another 3d printer has arrived in my workshop. it's a cr-10 and same as my previous printers i got it from the online dealer gearbest for this review. the rated print volume is 30 times 30 times 40cm, by what the cr-10 is made for printing large objects. the printer comes as a kit, but most parts are preassembled so that the build instruction is short and you can put it all together in one hour. tools needed for the assembly and included in the box are a side cutter, two wrenches, five hexagon tools, a screwdriver and a scraper that is useful to remove parts from the print bed.
what are the benefits of 3d printing, the frame is composed of aluminum profiles of the dimensions 40 times 20mm and 20 times 20mm. the orange rubber inserts are nothing but decoration and they have a strong smell after unboxing - at least for the first week. minor scratches on the aluminum frame don't affect the functionality - it's a machine for your workshop. no more than two parts of the frame have to be assembled.
the end switch of the z axis is at one of the bracings. that's all to finish the mechanics of the printer. the wiring is also done quickly. all plugs are labeled and protected against reverse polarity by what a faulty wiring is nearly ruled out. all cables are bound together by weaved tubes, so there is no cable spaghetti. the electronics, including the power supply, is covered by a solid metal case. with that, the insulation of the mains power is perfect. turn the selection switch to the proper mains voltage of your country before powering the device for the first time. all axes are guided by plastic rolls with ball bearings, but i noticed unwanted backlash at the mechanism of the print bed.
when having a look at the bottom side, you can see that 6 rolls are used to guide that axis - a strange number of bearings, because the middle rolls are an unwanted pivot point of the construction. that's why i removed them - additionally the remaining fours bearings were a bit loose, thus i tightened them with a higher torque. with that simple modification the backlash at the y axis is successfully eliminated. the z axis is driven by a single motor through one spindle - the second end of the x axis rides on three rolls without another motor. when lifting... ...and pushing down that end of the x axis by hand, we get an error of approximately 0.4mm, which is anything but negligible. however, while operating the machine, gravity pulls the axis in only one direction. when driving the carriage to the right... ...to the left...
...and back to the right, that end of the x axis moves for no more than approximately 0.02mm. even more important is, that the reading on the dial indicator varies for clearly less than 0.01mm whenever the carriage is at the right end of the x axis, which is nearly perfect. on the aluminum plate of the heated print bed there is a glass plate covered with adhesive tape, kept in place by four metal clamps. the height of the print bed is adjustable through knurled nuts. adjust the bed to a low position before turning the printer on for the first time. besides the two dismounted bearings, a couple of screws are left over, which gives good karma points. furthermore the kit comes with a spare bowden tube including a brass bush and a second nozzle, as well as a roll of adhesive tape. time for turning the machine on: even in idle mode the printer is very loud, because of the two fans in the power supply...
...as well as the third fan at the hotend that emits a high frequency noise. the printer is operated through a rotary knob and an lcd display. with that you can move the motor of the x... ...and y axis. you can also home all axes with one command. before adjusting the printer, you should preheat the print bed to 60 degrees, because the large aluminum plate expands noticeably while heating up. don't move the print head by hand while leveling the bed - use the control panel instead, because as demonstrated before, the right end of the x axis bends under load. besides that the procedure is a usual: the height of the print bed is adjusted in several runs...
...until finally nothing but a sheet of paper fits tightly between nozzle and print bed. the printer comes with a spool of white pla filament. however i am using a 1kg spool of blue pla to get a higher contrast for my first video sequences. to feed the extruder with filament, you must preheat the hotend to 185 degrees celsius... ...and drive the x-axis to the right end. the cr-10 has a bowden extruder. at this type of extruder the heavy stepper motor pulling the filament into the hotend is not part of the carriage moving along the x axis. both parts of the print head are connected through a tube, usually made of polytetrafluoroethylene, a type of plastics with hight temperature resistance and low friction. as soon as the target temperature of the hotend is reached, the filament can be inserted by pressing the lever on the stepper motor.
