The K40 laser – Part 1 – commissioning

This is not the usual sort of blog post that I make on this site. Normally, those posts are restricted to the use of electronics to enable some or other aspect of photography or just to photography or electronics alone. In this case, the post is about my first impressions of a K40 laser cutter and my first attempts to ‘sort it’! I bought this 40W CO2 Laser Cutter from Vevor here in France. The term ‘K40’ is a generic name given to a range of cheap Chinese laser cutters. Since they cost about 350 Euros, they are very….very….very cheap for what they are. If the machines were made in the USA, UK etc. the price would probably have a zero on the end. However, the fact that they are cheap really does show! The manufacturing quality, quality control and the optical and mechanical alignment on delivery, range from fair to terrible. That said, the results of others show that with some tweaking, the machine can be made reasonably accurate and extremely effective. I should say right from the outset that this, Part 1 of what is likely to be a 3 part post, is based on my initial impressions of the machine and that as yet because I don’t consider it safe (!), I have not even fired the laser. I won’t do that until I consider the machine electrically and optically safe. While I know nothing about laser cutters, I have fair experience with lasers as components of complex optical systems such as confocal laser scanning microscopes and other bits of scientific instrumentation.

This photo may help you to identify whether or not it is the same machine as you have. Note what I say about the machine being produced in different factories and changes that may have occurred over time.

Out of the box, it is pretty clear that some of the parts are nothing short of rubbish and that is where they should probably go (or better, into the recycle bin!). The machines are intended for cutting/engraving over only a near postcard sized area and arrive with a clamping bed designed to hold small objects such as the heads of rubber stamps. However, with that (useless) bed removed, the cutting area can easily be expanded to ~A4 or larger. I was a bit surprised by some of the YouTube and other posts about this machine, while some were misleading or just plain wrong, based on older specification machines and thus not particularly relevant to later models, many showed a horrible disregard for laser safety and should be taken down for that reason alone. I bought a pair of laser safety goggles designed to block 10600nm CO2 lasers – the machine should never be operated with the laser tube compartment or the cutting area compartment open without the user wearing safety goggles…never. People say that the orange observation window is laser safe and it probably is safe to observe the cutter in action through it without safety goggles but why take the chance? You only get one pair of eyes and a 40W IR laser could take them out in an instant – you cannot see IR light so you will have no protection from your eyelids closing and they would in any case instantly be burnt by the beam.

It is important to know that the construction of K40s varies in detail depending on the manufacturer, the exact ‘model’ and when it was made. Thus, much of the information you can find on-line is out-of-date – even that on what are considered ‘go to’ sites dedicated to the K40. It seems to me that the manufacturers have responded to some of the obvious criticisms of the machine with minor modifications to the frame and chassis. You really need to check your machine matches the description below (or on any web site from which you take advice) before trying to apply any fixes otherwise you are going to waste a lot of time and possibly damage things.

Mechanically the machine effectively consists of 3 parts that I will refer to as the enclosure or chassis, the X-Y frame and the gantry. In my nomenclature, the enclosure is the big pressed steel box the whole thing comes in. The X-Y frame is the rigid(ish) 4 sided pressed steel framework to which the stepping motors and the gantry are attached. The gantry is the arm that is powered up and down along the Y axis and that powers the laser head along the X axis.

This image shows the gantry and the X-Y frame. You can probably see that the gantry is not parallel to the frame. The black bar at the front is the cover for the Y axis stepper motor. To the left is mirror #2 and ‘laser head’ containing mirror #3 and the focussing lens sits on the gantry and is powered along it by the x-axis stepping motor that you can just see on the LHS side of the image. I have taken the bed out of the machine. It is pretty useless, has a tiny useable area and I will replace it. Note the fume extraction vent (grey) at the rear. It protrudes into the cutting area and I will cut it down.

Lets start with electrical safety. I have no idea how electrically safe the power supply and control boards inside the RHS compartment of the chassis are. Probably they are OK…maybe. What I did want to find out right away was how well grounded is this metal box? Voltages inside this box run up to 20000V. That could be instant death so the box needs to be grounded so that the metal work you can touch is at earth potential just like you are. There is a lot of stuff on the poor connection of the earth point on the back of the machine though my machine’s chassis was definitely grounded, it was not via this point.The red socket there is actually insulated from the chassis because it uses a socket that has a plastic mount that prevents the metal of the connector touching the chassis plus there is a thick layer of paint on both side of the chassis. Clearly, for some reason the manufacturer did not want a connection to earth at this point on the chassis. However, the wire connected to the plastic earth socket is connected to the chassis and derives directly from the earth pin on the kettle-style socket on the left of the machine at the back. The web also has lots of articles saying you should not trust the white auxiliary mains sockets to the right of the kettle connections. I do not know about that but on my machine they all have earth connections, again directly derived from the earth on the kettle-socket. I cannot see any reason why the ancillary red plastic earth socket should not be connected to the chassis directly where it penetrates it but it isn’t. I wanted an earth point I could rely on so I threw the plastic socket away, scraped the paint to bare metal on both sides of the chassis and created a ground point at that hole using the bolt – washer – solder connector – shake-proof washer – chassis metal – shake-proof washer – solder connector – washer – nut ‘sandwich’ shown in the diagram below. Note that I don’t intend to connect anything to the earth point I created – I just wanted to know that there is a properly constructed connection to earth for the enclosure. I also checked that the socket the laser cutter will be plugged into had a reliable earth – you would be surprised how many sockets do not! I checked the grounding of the chassis to the earth pin of the cutter’s mains plug with a multimeter and the socket to the grounding stake outside my workshop. 5 Ohms is generally considered the maximum resistance for such a connection to earth – mine was 2 Ohms – the ground is dry at the moment – I will water it.

