Hello everyone! By now you probably have seen me post around the form. I've probably posted in a thread that you created! I found the AXIS MakerTech 3D Kickstarter campaign as I find all the campaigns I back, by trawling through the site looking for cool things. I was intrigued not so much for the cost of the printer but for the ability to print two different materials at the same time. I have been wanting to write a review of the Kickstarter campaign and the printer for some time. However, before I did so, I wanted to be able to fully test and try each part of the printer and all the upgrades that I purchased. Now that I've gone through each portion of this kit I feel it's time for me to give my review of the AXIS printer. As an advance notice, I'm trying to be completely fair in this review even if there are points where it may appear that I'm using it as a cathartic device for relieving some of my anger with the printer. The Kickstarter Campaign As mentioned I was really interested in the printer for dual extrusion. Finding the printer on the Kickstarter only meant that I was interested. I next had some research. I quickly found MakerTech's site and that they sold a nice looking Proforge printer. This assured me that they were able to deliver a product. Considering my research on the company done I backed the project. I have backed 40+ campaigns on Kickstarter. I've had less than a handful fail and another similar amount that gave me a bad overall impression. I really have to give MakerTech credit here. From my standpoint they handled the campaign very well. Back at the beginning of the year who thought that we'd have a global standstill on trade because of this pandemic? Even still, at one point it only looked like a bump in the road. MakerTech communicated effectively. They gave monthly updates and were very upfront about where things were in progress. From the time when they indicated that shipping to US backers would occur and when I got the printer was acceptable and anticipated. Overall, I give MakerTech high marks for a well run campaign. I have no complaints and thought that it was definitely one of the better campaigns that I've backed on Kickstarter. Kudos! The AXIS As A Kit Due to the pandemic I had a lot of down time with respect to my normal hobby, RC sailplanes. I had really wanted to get into 3D printing, mainly to supplement the other. The AXIS was to be my first printer. I decided to buy another cheap printer so to acquaint myself with the process of designing and printing parts when the shipping became an issue with regard to the parts for the printer . The printer I chose was pretty much preassembled. I had it together within an hour. This was with me double and triple checking alignment, etc. Realizing that I was completely hooked with this new hobby I immediately began learning as much as I could about everything in the 3D printer world. I upgraded and tweaked the printer heavily. This allowed me to learn a lot about how printers function and how the firmware makes the whole process go. Having the cheap printer working well by the time the AXIS showed up was a huge advantage. Upon opening the box, a very well packaged box I might add, I was not daunted by the parts. Actually, I should say that the boxing of components into labeled boxes and bags with dimensionally accurate pictures of screws and other items was really welcomed! It was also nice to have guides that could be annotated by other users. Overall I thought the kit was really well prepared. I found no major issues in assembling the kit. As with any assembly there were hiccups and backtracks as well as some minor issues. Again, as I had experience with tweaking another printer, it wasn't hard for me to overcome some of the small roadblocks that others encountered. The minor annoyances with some wires being short or having to flip the LCD plug were pretty easy to get past. My largest annoyance with the kit was the lack of an a second static linear guide for the z-axis. This was replaced with a third eccentric. I did have some component failures which were annoying. I was receiving TMC comm. errors with respect to some of the TMC2208's that I had. I didn't really get to troubleshoot them as the MKS Gen L v2.0 board that came with the kit died not long after I had it installed. Having been playing with upgrades on my other printer I was lucky and had a BigTreeTech SKR 1.4 Turbo. I decided to not bother with any of the stock stepper drivers and went with 5 TMC 2209 drivers that I also had available. This allowed me to get the printer up and running after modifying the inductive probe wiring to work off of 5v. As for the quality of the components in the kit I felt there were some pluses and minuses. The MKS Gen L v2.0 board is not a bad board. It sucks that I got a lemon, but that can happen to any piece of electronics that you buy. The same could be said for the TMC drivers. In a perfect scenario MakerTech could have flashed a base firmware on each mainboard and did a quick QC on the drivers. However, that requires time, which had already been expended with the pandemic, and also personnel to do the check, which probably would have added to the cost of the kit. The same could be said about the short cables and incorrectly reversed ends. Not a deal breaker and not really out of the expectations for such a budget printer. The stepper motors themselves appear to be of good quality. It was nice to have an external mosfet for the heated bed in order to take the load off the mainboard. The power supplies were a bit on the paltry side and can be labeled as a disappointment. I don't care for the fans that were used for the case and hotend. The hotend itself, a E3D clone, was said to be all-metal, but clearly had the PTFE tube go to the nozzle, adding another disappointment. 3D Printing With The AXIS After watching the videos on the Kickstarter campaign page I had good hope on the quality of the prints that would be achieved. Looking at the Proforge printer that MakerTech makes I had no reason to believe that the AXIS would be much different. In reality, it couldn't have been more different. Upon assembly it was immediately known that due to the printer's motion design that there would be a lot of work to be done in order to get the printer to produce acceptable results. This forum is full of threads attempting to resolve these issues or provide tips and help on how to make the printer better. This is probably where you've seen me post! I could rehash most of those posts here, but what it boils down to is that the AXIS, while on paper looks acceptable, just doesn't deliver on what it sets out to do. I have no doubt that someone reading this will have their printer setup to the point where they're able to get something acceptable to come off, but I can tell you now, they are one of the lucky ones. I have collaborated with users on the forum on methods to physically make the printer better. I started off trying to keep the modifications to a degree that almost anyone could do them. I started with attempts at