The K40 laser – Part 2 – upgrading (aka making it work at all!)

It soon became apparent that the K40 was not going to be that useful unless several upgrades were made. First of all, the useless bed/clamp thingee that is supplied with the machine is way too small. It might be OK for tiny jobs but the gantry of the machine is capable of traveling over a much greater area. To enable the use of the whole area covered by the movement of the gantry, the old bed has to be completely removed. This done, some people then prop their work on a lab jack. This will enable the machine to function over approximately an A4 sized area and also allows objects of different thickness to be placed where one wants with respect to the focus of the laser. My solution was rather different. I bought a piece of perforated steel plate that is pierced over its entire extent by 8mm holes. Using the aluminum from the bed supplied with the laser and using the steel plate as a template, I drilled an 8mm hole at each corner of the plate and used these to mount 4 pieces of 120mm long threaded rod. These were locked in place with nuts on either side of the aluminium plate. I then placed some washers over these along with 4 springs. The steel plate was then slid over the screwed rods and nuts placed on each of them so that by screwing them up or down one can easily alter the height of the bed. In my initial experiments, I found that unless the material being cut is raised from any surface beneath it, you get a lot of staining from the tar etc. that is emitted when it burns. Thus, I decided to use small N52 neodymium magnets to raise the material to the correct height and stand it off from the steel plate. More magnets on top that are attracted to those beneath, hold the work-material flat. The whole setup is pretty versatile and the height adjustments can often be made by using 2,3 or 4 of the magnets beneath the material. If that isn’t enough, the bed height can quickly be adjusted with a spanner to any height one wants. To aid in setting the height, I cut some plywood gauges. The whole bed can easily be slid in and out of the machine. Care is required to align it (blue arrows on the Y axis) so that the laser head cannot hit it or the screwed rods. I will add some plywood rings to locate the bed on the chassis such that whenever it is removed, it can be dropped back in exactly the same location.

New adjustable laser bed in which the nuts on the screwed rods compress the bed against 4 springs. Note that the plywood here is held off from the bed by stacks of small N52 magnets and more on the surface act to hold the ply flat. The new laser head and air assist (blue) are also shown. Note also that the smoke extract vent has been shortened by about 2cms.

At present, my smoke extract is to say the least, ‘suboptimal’. While the new OEM fan that has replaced the old slide-in bathroom extractor setup is far better, I have it attached to 80mm diameter ducting. I wish I had made an adapter and used higher diameter ducting.

A second modification is required in order to use the whole area covered by the movement of the gantry; the smoke outlet duct needs to be trimmed back by about 2cms. I did this using a Dremel cutting disc and without removing the duct from the chassis. The reason I cut it in place is that like all the fastenings on the K40, none are captive so in order to remove and refit anything, you need access to both side of the sheet metal and I’ll say it again, the quality of the nuts, bolts and screws is truly awful! You will find statements on the web that the laser power a little way beyond on the focal point is too dispersed to do much damage. That is not true either where your eyes or where the big hole to be found in the bottom of the K40 chassis, are concerned. I ran the cutter and found that I had set alight to the bench under the machine! I think the big hole may be necessary for good extraction (I am not sure on this) so I placed a piece of aluminium sheet beneath the machine to protect the bench. I should probably emphasize here that a K40 should be kept under constant watch – never leave it running when you are not there and keep a fire extinguisher handy. To further improve extraction and so that the fan on the K40 has less work to do, I laser cut the parts for an in-line fan box to go near the exhaust point and installed an old 18W 12V 120mm fan I had lying around. Finally, I sealed up any leaks on the outlet side of the fans – mine leaked a lot from the outlet side of the K40 fan – I sealed the gaps with silicone.

The booster fan housing. Probably not a good idea to use 80mm plastic tubing – too narrow and also in my case, way too long but the booster fan helps. Note to self: plastic burns!

Next up for attention was the provision of an air assist. Blowing air over the point impacted by the laser creates a much better cut, putting out the flame caused by the heat of the laser, it greatly reduces charring, increases cutting efficiency and reduces the width of the kerf (the thickness of the cut line). My first attempt at an air assist consisted simply of a 3D printed clamp holding a piece of 6mm aluminium tube bent so that air from my shop air-compressor was directed at the focal point of the laser. While this worked really well, it turned out to have one major disadvantage – because the air is coming from one side, the air assist is more effective when it is chasing the laser when it is moving in the direction of the air flow. Though such a setup is recommended by a number of K40 aficionados, if you use it then the kerf will be different in the two axes. If you need parts that are as accurate as possible, the difference in the width of the cut line is significant – about 0.5mm. The way round this problem is to direct the air-flow in line with the laser beam. There are several designs for 3D printed air-assists that do this available on Thingeeverse. However, I chose to buy a new laser head from CloudRay. This had a couple of advantages; the laser head came with a new lens and mirror to replace the existing low quality ones supplied with the K40. While I was at it, I decided to buy a set of 3 new gold-on-silicon mirrors. Thus, mirrors 1-3 were replaced and I had one as a spare.

