cnc machine for shaping large pieces of polystyrene foam
by：QY Precision 2019-11-25
Summary This manual shows how to form polystyrene foam using a CNC machine. Starting with a brief introduction, the design software needed to create STL files is discussed. Next is the way to create G- Use the STL file as the code for the input. After that, we introduced GRBL, which is an arduino-based use G- Control the code of the CNC mill stepping motor. The step driver is introduced, and the creation of large CNC hardware is described. The next steps are completed. Enjoy it, Richard tegebecks PS. I plan to expand/update this note as I gain experience and make improvements to my machine. Come back and see if you\'re interested! Also, please leave a comment if you would like to know any steps more clearly and I will try my best to help. When I take photos/videos of almost everything I do, I may add more pictures. Shortly after being disappointed by a 3D printing company in the UK, I became interested in CNC machines. Interested in reprap efforts, I decided to have a drill Just as the tool slides in three directions, this will be an easy target to achieve. I did my research and ended up slapping a small machine with an old flatbed scanner, an old printer stand and an abandoned CD player. I learn from experience and use it to achieve the following goals: to make a large 3-axis CNC machine that can process polystyrene foam. The processed foam can then be covered in a protective layer (eg. Water-based latex) Before pulling the composite fiber mold (ing)from it. To design something on a 3D computer ,(CAD) Software needs to be designed. This may not be pleasant for a favorite apple person or an open sour apple person, but the system to be used (in 2012) Microsoft\'s old Windows is so boring. Sorry, no discussion. . . Made a democratic decision for Windows ( Only one voter is allowed: me. . . ) , The product of CAD application is very rich, if you study hard how to use the software package, I suggest you do a lot of research. If the budget is a limitation, it\'s hard to ignore Autodesk\'s 123D. It seems relatively easy to learn because it is free so can\'t be clapped. Note that as it is still in the testing phase (June 2012) It seems to be a hassle on the old, lower spec32bit system. The 64-bit version should be very good on a more powerful machine. On a professional level, the price of software has risen sharply, but people are paid for it. . . Autodesk\'s Inventor and Dassault\'s SolidWorks are popular products. Both are powerful and take a lot of time to learn. In short: as long as you can generate STL files, you can use any CAD software you like. It is worth mentioning that all the software mentioned in the following steps can be provided as open source or free software. . . Once the STL file appears, G- Create a codec. The program I am using is called Freemill: very easy to use, but it may cause some problems with your graphics card. If locked, try the following: redo the steps until the last step before locking again. Now narrow the program window or move it out of the screen to prevent the 3D model from being displayed on the screen. The program should work properly now. . . Grbl (\'garble\') This is great than any other reason: it allows the use of USB to connect the CNC machine to the laptop. Many established hobby solutions take advantage of outdated parallel ports that do not work properly on laptops due to power management issues. . . For Grbl, see: users can connect to grbl using the following: practical terms, grbl is running on arduino (uno) And it needs to be uploaded to the micro controller. There are several methods, and a good description has been described here: Support (Open Source) Arduino people, buy an original. . . On my original CNC machine, I use a small printer motor with a 12 volt power supply and a pololu drive. A lot of information is provided by Reprap guys. I need more power for my large CNC Machine: 24 V and larger drive. I tried some drivers based on the Toshiba TB6560 chip but I didn\'t make it. Since I only played for an hour or two, I might still get them to work before turning to another option. Suggestions are welcome. . . Update (May 7, 2012) : The driver of Toshiba is garbage. The first one didn\'t work, the second one burned, but the third one worked very well. I think the assembly quality may be poor: when boardsfirst arrived I noticed that the solder joints I had repaired were poor. Based on the same chip, I will get new drivers and even make them myself. All in all, this means I can\'t run the machine yet :-( I used 50x50x2 for the basic framework. Box Section 5mm. While I already have most of the other steel for this project, I need to purchase four lengths for the Box Section. Not cheap. . . To align the frame correctly, it is important to have a flat surface. It\'s better to have a thick and steel table. In my case, I use the table I made with the old door. As the frame was too big, I was lucky to get some help from my partner Kirkthanks again! Another aspect of frame/machine size is moving the machine after completion. Therefore, the wheel is connected in this step. The key part of this phase is to have the top members parallel. I did spend the whole night in order to get the results I wanted. For the track I redesigned The round shaft of a broken 0 printer was used. It came from local tips, initially it turned around for 5 and stayed in my partner\'s house for a few years. . . When the printer is opened, the drive belt disintegrates. It just confirms that it\'s a good idea to avoid not doing this. Standard components. In terms of the round shafts themselves: if they need to be replaced, I can do this by getting 16mm OD silver steel. The round shaft will position the slide frame in the side direction. The second axis is not Position or float in this direction. Therefore, it only needs to reach a high tolerance directly in the upward direction. In this way, the cold-rolled box section of 20x1. 5mm of the railway is my choice. For these two tracks, I weld them as accurately as possible. By using the gasket *, the final adjustment of the high tolerance can be realized later. * I make my shims by cutting freshly emptied beer cans; The thickness of this gasket is very constant, at about 0. 