New Machine Build DIY G-Code Laser Cutter

# Thread: DIY G-Code Laser Cutter

1. ## DIY G-Code Laser Cutter

Hello to everyone.

I am in a process of designing and building a DIY CNC Laser Cutter that primarily will be used for cutting parts out of plywood and balsa, but not exclusively.

Now, the first question that fall on everyones mind is why would anyone want to do that, when there is many
commercially available Laser Cutter/Engraver machines for every pocket. Well, there is couple of reasons, that I
will try to explain, briefly. I need a laser to make some airplane kits that involve cutting parts from plywood and
balsa. I was try to use laser cutting services around me (there is couple of them) and I face in with some
difficulties. First they all ask for Adobe AI input formats, and I design in Solid Works or Rhino. I was try any sort
of conversions and I can tell: this does not work properly. Lost about a month on this try and error attempts. I was
even try to make my design in AI. It was nightmare. AI simply is not designed for such complex task, even in
Rhino it is difficult if not impossible to do that. Secondly, all machine around me use timing belts mechanism to
move laser head, not precise enough for what I need, especially when you combine cheep machine and bad
maintenance :-). Finally, they have enough business that they do not want to spend much time adjust cutting
parameters for my delicate job :-) .

So, the goal is to design and build machine that:

1. should run on G-Code (file format conversion fall out)
2. will use ball screws to move laser head. (Much precise positioning, with price of high speed moving) I guess,
high speed engraving will just fall out. I assume that maximum speed will be limited somewhere between
100mm/s and 300mm/s. But, when I look at speeds used to cut such material, I can see that, for 100W power or
less, speeds goes between 3mm/s and 180mm/s (40W 2,3mm MDF versus 100W 2mm plywood, respectively).

So, basically this is just a CNC router machine with laser head instead of spindle. Hopefully, I will add spindle as
option.

Here, I plan to share and discuss all my design decisions, maybe change some of them, based on your
comments, for better design over all I hope, to share, complete design, in Solid Works, to share all calculation
and process that involve it, step by step, and to share some build log pictures, when build starts. If this is not a
right place for this, I ask moderator to direct me to right place.

At the end, I have one question:

Is there any reason why I should NOT put laser tube on gantry? I found some pictures of large industrial
machines that do that:

New model reci 150w laser cutting machine for metal 1325 on Aliexpress.com | Alibaba Group

but on smaller machines it is rarely seen. This is only situation I found:

.

At the end, there is some parts that I was modeled in Solid Works. It is fully functional models that can be freely

for further use. Maybe it will be useful to someone :-)

Have a nice weekend.

2. ## Re: DIY G-Code Laser Cutter

Well, the first machine you posted has the laser in the cabinet, its a fiber machine. Belt drive is pretty accurate, more accurate than the wood your cutting, and much easier to set up, and faster for sure. Any of the off the shelf lasers can be converted to Gcode, lots of hardware for that. My laser uses correl, and i can import DXF directly in, so never really been an issue for me.
Maybe a higher power blue laser would be better for this, no optics to really keep aligned.

Sent from my SM-T813 using Tapatalk

3. ## Re: DIY G-Code Laser Cutter

If I understood you correctly, you suggest that better approach would be if I convert some of the shelf lasers. Chinese laser with that size of working area coasts about 3.000,00 \$, plus import taxes (free shipping). I am not yet check how much would be conversion, but I will certainly check. About conversions, I was neither have problems with Corel Draw. I was have a problem if original files was made in Rhino or SolidWorks, then exported as DXF then loaded in Corel Draw or AI. That DXF, has substantial error, small but persistent. I think, but not sure that it was caused by line thickness. Whatever, files that was originally made in Rhino or SW, (dimensional error on the side) has caused that laser machine work erratically with lots of stop and go situations, not smoothly as it work if file is originally made in Corel or AI. Part that was cut (dimensional error on the side) has correct shape, but owner of the laser refused to cut that files. Others even not want to bother with me. :-( After load a file in AI (exported from SW or Rhino) and magnified some curve it looks like it is made from lots of small lines that do not chain each other, sometimes they chain (end of one is start of next) but on many points, line that is part of curve connect two adjacent lines with two start or two end, and curve connected that way cause that head must go back and forth all the times... That approach is definitely behind me. What I am interested is pure G-Code machine that is drive on MACH3 or some other similar program. So, If, after conversion of the shelf laser, I still has to use Corel Draw or AI or some other software of that type, it is not for me.

You suggested me an interesting idea, now I am going to make some tea for me, it will be a long night :-) Thanks for your respond.

4. ## Re: DIY G-Code Laser Cutter

I believe that most commercial laser cutters use a form of g-code under the hood; they provide the graphics programs as a front end user interface because most people don't want to have to deal with g-code. Rhino will save drawings in AI or DXF format; if they don't work on laser cutters you probably are doing something wrong. Check your tolerances. If your vectors are exporting as a lot of tiny segments, try using the "Join" command on them before export.

If you are building your own laser cutter, be sure to provide a safe enclosure for it; a laser beam can be quite destructive if it hits something reflective and bounces into your eyes or hits another unprotected part of your body. Timing belts are fairly precise if well-adjusted; the precision comes from the motors; the belts just transmit it. They don't have the stiffness of ball screws, but that's not a problem if there's no resistance to motion. Ball screws with high precision come with high prices, and if you run them at high speeds they can "whip", especially if they're thin.

5. ## Re: DIY G-Code Laser Cutter

Originally Posted by awerby
I believe that most commercial laser cutters use a form of g-code under the hood; they provide the graphics programs as a front end user interface because most people don't want to have to deal with g-code. Rhino will save drawings in AI or DXF format; if they don't work on laser cutters you probably are doing something wrong. Check your tolerances. If your vectors are exporting as a lot of tiny segments, try using the "Join" command on them before export.

