arkypita / LaserGRBL

Laser optimized GUI for GRBL
http://lasergrbl.com
Other
1.26k stars 497 forks source link

Finer engraving #1012

Open fdurville opened 4 years ago

fdurville commented 4 years ago

Would it be possible to increase the image resolution (Quality) to 50 lines/mm? I am currently using a laser which has a spot size that can be adjusted down to about 20-micron diameter. Thus having 50 lines/mm would allow me to realize extremely fine engraving...

arkypita commented 4 years ago

The current limit of 20 lines per millimeter is an arbitrary limit. It is easy for me to remove it. Keep in mind that 20 lines / mm correspond to about 500 dpi which is a rather high resolution.

Increasing the resolution also means that LaserGRBL will need to work with larger files, therefore more memory and more CPU for the process, but as long as you work on maximum dimensions of 30 * 40 cm there should be no problems.

What laser do you have, out of curiosity?

fdurville commented 4 years ago

The laser is a Q-switch fiber laser which I built myself. Frederic

On Fri, May 22, 2020 at 4:14 PM arkypita notifications@github.com wrote:

The current limit of 20 lines per millimeter is an arbitrary limit. It is easy for me to remove it. Keep in mind that 20 lines / mm correspond to about 500 dpi which is a rather high resolution.

Increasing the resolution also means that LaserGRBL will need to work with larger files, therefore more memory and more CPU for the process, but as long as you work on maximum dimensions of 30 * 40 cm there should be no problems.

What laser do you have, out of curiosity?

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-632894038, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6P2XHPCORG3DT644MTRS3MK3ANCNFSM4NIBHUGQ .

arkypita commented 4 years ago

I did some tests on LaserGRBL enabling the possibility to use a higher resolution. The processing becomes very heavy because you have to imagine that 10cm x 10cm at 50 lines / mm become an image of 5000x5000 px, which at 32bit per pixel are about 100MB.

Considering that the image undergoes several steps this means a slower and heavier work in memory.

Nothing unmanageable for a good modern PC, but definitely oversized for the normal user.

Let me know if you want to try a customized version of LaserGRBL I can send you one that admits higher values.

fdurville commented 4 years ago

Thank you very much for the response. I certainly do understand the issue of larger files when dealing with large images and higher res. And it also would increase proportionally the time it takes to engrave the whole image. In my case, the largest image I have engraved so far is about 3x3cm, and I would like to explore only small images (1x1 cm max) with the higher res (doing small engraving on small jewellery for example, my laser can engrave on gold). So yes, I would definitively love to try a customized version that can take higher res values. Frederic

On Tue, May 26, 2020 at 10:16 AM arkypita notifications@github.com wrote:

I did some tests on LaserGRBL enabling the possibility to use a higher resolution. The processing becomes very heavy because you have to imagine that 10cm x 10cm at 50 lines / mm become an image of 5000x5000 px, which at 32bit per pixel are about 100MB.

Considering that the image undergoes several steps this means a slower and heavier work in memory.

Nothing unmanageable for a good modern PC, but definitely oversized for the normal user.

Let me know if you want to try a customized version of LaserGRBL I can send you one that admits higher values.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-634053210, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6MS6GSECCDGYJBDA2TRTPFMPANCNFSM4NIBHUGQ .

arkypita commented 4 years ago

You can try this pre-release https://github.com/arkypita/LaserGRBL/releases/tag/v3.5.0

I'd be curious to see some of your works

fdurville commented 4 years ago

Thanks a lot. I will try it out on Monday.

You can see some of the things I have done here https://drive.google.com/open?id=1_NkvUWjl19VE4ofxgebzaSOk7sWZ0xN8. So far, nothing really impressive IMO, just trying different things and seeing how it looks.

Frederic

On Fri, May 29, 2020 at 5:27 PM arkypita notifications@github.com wrote:

You can try this pre-release https://github.com/arkypita/LaserGRBL/releases/tag/v3.5.0

I'd be curious to see some of your works

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-636199766, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6OSF6EKNBPKUZ5AYHDRUASC7ANCNFSM4NIBHUGQ .

fdurville commented 4 years ago

Thanks again for this update. I just tried it, and it works quite well. Here is the difference https://drive.google.com/open?id=1_NkvUWjl19VE4ofxgebzaSOk7sWZ0xN8 between res 40lines/mm (on blue anodized) and 20lines/mm (on orange anodized), as looked under microscope 20X. The difference is not visible under naked eye, but is clearly visible under magnification. Frederic

On Fri, May 29, 2020 at 5:27 PM arkypita notifications@github.com wrote:

You can try this pre-release https://github.com/arkypita/LaserGRBL/releases/tag/v3.5.0

I'd be curious to see some of your works

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-636199766, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6OSF6EKNBPKUZ5AYHDRUASC7ANCNFSM4NIBHUGQ .

arkypita commented 4 years ago

Hi @fdurville I'm glad to know that the change had the desired effects. What is the size of the rose in the photo?

