I spent a lot of time writing this article. I actually preferred to prioritise construction over documenting at the beginning. The machine must be completed quickly enough and the writing takes a lot of time!
These articles had to be structured according to the different parts of the construction, however as the original Instructables is full of errors and as I spent more time correcting things than moving forward, I will have to change format.
Now, I will tell the story of the building by following the chronological order of taken pictures. And in a second time, I will create a PDF where the entire process will be detailed (with the files at the end). It will be easier to set up and especially more useful for you!
I installed the linear rails and the X axis. The rails of the axis Y are centred on their profile to leave the necessary space on each side with the elements type motor, end-stop, pulley. Then I was able to mount the laser, the cable guide and the compressed air pipe. Due to the fragility and inaccuracy of some parts, I took over the 3D and I changed them (laser head support, cable guide support, X mirror support). Everything will be made available at the end of the construction.
A word on the door and on the actuators. The Instructables provides actuators supports in 3D printing. They explodes under the necessary force to actuate them… And if we look at the photos, metal supports appeared. No trace in the text, in the list of materials and no indication of where to find them. I tried to replace the actuators with a lighter version with metal supports without success (it does not hold at all the door in the air). So I will cut into a 3mm steel plate custom-made brackets for the original actuators.
For the bottom, I chose a 12 mm plywood plate for price, stiffness and weight. To be able to fix it properly to the frame, I printed a drill guide (for a good alignment with the aluminium profiles and a good spacing between each hole). Small PLA buffers have been added here and there to avoid scratching the surface were the machine will be.
The tank consists of a 100 mm PVC pipe with two stops glued on both sides. Two holes are then made to insert the water circulation pipes. Another small detail forgotten about the original BOM: consumables and special tools. Example here with the PVC glue which is essential to fix the ends.
I took what I had on hand in the usual DIY store to make a conductive backing plate for the electrical / electronic parts of the cutter. This will be the brain of the machine (or motherboard). The idea is to have a conductive support which would be connected to the mass of the domestic network to avoid current leakage (steel would have been better for cost and rigidity, but I did not find it at the store).
The purpose of this step was to position the elements according to the diagram made previously. I wanted to separate the high voltage (220V) from the low voltage (5V and 12V) to avoid interference and for an easier handling of the cables.
To connect some elements in parallel, I made small maps. Then to position the cables correctly I printed chutes in PLA (very perfectible … but I did not find the parts displayed on the Instructables). Everything is slightly elevated when there is no need for contact with the earth.
Once everything is positioned, you have to drill! And for the fixing screws of the plate on the frame, and for the fixing screws of the cards on the plate. And then we install everything. Again, what is presented here is not the end result and will be changed depending on the progress.
There are two sets of fans to install: first batch to cool the box that will contain the motherboard and laser driver, second batch to extract the toxic fumes and cool the water of the laser. Both are fans connected in parallel on 12V.
The goal is to properly light the workspace (and to add a finished aesthetic look to the machine). I decided to position a bar below the X axis to be able to precisely monitor the advance of the cut, and then all around the cutting surface. Unlike fans, LEDs are connected in series. For a reason of practicality in the wiring, I installed two lots: a lot that goes around the frame, a lot that is under the X axis. The two lots are connected in parallel.
After an approximate positioning to choose my fasteners, I adapt these (holes too small). You can see here the first iteration of improvement of the axis X besides. Then I put the cables clean so that they are not in the middle and / or do not interfere with the movements of the machine.
I finally received the plates! A big delivery that took up a lot of space. First and foremost I wanted to install the plate which seemed to me to be the most painful … As the holes to fix the radiator were missing on the plans, I positioned the radiator in function of the fans and I made the holes. Then I realised with great pleasure that it was impossible to install this plate now that the frame was finished … Yet the previous author insisted that the assembly was done in this order … So I dismantled a profile and I was happy to realise that it was also necessary to install this plate before the bottom plate (again, contrary to what has been explicitly requested) to have a chance to screw it!
It is no longer possible to work on the ground for obvious reasons. So I decided to install (by myself …) the machine on the table for easier construction! And now that it is very rigid, it is OK to have it on a smaller surface.
Here is the installation of the motherboard on the frame. And there are also the screen, the buttons and some of the Plexiglas plates. I also installed a little piece of wood to be able to work on the electronics later.
CO2 tube and cooling circuit
I set up pump, pipes, laser tube and fans. Do not rely on the photo, the Plexiglas plate under the tube must be on the other side of the profiles (this will be corrected later). The pump is attached externally for testing. First observation: the silicone hose requires riselons not to jump. Second observation: the pump sucks too hard, which crushes the inlet pipe and the flow is greatly reduced (not to mention the horrible noise). For the second report I ordered a hose to replace this section. Well, that solved the problem! But suddenly my pump runs at rated speed, which is much too powerful for a circuit like this … so the silicone hose exploded under pressure … So the circuit was never sized, it’s is the combination of eBay’s first 12V pump with the first AliExpress 8mm silicone hose, it’s fantastic. I ordered a much less powerful 12V submersible pump based on research. The idea is to maintain a correct cooling while being compatible with the current system. And as I was tired of drying my wood with every leak or explosion, I installed a silicone under-film just below. I also took the opportunity to install the first mirror.
Side panels and carbon
Still for questions of rigidity I chose plywood 12mm. It was enough to make the cut at 45 °, the installation of the vinyl carbon (I wanted it to be pretty) and holes for fixing. Then as I did not have the means for a folding Plexiglas on such a huge dimension, I ordered the cut plates rather and I made the installation with silicone for the holding and vinyl for aesthetics.