If you are building your own laser cutter, be sure to provide a safe enclosure for it; a laser beam can be quite destructive if it hits something reflective and bounces into your eyes or hits another unprotected part of your body. Timing belts are fairly precise if well-adjusted; the precision comes from the motors; the belts just transmit it. They don't have the stiffness of ball screws, but that's not a problem if there's no resistance to motion. Ball screws with high precision come with high prices, and if you run them at high speeds they can "whip", especially if they're thin.
I agree that you shouldn’t use ball screw drives vs belts on CO2 lasers. If you want high accuracy invest in high resolution drive/stepper combinations.

Sent from my iPhone using Tapatalk

6. ## Re: DIY G-Code Laser Cutter

[QUOTE=Dean448;2171436]I agree that you shouldn’t use ball screw drives vs belts on CO2 lasers. If you want high accuracy invest in high resolution drive/stepper combinations.

As I understand you and underthertire and awerby, all of you said me the same. For Laser machine timing belts are mandatory, because it will give me enough speed for scan moving that is used for engraving and if it is do as it should, it can give me cutting precision that is equal or better than cutting process technology itself. If I choose this solution, I will have a dedicated laser machine, with all capabilities of a decent laser machine. That way, the question that is left is only controlling/driving mechanism that has both, software and hardware implications. If I goes this way, numerous solutions is available form buying a high class Laser machine, to modifying some cheep machine with my will. The end result would be dedicated laser machine, that potentialy will solve the problem with cutting/importing plus, I will have option to engrave with decent speed.

Other solution that I was initially opted was to go to ball screw solution, that will solve the problem with cutting/importing (as it will with timing belts), but I will have to sacrifice high speed engraving and maybe engraving completely, with option to add spindle on and have router too on the same machine.

So, choice is between engraving and routing as an option. The choice has also implication on what for I will use this machine. If I choose dedicated laser with no spindle option, I will have opportunity to make some money on the laser service market, when not cutting for myself. To have two machine in one is pretty temptation to me. Something that I was thinking just as far option (Z axis spindle) now has to bee built in from start or complete project will be meaningless because of the self laser machine (with or without some modification) offer a solution to my initial problem.

I have to rethink this idea again, from start. Thank you all for helping me.

BTW, no one has said anything about mounting laser tube on gantry.

7. ## Re: DIY G-Code Laser Cutter

Originally Posted by underthetire
Well, the first machine you posted has the laser in the cabinet, its a fiber machine. Belt drive is pretty accurate, more accurate than the wood your cutting, and much easier to set up, and faster for sure. Any of the off the shelf lasers can be converted to Gcode, lots of hardware for that. My laser uses correl, and i can import DXF directly in, so never really been an issue for me.
Maybe a higher power blue laser would be better for this, no optics to really keep aligned.

Sent from my SM-T813 using Tapatalk
So I am upgrading an Epilog Co2 laser to use either a 455nm laser (Chinese) or a fiber laser (Probably a q-switched FAP 808nm or chinese) so a couple questions.
1. The Synrad laser is RF input (5v TTL with a tickle pulse) Local EE expert says I can just make a RF coax to 2pin header connector for input to the 455nm laser controller. Is that correct?
2. Can I do the same as the above for the FAP q-switch? I have zero familiarity with q-switches...
3. Or do I need a separate controller for the Q-switch?
4. I NEED to engrave anodized aluminum, AND stainless steel. The 455nm will for sure engrave both, but not sure how well (Compraed to the 35W Co2), or as permanent. I need a permanent mark.
5. Will the 808nm FAP or Chinese engrave SS and anodized AL? I know it will drill holes in SS, so presumably it will engrave really well.

Which would you do? I may just buy a \$500 chinese 455 engraver instead, but it would be nice to pop just the head on the Epilog, and use it for cutting also. Refilling the Synrad Co2 is nearly \$3K when it is time again.

8. [QUOTE=redbaronfc;2171544]
Originally Posted by Dean448
I agree that you shouldn’t use ball screw drives vs belts on CO2 lasers. If you want high accuracy invest in high resolution drive/stepper combinations.

As I understand you and underthertire and awerby, all of you said me the same. For Laser machine timing belts are mandatory, because it will give me enough speed for scan moving that is used for engraving and if it is do as it should, it can give me cutting precision that is equal or better than cutting process technology itself. If I choose this solution, I will have a dedicated laser machine, with all capabilities of a decent laser machine. That way, the question that is left is only controlling/driving mechanism that has both, software and hardware implications. If I goes this way, numerous solutions is available form buying a high class Laser machine, to modifying some cheep machine with my will. The end result would be dedicated laser machine, that potentialy will solve the problem with cutting/importing plus, I will have option to engrave with decent speed.

Other solution that I was initially opted was to go to ball screw solution, that will solve the problem with cutting/importing (as it will with timing belts), but I will have to sacrifice high speed engraving and maybe engraving completely, with option to add spindle on and have router too on the same machine.

So, choice is between engraving and routing as an option. The choice has also implication on what for I will use this machine. If I choose dedicated laser with no spindle option, I will have opportunity to make some money on the laser service market, when not cutting for myself. To have two machine in one is pretty temptation to me. Something that I was thinking just as far option (Z axis spindle) now has to bee built in from start or complete project will be meaningless because of the self laser machine (with or without some modification) offer a solution to my initial problem.

I have to rethink this idea again, from start. Thank you all for helping me.

BTW, no one has said anything about mounting laser tube on gantry.
I suggest you export your original items at a larger scale. I was having the same problem with small circles and other details in the drawing. Now I make my drawing in corel 10x bigger before saving and then after I import it in rdworks I size it to 10%, I am able to get those little details back.

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•