This is an image taken from the Electrical Engineering Stack Exchange (assumed Creative Commons) and it shows how to make a good earth connectiom to a chassis.
This is a shot of the back of the elctronics compartment of the K40. Note the earth wires (yellow/green) run from the earth pin on the kettle socket on the right to the first auxillary mains socket, then to the second one and then to the red plastic socket on the rear of the chassis. The earth connections all measured as good on my machine but the earth socket is insulated from the chassis. The chassis is earthed somewhere inside the compartment but it isn’t clear where. I made a good earth connection using the arrangement shown above. NOTE – the big red wire is the HV supply to the laser!!! Here lie monsters – 20,000 volts – never operate the machine with the electronics door open unless you know exactly what you are up to.

Now I was happy the machine was well-earthed, I switched it on but only after making sure the laser would truly be OFF. The machine homed appropriately, the lights came on etc. I made sure the laser ON switch was at OFF. I consider that switch unsafe – the travel of the push button is tiny and you could easily mistake OFF for ON. Using a multimeter I checked the lid interlock safety switch went open circuit when the lid was not closed – it did. For now, that was as far as I went with the electrical connections. The electrical compartment on the RHS of the machine and that for the laser were both delivered with bolts to keep them shut. I have removed those bolts and will install two extra safety switches that will cut the power to the laser if the lids are lifted a millimeter or two.

The next thing was to check the alignment of the mechanics of the machine. Pretty clearly there was a significant problem. The gap between the gantry and the bottom bar of the frame was clearly much greater at one end than the other! It was about 5 to 10mm out. However, before sorting this I checked that the frame was square. It wasn’t far out but it was not square. The construction of the machine is such that whether the frame is square or not depends both on the welding and bolting together of the folded pressed steel fame and how the frame has been bolted to the enclosure. There are four 10mm bolts connecting the frame to the chassis. If the frame is little bit out of square (say 1mm or less), as mine was, it can be squared-up by releasing the front two bolts and applying some judicious leverage and then re tightening the bolts. What’s happening here is that the bolt holes in the chassis and frame are a little over-size and that can be used to advantage. I guess the rear bolts could also be used in the same way. Note that there are no washers and the nuts are not captive – you will need 2 spanners. If the frame were more out of square than this I guess you would have to take all 4 bolts out and remove the frame to get it square with some judicious leverage. So with the frame squared-up in X and Y, I checked that the Z axis ran level in the plane the frame. It did, but I suspect I will make further checks on this when the bed is back in the machine because ultimately one wants the laser to be focused on the bed over its entire extent.

Checking that the X-Y frame is square. The little bit of blue light that you can see seeping through along the bottom edge of the square is because the frame was not quite square.

With the frame reasonably square it is possible to sort out the gantry to ensure it runs at right-angles to the frame. As I said above, mine definitely did not. There is an easy way to make sure the gantry runs true. I removed the two bolts connecting the black Y axis stepper motor cover to the XY-frame. Under this is a motor with shafts that connect the LHS and RHS belts that drive the gantry on the Y axis. The motor is connected to the belt drive on the LHS via a long shaft. There is a crude rigid coupling connecting the motor shaft and the longer drive shaft. It has 8 rather crappy screws. I undid all of these (you might just have to unlock 4) thus uncoupling the LHS and RNS belt drives. This done, you can rotate the Y axis motor shaft until the gantry is square to the frame and re lock the screws doing your best to get the shaft central within the coupling. After doing this, everything was within about 0.5mm of true.

The Y-stepper motor has two output shafts. The LHS one is coupled to the drive pulley for the LHS end of the gantry via a long shaft. Undoing this coupling allows one to move one end of the gantry independent of the other and to bring it square with the frame.