pull in the filament until a strand of plastics exits the nozzle. now, the printer is prepared for the first job. in order to print one of the files that come with the printer, insert the micro sd-card on the right side of the control box. after that, you must initialize the card. select "change sd-card" whenever you want to remove the card, to avoid data corruption. after initialization you can select one of the files. the job starts as soon as the target temperatures of print bed and hotend are reached. my advice is to use one of the sample files that come with the printer for your first test run, since they work properly on the device. however, the "cat" file confused me - not because the cr-10 could not print cats, but the parameters chosen for that sculpture are not as i expected them to be:
the first filament strand extruded around the print area to pull out the air in the hotend, is very close to the edge of the first layer of the figure - it simply doesn't stand out. furthermore, the first layer looks very rough, because it is printed with low infill and low layer height. i expected to see smooth layers from the beginning of the print, but that did not happen. because of the rough output i canceled the print three times in order to check the bed leveling as well as the extruder. as i could not find anything wrong with the adjustment of the printer i simply let things slide and after a couple of layers it all started looking good. the "cat" file on the sd-card grows layer by layer. the sculpture has a hollow body - the infill it set to zero. with growing height, the print looks better and better and my skepticism disappeared over time. the printer has a 0.4mm nozzle, the layer height is set to 0.12mm and the hotend is adjusted to 200 degrees celcius.
there are no visible strings between the ears of the cat. the print is paused while retracting the filament a couple of millimeters before the print head passes a gap. with that, oozing is eliminated nearly perfectly. after 3 hours and 35 minutes, the cat is printed. there is a reason for the rough structure of the bottom layers i found faulty, first: you can remove the cat from the print bed with ease. all details of the cat, including the chinese characters meaning fortune, are clearly visible. good prints should not be based on good fortune, but on knowledge about filament printing as well as on a solid mechanics so that the results are repeatable. next thing i stored on the sd-card in order to get printed is a link of my cable chain.
the layer height is set to 0.1mm. there is almost no oozing - the few strings being visible are very thin. one of the advantages of a bowden extruder is that the mass of the moving parts is very low. with that, the acceleration of the print head in x direction is set to a high value, which cuts the unwanted strings coming out of the nozzle. the scrapper is very useful to remove the prints, because there is a really good adhesion on the print bed. same as the links printed with the tronxy p802 and the anet a8, the green parts made with the cr-10 fit into the cable chain perfectly. one working print of a link could be nothing but good fortune, however multiple prints with same quality point out that the cr-10 performs not bad. with a print bed of the dimensions 30 times 30cm, the cr-10 is made for large objects. here i am creating a wing part of a model plane.
the base area is 8 times 25cm. the infill is set to 10 percent, the fill pattern is honey combs. the layer height is set 0.3mm in order to reduce the total print time. a "winglet", which is an end plate at the tip of a wing, is printed layer by layer. the cross section of the airfoil is very thin at the rear end - at this point, the print head changes direction rapidly the cr-10 prints that edge of the winglet precisely. the maximum height of the winglet is 20cm. the print was finished after 20 hours - you need lots of patience when creating large objects. especially at the rear end of the horizontal part of the winglet you can clearly see the layered structure of the print.
while designing an object you should always have an eye on the direction of layers and surfaces, however my design failure shows that the printer did a good job - all edges are running in parallel. the surface of the airfoil is smooth at the vertical part - the perimeters are running perpendicular to the layers, by what the low print resolution doesn't matter that much. a step is formed near the rear edge of the airfoil... ...however a look at the design in slic3r shows that this is ruled by software. the rear edge of the airfoil is dead straight which is an indication that the frame of the printer is sturdy and there is no noticeable backlash. the front edge is also dead straight and the shape of the airfoil is replicated with no visible deformations. however the weight is too high for an airplane which is why i must change major parts of the internal structure of the design to get it more lightweight. in comparison to my first 3d printers, the limit of the cr-10 is closer to the sky - literally: next object printed is a rocket mockup.
it's standing on the print bed with four wings and the rocket engine. once more the layer height is set to 0.3mm. the 6mm wings are getting closer to the rocket body layer by layer forming an angle of 69 degrees. the maximum diameter of the rocket body is 75mm. the object is printed as hollow body with a wall thickness of two filament layers. the print speed is 60mm per second for the inner layer... ...and 40mm per second for the outer layer. with a height of 40cm the rocket brings the cr-10 to it's limit. as you can see, the bowden cable touches the upper cross bar of the printer at a height of 35cm.
that doesn't really matter for now as there is not much movement along the x axis while printing the tip of the rocket. however you should consider that limitation when printing objects going above 35cm. after four and a half hours the print it done. the thin wings at the bottom of the rocket are straight. the rocket body is circular and has a smooth surface without gaps in the thin walls of the print. the filament shrinks while cooling down which makes it very tricky to create dimensionally stable, large objects with a filament printer. you can find more information about those issues on my project page.
as a conclusion, the cr-10 is an allrounder for large... ..as well as small workpieces.
i really like the print quality of the cr-10 and you can have a closer look at the full resolution photos of my sample prints on my pages. thanks for watching and: "i'll be back!"
