[Echd]

Woodburning, pyrography, laser work, shou sugi ban... I'm sure you would find plenty of interest for any woodworking here. A few of the members here have a number of much more potent co2 lasers and are very helpful on their use.

 

[Echd]

I had some spare time today so I was fooling with Vectric. @riggsp , what is the max spindle speed you use for vectric? I started at 255 (which an older image on the jtech website said was the max for GRBL machines, which mine isn't) and results were highly anemic. I was confused until I remembered that my CNC machine's controller was updated to have lightburn use 1000 as the max s value. I manually altered some files and the vectric files then cut without issue, but I wanted to make sure I wasn't going to burn anything out in my overeagerness.

 

[riggsp]

I had some spare time today so I was fooling with Vectric. @riggsp , what is the max spindle speed you use for vectric? I started at 255 (which an older image on the jtech website said was the max for GRBL machines, which mine isn't) and results were highly anemic. I was confused until I remembered that my CNC machine's controller was updated to have lightburn use 1000 as the max s value. I manually altered some files and the vectric files then cut without issue, but I wanted to make sure I wasn't going to burn anything out in my overeagerness.

I set the max speed at 1000 rpm which on my control equates to 100%...since all speed settings are intensity and percentages, your 255 was slightly over 25%...I did a burn intensity test using intensity percentage and feed rate for ranges of 10% and 25 ipm feed rate for a low to 100% and 100 ipm for the high to see what the burn looked like for engraving...best setting, cleanest burn, was 50% at 50 ipm.
I think you'll be ok with the settings you are using with 1000 as 100%
play with the settings on various wood types and keep the samples as setups.

 

[Echd]

Yep, I believe that to be the case. Oddly the jtech setup website mentions 255 as max, which had me really confused when my lightburn files were doing a much better job than my vectric ones! As the gcode they generated wasn't following the exact same syntax I couldn't compare them directly.

I like the interface and am more familiar with vectric but the implementation I am using is a bit "hacky". Seems so weird to see a laser dance all over instead of just rastering, but it is treating it as an endmill in vcarve after all. The quality of the finish is good but it is definitely slower than lightburn, especially when lightburn allows you to fast travel over Un burned space. A 38 minute lightburn job hovers around an hour for vectric using my settings.

 

[riggsp]

Yep, I believe that to be the case. Oddly the jtech setup website mentions 255 as max, which had me really confused when my lightburn files were doing a much better job than my vectric ones! As the gcode they generated wasn't following the exact same syntax I couldn't compare them directly.

I like the interface and am more familiar with vectric but the implementation I am using is a bit "hacky". Seems so weird to see a laser dance all over instead of just rastering, but it is treating it as an endmill in vcarve after all. The quality of the finish is good but it is definitely slower than lightburn, especially when lightburn allows you to fast travel over Un burned space. A 38 minute lightburn job hovers around an hour for vectric using my settings.

Check your g-code file, your tool data base, and post processor...you shouldn't have any "Z" moves unless you are trying to burn 3d files...if you have them, make sure the feed is the same as the "X" and "Y" moves...your machine will only move as fast as the slowest axis setting...may be worth looking at just in case.

 

[Echd]

I have set it to drop z height for cut moves l, there is no z movement for engraving. Jtech instructed me to do so when I asked if doing so was preferable.

Here is the advice they provided earlier to me

Make sure you are going 10 inch/min with multiple passes. If you make your machine rock solid, then you can move the z down per pass. Have the focus set a bit above 1/8" and then divide 1/8" by the number of passes to get the z step per pass" number.

So I would set my z height at 0.25 with the focus tool, then had the z height drop to 0.125 for the cutting pass, and drop 0.01 per pass so it ends up barely being above the material surface at the end. It does seem to improve cutting over starting with the "deeper" focus.

The move speeds are appropriate to pathing of vectric, but their paths are calculated differently from those of lightburn, at least when one is not using their laser module. Whereas lightburn will do a normal raster pattern (which is efficient for a burn) and speed up in white space, my vectric code acts more like an endmill and will hop from area to area- which is normal for an endmill.

I suspect the kerf size is probably also smaller for vectric than lightburn, which is of course adjustable. I have mine set in vectric for 0.006. I'll have to double check for lightburn.

 

[riggsp]

I have set it to drop z height for cut moves l, there is no z movement for engraving. Jtech instructed me to do so when I asked if doing so was preferable.

Here is the advice they provided earlier to me



So I would set my z height at 0.25 with the focus tool, then had the z height drop to 0.125 for the cutting pass, and drop 0.01 per pass so it ends up barely being above the material surface at the end. It does seem to improve cutting over starting with the "deeper" focus.