Why was the gantry so far out-of-true? It is possible that this was because the belt on the RHS was lose and the drive pulley had skipped some of the notches on the belt after being bounced about in transit (but see below re: optical alignment). Either that or it was just how it was put together (more likely – see below). It is relatively easy to change the two Y belt tensions. There are two holes in the back of the machine that allow you to access the tensioning screws, a process made much easier if you put a small torch in side the machine to illuminate them, get your eye level with the holes so you can see to position a screwdriver. The RHS tensioner on my machine seemed a bit stuck but I was able to get the belt reasonably tight – I figure you should be able to depress the belt about half the gap between one side and the other using a light finger press. You can tension the X belt by positioning the gantry such that you can access the tensioning screws (not the Allen-headed bolts that hold the end of the gantry to the slider!) via the big hole in the LHS of the power supply compartment.

Sticking a small torch inside the main compartment of the machine makes it easy to spot the tensioning screws for the Y axis drive belts.
The tensioning screws (cross-heads) are accessible through the hole in the wall between the electronics and main compartments of the chassis. Mine did not need adjusting.

With these preliminary mechanical adjustments done and without which there was little hope of aligning the optics, I turned my attention to the light path. Just eyeballing it, it was pretty clear there was no way the light from mirror #2 on the gantry would ever hit mirror #3 in the laser head because that head was clearly twisted through several degrees. Given how tightly it was screwed to its mount, I doubt that this had shifted in transit, rather I think that perhaps an attempt had been made at the factory to get the laser beam to hit the laser head by turning it in its mount. This makes some sense given how out of square the gantry had been to the X-Y frame. Classically, to align the light path you use tape over the mirrors to see where the laser burns holes in it (the ‘burn’ technique). I had absolutely no wish to activate the laser while it was so clearly miles out of alignment. Instead, I designed and 3d-printed a laser alignment tool so that I could use ‘the reverse alignment method’ (my laser alignment tool is available to download on Thingiverse). This technique involves shining a red laser back along the optical path to the laser tube. In theory if you set up the mirrors so the light lands back at the exit of the laser tube, then the CO2 laser beam should retrace that path. However, there is a flaw in that argument, the laser beam may not enter the 1st mirror at the same angle as the laser pointer alignment beam left it. As a result, you might end up having to rejig the laser tube’s positioning to get the angle right. In practice that is a pain and the ‘burn’ technique is more practical. However, this does not mean that the reverse alignment technique is as some have claimed ‘useless’ (‘a solution looking for a problem’). It will get you from a hopelessly out-of-line situation to one in which you are not too far off. My reverse alignment laser showed things were as I had expected, hopelessly out of line.

The 3d-printed laser-alignment tool in place on the laser head less its lens and holder – you don’t want the lens in the head for this process. I could add – “bingo!” – the laser takes the correct path from the laser head mirror to mirror #1 in the laser compartment.
The laser head disassembled and cleaned up. The quality of the parts is pretty crappy. You can see two burn marks on the mirror that must have resulted from the machine being tested.

My first steps were to completely dissemble the laser head, deburr the mount, clean up the turned parts, the 3rd mirror and the lens. I could see where the CO2 laser beam had struck the mirror so clearly someone at the factory had at least tried to align it! With the laser head cleaned up, I cleaned the other two mirrors – they were all filthy and were cleaned with propyl alcohol on a cotton wool bud. Next using the reverse alignment tool, I aimed the laser pointer beam at mirror #2. It would not hit that mirror but rather always fell below it. I decide to ‘shim up’ the LHS laser head mount hex spacer (see picture). One turn on the hex spacers was enough to bring the laser to the centre of the mirror. A few tweaks on the adjustment screws for mirror 2 brought the beam onto mirror 3 and minor adjustments on mirror 1 brought the beam into the centre of the exit from the CO2 laser tube. Actually, centring doesn’t matter much. What is important is that the beam travels parallel to the X and Y axes and does not thus move when the laser head is moved from a near to a far point and the same for when the gantry is moved from close to far from mirror 1. There are some good YouTube videos on how to achieve this (do a quick search to find one you like) so all I will say here is that you start with masking tape on mirror mount #2 and get the reverse alignment spot steady on that mirror at all distances, move the tape to mirror #1 and arrange for the same there. Now, the machine IS NOT aligned for the CO2 laser but it probably isn’t far out and it is time to use the ‘burn’ technique. More of this in Part 2 of this blog entry. My conclusion at this point is that the optical components are all cheap and nasty and at some point better mirrors and a better lens will be required. They are ‘functional’ but no more than that. However, the laser tube looks good!

This is the laser tube. Not something with which to lightly mess around. It isn’t as easy to change the alignment of the tube as it is to alter that of the mirrors, Hence, doing a reverse alignment with a red laser is not a complete solution.

The next post will deal with firing up the laser, getting it properly aligned, sorting out the bed and some other things.

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 3D printing, laser cutting and tagged , , , , , . Bookmark the permalink.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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