changing the friction mounts to using various sized bearings with cups to guide the build plate. I changed the lead screw coupler to an aluminum one. The lead screw and the nut that guides it were swapped out. I removed the lead screw bracket and made several attempts at designing brackets to keep it aligned with bearings. These only provided small changes that really had no marked improvement. The modifications then turned to trying almost anything. I've switched from the friction guides on the x and z axis to linear rails. The redesigning of motion systems and belt routing had to be done. I purchased a borosilica glass build surface in order to insure a flat, level plane for the printer to print upon. I replaced the thermistor in an attempt to mitigate constant thermal runaway errors. The power supplies were replaced with a Mean Well LRS-350. In the end I sunk over an additional $170 into the kit. This amount doesn't include the items that I already had on hand that I had to use to supplement the kit, a mainboard, stepper drivers, various connectors, etc. I have also scoured through documentation for the Marlin firmware and slicer settings trying to come up with the best results. To say that I've flashed a few versions of the firmware would be an understatement. I mounted a micro SD extension so that I could flash to the mainboard without having to turn the printer over every time and insert the card! In the end the level of effort I put in didn't generate a large difference in the output of the printer. Outside of the printer itself are the issues I have with the slicer recommended by MakerTech, their version of Cura. This had been a concern of mine during the campaign. I had even requested that they submit a PR to Ultimaker's repo so that it would be easier to keep up to date and work better in an ecosystem were users had other printers. In the end I found their version a little more frustrating. The coexistence with it and my "mainline" version of Cura caused issues. This was problem when Ultimaker release 4.7. I then had issues with the loss of my entire profile for my other printer and the material properties that I had been cataloging. I tried using Prusa Slicer as well in order to tame some of the settings that I wanted to try in order to get better prints. This was more of a headache. This isn't a direct reflection on MakerTech since they don't recommend that slicer. Upgrades I opted to get all the upgrades. At the time I thought this would be the first printer I would own. I researched each of the options and knew that I would want or "need" them. Out of all of them I think the heated bed and LCD are the best and should be included as a standard. The LCD is literally a must have. Using pronterface to control the printer outside of setup is just painful. You could argue that you could do everything via Octoprint but still, there are a lot of setup items that can be much better handled via the LCD. With respect to the heated bed, there are plenty that say that a build-tack type surface with no heat will still have PLA adhere. However, I would caution to say that it's much more likely first timer and cost conscious users would want to be able to not have to worry with constant first layer adhesion issues. I was most disappointed in the dual extrusion upgrade. This is what really attracted me to this printer. Granted I haven't tried more than a few times to get it to work. I gave up after having what seemed like really basic issues. Jams and settings just didn't seem to be something I could overcome. The piece from TriangleLabs just doesn't seem to get good reviews or marks. When I saw TeachingTech on YouTube do the video of MakerTech's dual extruder system I had hoped it would function something like that. Being not too familiar with 3D printers at the time I didn't realize that it was a completely different tooling setup than what was on the AXIS. The must frustrating upgrade was Octoprint. This should have been easy. I have about a dozen Raspberry Pi's operating in my house at the moment. I have 2's, 3's, and a 4. They all work flawlessly on Raspbian with wired connections, wireless adapters, or built-in WiFi. I didn't realize it when pulling the board out that this was a Pi B+. After getting through the customary setup with Raspbian and additional quirks of Octopi I was greeted by the utter slowness of this board. I hadn't even launched into the Octoprint web interface when I was concerned with how slow commands over SSH were executing. The power supply that was provided couldn't even run the Pi! I can't say I was disappointed here because this should have been the easiest thing to test and fix before anything was even sourced by MakerTech. Could it be that the card that was included was too slow? Maybe I could swap in a different Pi that I have. These could be addressed but as mentioned with the upgrades that I've done above it's just beyond disappointment and into frustration and disgust. Conclusion and Verdict There were some very positive things about the Kickstarter and the printer. I was very pleased with the overall Kickstarter campaign. I think we really need to give MakerTech a big thumbs up here. Obviously we all would have liked to get our printers way back in April, but with the situation the world was in at that time one could hardly blame them for the delay. The AXIS kit was also well packaged and had good documentation that allowed for community updates. Along with the positives were a good amount of negatives and drawbacks. The friction design for the motion system just do not provide the best solution for achieving an acceptable print. The cost nature of the project really hurt when sourcing cheaper components such as the power supplies, lead screw coupler and nut, and lack of pulleys replaced with bearings and washers. The upgrades that shouldn't have been as affected by cost concerns were also negatives. The bed and LCD I saw were necessary for any 3D printer. The dual extrusion upgrade was pretty much dead out of the gate for me. I'm really looking to see if anyone can conquer it as it is an area that I have the least knowledge in 3D printing compounded by my overall limited knowledge. The Pi included in the OctoPrint upgrade is almost worthless. You can pick up a 3 A+ for less than $30. It has WiFi built in. So what do I think of this printer? I just can't see it getting a recommendation from me. I am still struggling to get good prints after having the AXIS together for a while and attempting to upgrade it. With the amount of money that I've sunk into it for upgrades or fixes it does not win a value proposition. There are extremely well put together printers that cost under $300 that provide the ability to produce good quality prints. They can be built much quicker than this kit. They have almost all the upgrades already included and installed. I must say that I'm really glad that this was not my first 3D printer. I don't know if I'd be as interested in the hobby if I was faced with the hurdles that I've had to attempt to get past with this printer.