The move speeds are appropriate to pathing of vectric, but their paths are calculated differently from those of lightburn, at least when one is not using their laser module. Whereas lightburn will do a normal raster pattern (which is efficient for a burn) and speed up in white space, my vectric code acts more like an endmill and will hop from area to area- which is normal for an endmill.

I suspect the kerf size is probably also smaller for vectric than lightburn, which is of course adjustable. I have mine set in vectric for 0.006. I'll have to double check for lightburn.

Setting the "Z" height by the gauge which is actually .220 instead of .250 is the set point if you are not doing "Z" moves when engraving...for cutting, reduce the set point by .125 (1/8") then, as instructed divide the remaining distance (.095") by the number of passes and that will give the cut depth per pass...just be careful you don't move the laser housing into the part (the magnets will let go...don't ask me how I know that).
If you don't use the Vectric laser module you won't have the option of a raster toolpath like Lightburn...the software only sees the laser as another endmill and will follow the profile path as determined by the computer to be the most efficient path...that's why it seems to move from place to place while burning...the Vectric laser module will do raster toolpaths as part of it's strategy and will at the same time vary the intensity if the image being burned has greyscale changes where the normal toolpath will not (the intensity will stay the same throughout...any intensity changes have to be programmed in)...you will need to make sure your post processor is modified to handle that... Just as a note, I set my kerf to .011" but that number is arbitrary and is there to give you an allowance for kerf offsets when doing raster patterns and hatching/crosshatching.

 

[riggsp]

Just did this this morning...the image size is 2" x 2" on unfinished baltic birch ply...intensity set to 50% feedrate 50 ipm for the main outlines and 30% and 50 ipm for the hatching infill... I did this to see how the resolution would be for a small size with lots of detail.

 

[Wilsoncb]

After seeing this post I started thinking about laser cutting wood vs router. The thought was you would not have to worry about buying tools all the time. I understand your would also not be able to do anything other than engraving and cutting shapes. But nonetheless I was intrigued.

After looking into it, It doesn’t look like this would ever be a good idea. However, I would like to hear from anyone who currently has a router/laser machine. From what I can see if you cut solid wood with it, the edges are pretty well burned. Seems like cleaning up the burnt edges would be a real pain.

Also I thought some of you might find this interesting, while looking into this I see where Xtool is coming out with a 50w CO2 laser (no router). It says it is capable of cutting 35mm thick basswood.

xTool P2: CO2 Laser Cutter and Cutting Machine for Desktop Use | 55W

xTool P2 is a versatile and smart desktop 55W CO2 laser cutter featuring 26*14 inch bed size, 600mm/s maximum engraving speed, auto-passthrough, curved surface engraving, dual 16mp cameras, built-in air assist and Lightburn support. With its superb cutting capability, it laser cuts 18mm black...

www.xtool.com

 

[Echd]

It is a specialty tool. However, for engraving, it is a look that many people like quite a bit. Engraving with a bit instead often doesn't look as good, and doing a traditional or vcarve inlay is much more work.

Consumer lasers are emphatically not as efficient at cutting as bits on a router are. There are times it is handy- you don't need as solid of a workholding solution, for instance. For very small parts you don't need to clean up "tabs". For very small or complex curves they may be able to cut a part that will interlock that a bit would have a hard time doing. Cutting with 1/4 and 1/8 endmills is never much of a problem. Cutting with 1/16th, 1/32 cutters? They break easily!

The Cutting is a neat party trick of limited use. Look at lasers like this as chiefly for engraving and burning with occasional cutting when needed. Now, some people have very niche needs where they do use them as cutters- often from very thin sheet goods. Think Christmas ornaments or interlocking "flat pack" designs. In those cases a CO2 laser like you linked can be just what the doctor ordered.

For the hobby user, efficiency isn't always the priority or important at all. I'm working on a mother's day present for instance. It will be a functional jigsaw puzzle of about 80 to 100 pieces in 3/8 hardwood into which the image of my niece will be photo transferred from laser toner. The cutting path is projected to take about 6 hours- absolutely not efficient, but I'm not a wood shop. It can be done with a bit, but would require absolute tons of cleanup manually to make properly fitting parts.

Here is a test cut with a simple engraved picture on top. This idea could easily he made into a simple coaster. I was just playing with kerf size and focus to see how loose or tight a piece needs to be.

But mostly think of these, particularly diode lasers, as engraving tools with some cutting utility on occasion. Could you dimension lumber with a scroll saw? Sort of I guess. Would you? Probably not without a really good reason.