1mm in size. I use the flat surface of the desktop again for alignment. In order to stop even a little bit of deflection, I supported 50x50x2. There are about 25x25x3 boxes of 5 boxes of parts. At the positioning end, I installed a pillow block with a linear ball slide with an opening. I installed roller bearings on the floating end. Please note that only one pillow block is used to maximize the use of the round shaft. This can be achieved by having a longer floating guide rail with a pair of bearing groups. Shortly after putting the main carriage on top of the base, I noticed some issues with the ball slide. Some of the balls stopped spinning and ended up hitting my lovely shaft. Dust must have blocked the runway. My conclusion is that it is better to have a regular bush on the pillow. The most cost effective way to upgrade the pillow block is to make some enclosed bushes to install. The od of the bush is 2mm smaller than the ID of my pillowcase. So I made a 1mm thick gasket. Although some efforts have been made, I think it is worth it. I will also replace the ball bearing for the nylatronsliding block later. I decided to buy a chain drive system for various reasons. There are mainly :- The chain is not stretched so much compared to the belt The chain is very strong- The chain can be easily adjusted to any length At this stage of construction, the simplicity of Assembly requires some precise processing. The bushing needs to be installed at a hole of 10mm. The sprocket needs to be re-welded with 8mm holes. By getting 8mm OD silver steel, I managed to avoid the processing of the shaft. I work hard to produce a tension device that is easy to adjust. To straighten out the long axis on the base, I used a gentle persuader * for adjustment- X mark the scene! June 28, 2012 Update: The chain on the vertical shaft is driven by a sprocket and installed directly on the motor shaft. The other two shafts have two chains per shaft, and the coupling needs to be installed between the motor shaft and the chain axle. My original sofa twisted too much under high motor torque, resulting in a rebound of at least 2mm ( See attached video). The new Spider coupling has finally arrived in Hong Kong and is now in use. The rebound is now acceptable (less than 0. 2mm). * Engineering term spindle stand for hammer is mainly composed of 40x27x1 in the cold rolling box section. 5mm, fixed on the linear slide with bolts. When installed to the sub-compartment, there is no obvious play. I tried to rotate a 315mm long drill bit at 30000 rpm without vibration during operation. In order to be like a ball head knife, once the drill bit is round, it should be great for processing polystyrene. The length of the drill is not used for large cutting; Cutting should be done using only the ends. However, it would be great to create a deep cavity at a shallow release angle. In order to process solid wood, ordinary router bits can be installed. I did not finish the bearing on the floating end. I deviate from my original design when I want to stay away from the rolling bearings and install the nylatron pads. I also have to finish the vertical drive. It will be chain driven, but thread-based drive is also possible. Update: Z-June 12, 2012- Axis complete. I haven\'t found a counter yet. Weight, but no motor seems happy. It just means that if the power goes off, the carriage will drop with a bang. In addition, the 315mm-long drill bit pushed too much. Even at the slightest pressure/speed, once it touches the foam, it shakes too much occasionally. Just a little short. I will shorten the long time step by step and see what I can get. . . Step added on June 12, 2012: as mentioned in one of the previous steps, I did run into a problem with the broken Step drive. To run the machine for the first time from a mechanical point of view, my teammate Russell (many thanks! ! ! ) Take his 3-axis rig. It consists of a window tower and a 3- Shaft parallel plate with Toshiba TB6560 chip. The board is set to use 1/8 subdivision to make the step size from 0. 4mmto 0. 05mm. We also set the gap on the horizontal axis to 4mm per root. Due to the distortion of the coupling, the tooth side gap exists purely. This will be sorted when the coupling is replaced by a spider coupling. Big cars can easily run at the highest speed we dare to run. They barely warmed up and didn\'t seem to skip any steps. On the y- I managed to install the block on the round shaft. I will also be on the axis for x- The shaft, but this means to remove it in order to tap the thread at the end of the shaft. In the first trial, I installed the temporary block by clamping the rectangular box of about 50x25 to the main frame. The drill bit, 315mm long, shook too much. At some stage we replaced it with Dormer a1 10 1/4 \"bit and I rounded it up so it would turn into a ball nose bit. I was able to do deep cutting at high speed. View pictures and videos. . . Step added in June 18, 2012: The ultimate goal of a large CNC machine is not just to make shapes with polystyrene foam, but also to do something useful with a shaped foam. First of all, this will be the creation of composite molds and/or molding. In this step, I will show to create a single purpose polystyrene mold for a custom FRP welding mask. After designing the mold for the new mask, I created an STL file and opened it in FREEMILL. Here I created two sets of G- Code: 1 Rough Cut ( Slice at an interval of 4mm) The next group performs the last cut (at 1. 5mm intervals). The final cut is 30mm deeper than the rough cut. Rough machining takes more than an hour to complete, and the last cut takes about two and a half hours. In fact, the final cut creates obvious steps in the part of the mold. This will be improved by using 0. The interval is 5mm, but it takes three times. Seven and a half hours. Instead, I quickly polished the steps with a piece of sandpaper, which took about three minutes. . . Before being able to use a mold to make a FRP part, it needs a protective layer, in my case latex paint. I mixed a little blue in the paint sample into the plain white latex paint and used it for the first layer. I then added a layer to the white latex until the blue doesn\'t show up anymore. In this way, I am confident that the protective layer of the paint is thick enough. June 20, 2012 Update: The Mold needs to be waxed and polished. I prefer to do this with honey wax, which is what bees do. The three layers are usually good enough for a shot, and I did apply the four layers to determine. The first part of the wet shop- Up is usually the application of two-layer gel coating. I used two layers of rather thick glue coat with a red pigment mixed in it. Having a thick layer will allow the parts to be polished later to remove any defects on the surface. One or more layers of fiberglass soaked with laminated resin follow the coat. In my case I used two layers of mat and laminated resin and mixed some white paint. After the resin is hardened, the parts can be pulled out from the mold. It can then be trimmed down, polished to remove defects on the surface and polished to shine. In the case of welding the mask, I need to cut a rectangle for the screen. The actual screen and screen wrap will be cut from the \"donor\" mask and combined to the inside of the FRP case. The assembly will be completed by connecting the handle to the housing.