Evacuation of fumes
Two crucial points to work for the evacuation of toxic fumes: the extraction area and the tightness of the pressure chamber. The fans draw air from the work area to the radiator area. There is therefore a depression on the side of the working space and an over-pressure on the side of the radiator zone. The work-space does not have to be airtight for two reasons: the first is that the airflow would be from the outside to the inside, and so I have no risk of leaking toxic smoke, the second is that I have to bring air in order to create a stream.
To adapt to a standard extraction pipe size, I designed and printed a suitable adaptor (with grooves to attach riselons that will allow a good fixation of the pipe). And then, I have silicone all the over-pressure zone. And then, as recommended, I installed a joint strip between the rear Plexiglas plate and the aluminium profile. Of course, by screwing I apply too much pressure on the Plexiglas and it breaks. I will change later that part that does not suit me at all.
We can see the installation of various components concerning the X and Y axes. First major change compared to the original: the geometry of the elements that hold the belts. I do not know how he managed to apply tension, but with the original media, it broke for nothing.
To maintain a timing belt in place, there are several strategies. The one set up here (even with the previous author) is to go around a pin and embed itself. The previous pin was very thin and without reinforcement at its base. Here I opted for a thicker picot, with a slightly different geometry and a rounded at the base. All this helps to strengthen the whole.
Then I designed two supports: one for the cover sensor and one for fixing the Y-axis cable guide.
Installation of the mirrors on the supports. Again, no instructions on that, I had silicone joints that I did not know how to use … So I just unscrewed the elements and make the installation as it seemed consistent. I invite you to look more at this point or to tell me what you think if you know things on the subject!
The original feet are unstable, fragile and pivots. So I made new ones which are much more stable.
It’s time to properly tidy up all the cables in the work-space. The small parts in PLA are supports that I created but I am not pleased with them …
Water flow sensors
I added two sensors for the cooling circuit: one tells me if the water circulates or not (and directly cuts the laser if it is not the case), the other tells me how much it flows (rather for surveillance). Of course I bought sensors without adaptors … So I made copper adaptors for the silicone pipe.
It’s this point that definitely convinced me of the relative quality of the original Instructables article. Indeed, I went to weld the cables of the laser driver to the laser tube and I needed to know where was the anode and where was the cathode (to connect the red and the black to the good terminals). I find a good diagram on the net and then I said to myself “oh yes, there probably is a picture of it on the original post!”. There were, and it showed the opposite of what I found. I told myself that I was going to search, and after 10 examples confirming the meaning, I was able to conclude that the author left on his guide, a branch upside down (connection to the most expensive element of all achievement). At best nothing happens, at worst the laser goes to the other side where the tube is ruined.
So it’s easy to know where the cathode is connected: on the side of the glass propeller. And the anode plugs on the side of the laser output. I advise to strip 1cm of wire, to wind the strands around the graphite bar and to bring a generous dose of tin. We must feel resistance when pulling (slightly!) And then cover with the retractable sheath.
I also wired the water flow sensors and decided to put labels on all my cables so I could easily know what I was handling when I plugged in the motherboard.
The laser tube operates optimally between 15 ° C and 35 ° C, so I need to install a temperature sensor in the coolant to monitor it. The LM35DZ sensor will do the trick, but I have to make it waterproof. The idea is to make a copper shell and then seal with silicone.
Electrical, electronic, motherboard
I started by installing the 220V. The arrival is made by a European C13 cable in a box that contains an ignition switch and a fuse. The fuse is located on the phase and the switch is on the phase AND on the neutral. I started by connecting the ground to my aluminium plate and then paralleling the power supplies. Then I installed my neutral and my phase on the emergency stop button (be careful to connect it in NC “normally closed” and turn the NO to NC to have a protection on the phase AND on the neutral ). The RPI is supposed to be fed in 5V by the stabilised 5V power supply, but this one is of too bad quality and does not hold the tension on a high requested intensity … I was under-voltage all the time ( connection via a transformed micro-USB cable). After measurements, I realised that if I adjusted the power supply to 5.3V, it went down to 4.6V after switching the Raspberry ON. So I ordered an official power supply that I paralleled on the 220V circuit. The air pump is also connected in parallel to the 220V with the phase that passes into the relay.
The idea and use the RPI to manage the relay and view some system statistics. So I have to plug the items on the GPIO of it. That’s the card I made for the connection. I checked the compatibility of the RPI with the elements of the system before that. And it will feed the temperature sensor and water flow.
Installation of the various elements including the engines. I use a special clamp, but I’m not satisfied with the electrical contact that offers this connector (but maybe I’m doing bad …).
The M2 engine is coupled to the M3 engine, so I have to weld on the board a bridge between EN2, DIR2 and ST2. Of course, M2 and M3 do not turn in the same direction … so I reversed the wiring of M3 (M2 -> BNVR – RVNB <- M3).
The original author opted for a Y limit switch upside down, I went back to the place for simplicity and consistency.
BONUS: presentation of the project at Fablab
The Fablab Platform C asked me to present my cutting project in their premises. The idea for me was to convey the story behind this realisation rather than a thorough technical presentation. It was a very nice moment!
Encore une fois, n’hésitez pas si vous avez des questions. Vous pouvez m’écrire par commentaire et par mail !
So much info in this article. I think I’ll start from scratch with the V2, now that I have complete control of the machine. Because there are too many errors and inaccuracies on the original article. And I would like to add the ability to easily increase the power of the laser while maintaining the same frame. Why not also opt for an industrial cooler after seeing the cost of these!
What will be next :
- steel back plates
- new water pump
- SmoothieBoard setup
- RPI interface
All the articles regarding this series are here : DIY Lasercutter