I think that few users will be able to take advantage of this possibility because a very special and expensive laser and a very high resolution of the motors are required, however it is an additional function that I have enabled with little effort.

Out of curiosity, what is the resolution of your engines? What is the smallest step they can take?

Currently LaserGRBL produces gcode with two decimal numbers after the decimal point, therefore the positioning that LaserGRBL produces is in 10 micron steps. Do you think it could also be useful to increase to 3 decimal places?

Anyway thanks for the donation!

arkypita commented 4 years ago

Errata: LaserGRBL already support 3 decimal places if needed

fdurville commented 4 years ago

I understand that most users will not need that fine resolution... and I really appreciate your making the change. The resolution of my stage step-motor is 400/mm, this is a bit arbitrary as it is easy to change the "micro-step" on the driver board, or in my case on the CNC-shield board (I use an Arduino UNO with a CNC-shield) for each motor. Frederic

On Tue, Jun 2, 2020 at 6:47 AM arkypita notifications@github.com wrote:

Hi @fdurville https://github.com/fdurville I'm glad to know that the change had the desired effects. What is the size of the rose in the photo?

I think that few users will be able to take advantage of this possibility because a very special and expensive laser and a very high resolution of the motors are required, however it is an additional function that I have enabled with little effort.

Out of curiosity, what is the resolution of your engines? What is the smallest step they can take?

Currently LaserGRBL produces gcode with two decimal numbers after the decimal point, therefore the positioning that LaserGRBL produces is in 10 micron steps. Do you think it could also be useful to increase to 3 decimal places?

Anyway thanks for the donation!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-637455198, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6KGXPO6PDOHOHY66ADRUTKDTANCNFSM4NIBHUGQ .

StuartB4 commented 4 years ago

Just wondered what stepper drivers you are using, and have you considered the Trinamic drivers, microstepping up to 256. https://hackaday.com/2016/09/30/3d-printering-trinamic-tmc2130-stepper-motor-drivers-shifting-the-gears

fdurville commented 4 years ago

I use "generic" (i.e., from china) A4988 drivers... (that is based on the Allegro A4988 chip). And they can also be set to up to 256 micro-steps. To my knowledge, all step motor drivers are very similar from an electrical / electronics standpoint, as there are only very few different chips, and those chips differ only in their current handling (and heat dissipation) capabilities.The ones you mention do have a neat heatsink to improve their heat dissipation capabilities, which maybe useful in intensive usage (and also improve their overall reliability). FD

On Tue, Jun 2, 2020 at 1:43 PM Stuart notifications@github.com wrote:

Just wondered what stepper drivers you are using, and have you considered the Trinamic drivers, microstepping up to 256.

https://hackaday.com/2016/09/30/3d-printering-trinamic-tmc2130-stepper-motor-drivers-shifting-the-gears

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-637701094, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6MO5WNEQ55UMBWXR7LRUU223ANCNFSM4NIBHUGQ .

fdurville commented 4 years ago

I just realized that what I made a mistake: the A4988 driver has only up to 16 micro-steps. FD

On Tue, Jun 2, 2020 at 6:16 PM Frederic Durville fdurville@gmail.com wrote:

I use "generic" (i.e., from china) A4988 drivers... (that is based on the Allegro A4988 chip). And they can also be set to up to 256 micro-steps. To my knowledge, all step motor drivers are very similar from an electrical / electronics standpoint, as there are only very few different chips, and those chips differ only in their current handling (and heat dissipation) capabilities.The ones you mention do have a neat heatsink to improve their heat dissipation capabilities, which maybe useful in intensive usage (and also improve their overall reliability). FD

On Tue, Jun 2, 2020 at 1:43 PM Stuart notifications@github.com wrote:

Just wondered what stepper drivers you are using, and have you considered the Trinamic drivers, microstepping up to 256.

https://hackaday.com/2016/09/30/3d-printering-trinamic-tmc2130-stepper-motor-drivers-shifting-the-gears

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/arkypita/LaserGRBL/issues/1012#issuecomment-637701094, or unsubscribe https://github.com/notifications/unsubscribe-auth/ABJ7C6MO5WNEQ55UMBWXR7LRUU223ANCNFSM4NIBHUGQ .

StuartB4 commented 4 years ago

Ok. I use the DRV8825's, they can go up to 1/32 stepping. They can also run at a slightly higher voltage and current than the 4998's.