Results 1 to 9 of 9

Thread: 3d engraving with DC tube lasers

  1. #1

    3d engraving with DC tube lasers

    I know to do real 3d engraving you need an RF tube laser, but I was just curious what the physical limitation was on varying the pulse power on a glass tube laser. I tried to Google it and couldn't find anything substantive. I've always heard it's that the tube can't stabilize to the new power level quickly enough, is that correct?

    I'm wondering from a purely academic standpoint. The Glowforge does multiple power level engraving. I'm assuming it's because they engrave at glacial paces, meaning they have oodles of time to switch the tube's power level. They claim it's a magic power supply that they custom designed, but I don't buy it. I'd love to learn something new though about physical limitations to changing the power level on the fly like the RF tubes do.

    And also I realize dithering gets you most of the way there for nearly all real world applications, but hey I'm an engineer and am curious about the underlying physics

  2. #2
    Join Date
    Sep 2016
    Location
    Georgia, USA
    Posts
    394
    You might enjoys watching Russ' recent two part video with his experiments trying to understand 3D engraving on a glass tube laser. He isn't teaching or saying he is always right but just letting people follow along as he experiments and learns.

    https://www.youtube.com/watch?v=pzropA_AcL8

    https://www.youtube.com/watch?v=6F32EjrQLDQ

    I have seen better output from attempts at 3D engraving with glass tube lasers but Russ is always entertaining to watch.
    700mm x 500mm Ke Hui KH-7050 Laser
    80W EFR F2
    S&A CW5000 chiller
    Chuck style of rotary attachment

  3. #3
    It's the rise time of the tube from excitation to emission

    DC takes time, RF is almost instantaneous , DC also suffers with Q Spike, the initial pulse from the trigger current fires a higher power blast for a fraction of time
    You did what !

  4. #4
    Join Date
    Mar 2011
    Location
    Olalla, WA
    Posts
    1,532
    Quote Originally Posted by Bert McMahan View Post
    I know to do real 3d engraving you need an RF tube laser...
    What is it about 3d engraving with a glass tube laser that is not "real"?
    Shenhui 1440x850, 130 Watt Reci Z6
    Gerber Sabre 408

  5. #5
    Quote Originally Posted by Dave Sheldrake View Post
    It's the rise time of the tube from excitation to emission

    DC takes time, RF is almost instantaneous , DC also suffers with Q Spike, the initial pulse from the trigger current fires a higher power blast for a fraction of time
    Thanks Dave. Do you have any suggestions on where I can read more about this? I'm a bit confused on the specifics; during a raster, doesn't the tube come on and off fully between on and off dots? And if so, why can't dot number 2 be at a different power level than dot number 1? I'm not trying to argue, I just enjoy learning the fundamentals. I was hoping you'd chime in


    @Rich, sorry I misspoke. By "real 3D engraving" I meant the variable power rastering that the RF tubes seem to do much better than the DC tubes. I may just be ignorant on this but in my newbie experience with hobby lasers the power level is a global setting for the whole raster, not a per-pixel setting, which as I understand you can do on the RF tubes. The 3D mapping on the RF tubes always seems to look better than the dithering method on DC lasers, but then again maybe it's all in the software.

    Again I fully admit my total lack of understanding on the matter; hence this post. If I'm mistaken and normal DC tubes can do per-pulse power modulation I'd love to be corrected. I found a lot of marketing jargon but very little in the way of physical explanations

  6. #6
    If I'm not mistaken (if I am Dave will correct me ), (let's say 'most') RF lasers that we engravers use, don't vary their power in 3D or photo situations, they simply vary the timing of the firing. Or more simply, they too engrave in halftone (aka dithered). Why RF fired laser's shine is because they're always firing in pulses, and as such they can fire a single 'dot' and that dot will be pretty close to as deep or as noticeable as any other engraving it's producing. A DC laser just can't fire that fast.
    ========================================
    ELEVEN - rotary cutter tool machines
    FOUR - CO2 lasers
    THREE- make that FOUR now - fiber lasers
    ONE - vinyl cutter
    CASmate, Corel, Gravostyle


  7. #7
    why can't dot number 2 be at a different power level than dot number 1?
    It can, the problem is that q-spike or trigger ignition that takes place at above maximum power so that initial blast makes the dots all pretty much the same.DC lasers tend to work like news print, lots of dots, the closer they are together the darker it looks.

    On an RF there is no q-spike, the power requested from the controller is what you get, if you ask for 10% you get 10% on a DC if you ask for 10% you get a 110% spike that drops back to 10% after a very short time. This is why if you look at a raster done with a DC tube into acrylic at the edges where it changes direction you will see a ridge that is deeper than the rest of the line (test with clear acrylic and it all becomes very noticable)

    For starter research have a read up on PWM in lasers, that will get you started on the 3 basic factors (Amplitude, Duty Cycle and Frequency) once you get a handle on that the rest becomes pretty simple when you take into account the constraints of DC ignited tubes (common DC tubes rather than top end GSI or Coherant stuff)
    You did what !

  8. #8
    Quote Originally Posted by Dave Sheldrake View Post
    It can, the problem is that q-spike or trigger ignition that takes place at above maximum power so that initial blast makes the dots all pretty much the same.DC lasers tend to work like news print, lots of dots, the closer they are together the darker it looks.

    On an RF there is no q-spike, the power requested from the controller is what you get, if you ask for 10% you get 10% on a DC if you ask for 10% you get a 110% spike that drops back to 10% after a very short time. This is why if you look at a raster done with a DC tube into acrylic at the edges where it changes direction you will see a ridge that is deeper than the rest of the line (test with clear acrylic and it all becomes very noticable)

    For starter research have a read up on PWM in lasers, that will get you started on the 3 basic factors (Amplitude, Duty Cycle and Frequency) once you get a handle on that the rest becomes pretty simple when you take into account the constraints of DC ignited tubes (common DC tubes rather than top end GSI or Coherant stuff)
    Very interesting, thanks for the explanation! I got to thinking about all of this because the Glowforge is claiming to do multiple power levels with a RECI tube due to their magic custom power supply. I was wondering about the fundamental processes happening- I was aware of the spike but hadn't considered how it affects the engravings. I just knew that most DC tubes didn't have as good of control over their engraving processes as the RF tubes did, and was wondering what was going on there and how you might be able to compensate for some of the bad effects using software tricks.

    I wonder, if there's a spike when the ignition starts, is there a way to leave the tube at a very low power state the whole time the engraving is on, and instead of varying the power between Off and On you could vary it between "Very Low" and "On", assuming the "Very Low" state wasn't actually ablating the material. I'd assume you'd need a very low power tube to be able to get it to start lasing at a low enough level to not burn your material. I know with my 80W tube it's hard to control it when I want very low power engravings, but then again I don't have a very fancy controller

  9. #9
    You can provide any DC tube with a pre-ionising current, it will reduce the rise times but there is also the danger of photon saturation in narrow body tubes that have a small amount of gas available to lase. It won't remove the required current spike but you can flatten the curve out to some degree. The only real way GF has achieved this is by implementing PWM properly but that also comes at a net cost of speed as the tube response time is effectively pretty slow.

    On an expensive machine with a well made high end controller and PSU it is easier to do but the physics of DC tubes will still restrict the speed you can do it "at"

    RECI and EFR tubes (wide body) are designed to have a 4mA current run through them at all times they are running, it does help but not much without the use of proper PWM control (most Chinese machines are wired analog) the Chinese controllers have a good PWM function available but very few manufacturers actually use it to it's full capability due to the limitations of cheap DC tubes
    You did what !

Posting Permissions

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