Close up of the new CloudRay laser head and air assist which blows air vertically down into the cut thus avoiding assymetries in the kerf. Note in order to keep the lens at the same height was with the OEM head, the CloudRay replacement head needs to incorporate part of the old system (the shiny aluminium tube).

The new lens and mirrors were a revelation. The power required to cut things fell by about 50%. The new air-assist head ensured that objects that were meant to be square came out square. A further revelation was the discovery of the ‘ACD method’ of aligning the laser (see on YouTube). Once you have understood this method, alignment becomes a doddle. It is THE K40 video to watch.

Next, I decided it was a good idea to be able to keep an eye on the cooling water flow and temperature. To this end, I bought a cheap dual thermometer from AliExpress and also a flow switch. I simply taped the temperature senders to the inlet and outlet tubes for the laser and then encased them in pipe insulation. The thermometers are very accurate and you can easily measure the difference in the temperature of the cooling water on entry and exit from the laser. I have wired the flow switch to a flashing LED/buzzer to warn if for any reason the cooling water flow should stop. Since my workshop is cold in winter, I also bought a cheap aquarium heater and placed this in the cooling water reservoir. For cooling, I used de-ionised water. Deionised water is fine – you can find a lot of nonsense about how you have to use distilled water – it’s nonsense, either will do.

Flow switch for the cooling water with two 3D printed barbed hose-couplings. Loss of water flow would be catastrophic for the laser tube!
Temperature monitor with the sensors mounted on the inlet and outlet tubes. Word has it that the water should not be at less than 15oC or greater than 25oC. I have an aquarium heater in the coolant tank to keep the temperature above 18oC. This guage will alos be mounted on the control panel that is to come! By having the aquarium heater and the circulation pump wired to thermostat I can probably arrange to frost protect the laser tube…work to be done!

While on the subject of nonsense, there is a great deal of misinformation concerning the technicalities of using a K40. The 1st one that I would emphasize is the notion that it is safe to operate with the lid open. There are dozens of videos of people doing this on YouTube. Just don’t do it unless you are wearing eye-protection – think about the burns in my bench caused by the defocused laser and imagine the beam hitting a reflective surface and bouncing into your eye. Other nonsense; there are videos suggesting the laser head lens goes up the wrong way – it goes in flat side down. While on the subject of the lens there are warnings on the web that the lens is toxic. It is, but I wouldn’t worry to much, ZnSe is toxic but the LD50 is 5g per kg. That is no reason not to treat it with care and if you must use your fingers, wash your hands afterwards, better still wear gloves, but even if you ate the lens it would not kill you (don’t try that!).

Next, I decided that the potentiometer supplied with the machine doesn’t really provide much accuracy when trying to set the laser tube current. I swapped it out for a 5KOhm 10 turn pot with a counter (see pic). This allows for much greater precision and reliability in setting the laser current. Eventually, I will add a voltmeter to monitor the control voltage to the laser and give a more visible display of the power setting. Note that on a 10 turn pot, the ‘wiper’ is the terminal at the end of the potentiometer not the one in the middle as it is for a normal pot. Further, when wired up test everything is OK with a tiny blip on the laser test button. I suggest that because I was supplied with a faulty pot and it cause the laser to run at max output.

10 turn pot to control the laser control voltage along with a turns counter. The lettering on the counter is too small! It will work along side a voltmeter on the panel that is to come.
Wiring for the potentiometer.

With the ‘upgrades’ above, the K40 will cut 3mm baltic ply with only 9mA of current at 10mm/S. The edges of the ply are not horribly burnt. The top surfaces are clean, as are the back ones. This is a major improvement over the results I obtained from the unmodified machine (14mA at 6mm/S). When I measured the width of the kerf, it was practically zero and a 50mm square piece of ply was both the right size and truly square. I have one remaining thing to do, and that is to fix the broken Y-tension belt adjuster. For that, the whole gantry has to be removed so, my next post will be about what can be tuned when that is done and the addition of a new control panel incorporating temperature and control voltage monitors.

The kerf is now less than 0.06mm and everything is square making it possible to do nice finger joints! A good tip (I can’t remember who suggested it) is to cut the fingers a tad too long so that the burnt wood can be sanded off.

About petermobbs

A tinkerer and maker of 'things' who loves macrophotography and the natural world. Retired scientist and former Chair of Physiology, Dean of Preclinical Medicine and Dean of Life Sciences at UCL, now enjoying the beauty of the Gresigne in South West France
This entry was posted in Photography and electronics and tagged , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s