A Lesson In K40 Laser Repair

The K40 laser cutter has become ubiquitous in hackerspaces and well-equipped home workshops over the past few years, as a relatively inexpensive introduction to laser cutting and a machine that is readily hackable. Tokyo Hackerspace have one, but sadly their laser tube failed after relatively little use. Replacing a laser tube might be a routine component change for some readers, but it’s still worth looking at in some detail.

Their tube had failed at its output lens cooling cap, a component that is glued onto the end of the tube rather than bonded, and which had snapped off. There had been no mechanical stress upon it, but it was found  that the arrangement of their cooling system caused it to drain between uses and thus air bubbles could accumulate. The resulting cooling inefficiency caused enough thermal stress for the bond between the tube and the end piece to fail.

The in-depth analysis of what caused the failure and step-by-step description of the procedure should be of interest to any K40 owner. Little things such as ensuring that the tube is rotated to the right angle for all air bubbles to make their way out of it, or making sure that when the pump is switched off the water isn’t all pulled out of it by gravity seem obvious, but these are traps that will have caught more than one K40 owner.

We’ve covered many K40 stories over the years, but a good place to start for the novice might be this commissioning story, or even this tale of a hackerspace’s modifications to their model.

25 thoughts on “A Lesson In K40 Laser Repair

  1. I find it’s almost impossible to get all the bubbles out unless you rotate the tube to two different positions every time you due up the pump. A couple of little bubbles don’t seem to matter to much though. it

    I’m a bit disappointed that considering the laser fired up they didn’t try to repair the end. Anyone know if this is possible.

        1. Agreed it is a seperate bit of glass tubing that is in line with the main tube to cool down that half mirror on the output side of the tube..just glue it back on ..they do have a tendency to fall off especially on the cheaper tubes!

    1. A member here: if I remember right, they did try to reattach it but found two issues
      1: didn’t have on hand any kind of glue or bond that would both deal with glass and also handle potentially high temps (it was the high temp that cause it to come unglued in the first place)
      2: perhaps more importantly, the end of the tube had become “smoked” by the burn and was no longer optically clear.
      So, in all the assessment was not worth trying to fix. Especially when a new tube is $100 USD shipped. By the way, new tube cuts like a hot knife through butter.

    2. I am the person who wrote the article and did the repairs. One of the other members tried to glue it back together and still couldn’t get it to work. It is my opinion that it probably came off without us noticing, and that it didn’t block the laser’s path to the mirror, so we kept using it and using it not knowing there was a problem until one day it just stopped. The output mirror/lens was very black as well, so I was fairly confident that it was not recoverable.

      I found that lifting the right side of the laser cutter about 20 degrees into the air for 20 seconds got the remaining air bubbles out. For the last 2 weeks that has been my habit. Turn on the power, let the water flow for 2 minutes, then lift and hold. I am designing a new workspace for it which will include placing the water tank above the water inlet/outlet level, but until then.

      1. Adding something to lower surface tension will help with bubbles. A wetting agent like Photoflow or Calgon or a few drops of shower hair conditioner, for example. Any of the additives used to make “wet water” should be perfect. They will all prevent sticky bubbles that cling to the walls of tubing.

        1. People who drive old cars (1920-1950?) often add a product called Water Wetter to their engine coolant to help keep engine temperatures down. Maybe it has to do with reduced bubbles, I don’t know, but I don’t doubt what you said.

          1. From what I’ve seen from side by side demonstrations and double blind experiments, that stuff does exactly nothing. In fact there’s some reports of it running copper parts in the cooling system.

            Automotive additives are a nasty, nasty game. For each propriety potion that’s legit there’s three hundred brands of snake oil. Watch out and remain skeptical until you see overwhelming proof.

  2. I really doubt bubbles were the issue, these lasers are just cheap and the epoxy broke down. They are really crappy tubes, I mean what do you expect for $100 shipped? Really on these smaller tubes you dont even need to cool the optics. Probably the tube just failed from age. These tubes are what is called a soft sealed tube, the optics and vacuum are sealed by epoxy and the helium in the tube migrates though the epoxy and things go out of whack in the tube. (Helium will still escape the tube even though the inside of the tube is at a lower pressure) The tubes also dont have a catalyst in them that returns the CO into CO2.

    These tubes actually go bad sitting on a shelf. I had bought one for a project and never used it. I tested it every once in a while and the power dropped over the course of a few years until it was useless.

    One last thing is watch the current the tube is running at. Many times these lasers are not set up to limit tube current and the power supply just drives them full on which kills the life of the tube as well.

          1. Sure, and I’m an extraterrestrial ai with the ability to see the future. You can’t tell on a forum. You claim that it doesn’t even need cooling when it’s clearly scorched and then start going on about unrelated stuff like helium leakage.

    1. You ought to read the fault description and analysis before commenting.
      It’s a known fact that the tubes get worse over time. But this is not limited to “cheap China tubes”. Every book I have ever rad on lasers going back to the late 1970s mentions that.
      And we had already noted some weaker power over time.
      In this case it was working, and then it wasn’t.
      Inspection shows that the lense cooling envelope had come off, that both mating surfaces were blackened, and that the tube only ever filled about %50 with water. Literally top half of the lense in air bottom half in water.
      Our mistake for sure.
      But this has nothing to do with age of the tube.

    2. K40 has a knob to limit current. And a display showing the tube current. And the controller mounted in this case is a Smoothie board with variable current control. So the tube current has several layers of control.

  3. This happens to that face covering the half mirror all the time . It’s very annoying but most of the time you can just glue it on after making sure you have cleaned it up … Use a bit of Epoxy if required!

  4. I’m guessing you don’t have one. The display does not show current unless it is a true ammeter, which most aren’t. It’s just some arbitrary number 0-100, where 100 is usually far more than the tube’s rated current. And no K40 comes with a smoothieboard out of the box that I’m aware of.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.