I write "advanced" but that doesn't mean this discussion isn't for the uninitiated or those who have just started printing. I hope this discussion can be an in-depth and beyond-the-basics type of topic. Let me state up front that I own another printer with a BLTouch probe and have been working with it and a BTT SKR 1.4 Turbo board to level an Ender 3 Pro. Most of my knowledge of the Marline firmware and bed leveling come from the setup of this probe and printer. Current State of Bed Leveling with the Makertech Axis The AXIS printer uses an inductive probe with a PNP normally open transistor switch. This probe will close or trigger when it is at 4mm from a metal object. Materials commonly used as build surfaces which are not conductive, such as masking tape, glass, etc, will not be detected by this probe. In the Configuration.h file that is provided by Makertech ABL (automated bed leveling) is configured to use a 20 point grid in a 5, 4 configuration to achieve leveling using the bilinear method. The bilinear ABL configuration is, from my observation, the most common way to use a probe to create a mesh of the height of the bed at various points. The printer's firmware alters the z-axis height according to this mesh to produce a uniform set of initial layers onto the build surface. The limit in height that this mesh affects the z-axis is known as the Fade Height. The amount of leveling applied is gradually faded until this height is reached where the printer will then assume the print is level. The Fade Height is adjusted with g-code and is set to 4mm. The g-code for the AXIS printer is found in the printer profile in Cura, M420 S1 Z4. In the Marlin firmware there is a setting RESTORE_LEVELING_AFTER_G28. Having this defined (uncommented) means that after the printer does its homing routine that the bed leveling mesh is restored. With the current configuration of the Marlin printer a leveling g-code command is sent to the printer, G29, which causes the bed to be probed and the mesh stored. During printing a G28, home the printer, command is sent followed by the Fade Height command. Questions and Thoughts on Current Configuration I have had a few thoughts on the current configuration of bed leveling that I was hoping we could discuss. These are mainly around the g-code sent to the printer. First, I'm new to an inductive probe. With a probe such as a BLTouch physically touching the surface I find it important that the bed leveling mesh be generated often as a bed's surface can change over time. Since our beds are most likely covered by some non-conducting material it doesn't make as much since to do this. However, I still feel that a pre-print leveling routine should be performed. I have added a G29 command after the G28 command in the printer's configuration within Cura. Next, since the mesh is reloaded after G28 and the G29 command comes after it I do not believe the Fade Height command is necessary. This value can be set once via Pronterface or the LCD and then stored in the EEPROM. It should be recalled automatically. Therefore, I've removed it from the slicer's configuration of the printer. Something I've noticed, and this is where I'd like feedback, is that I can see a valid mesh get returned by the G29 command, I can see the z-axis move during the print, but I don't feel like it is correct or by enough. There are spots on my bed that I know the nozzle will hit even after a level command. What I don't know is if this is caused by the non-conductive surface on the flex plate. I can also see this on my other printer which has a similar configuration. Is there some incorrect assumption that I made above? Please share your thoughts. Improvements to the Current Firmware Configuration There are several configuration changes that I feel make the process of getting a level bed much easier. Some of these will require the LCD to be installed on your printer in order to take full advantage of the feature. Babystepping is a change that is probably helpful to all users which have the LCD installed. Here's the comments from the Marlin firmware: "Babystepping enables movement of the axes by tiny increments without changing the current position values. This feature is used primarily to adjust the Z axis in the first layer of a print in real-time." What this is essentially saying is that you are able to adjust the nozzle height from the bed using your LCD and knob. The configuration of babystepping is done in the Configuration_adv.h file. The following is an example of a configured babystepping section: #define BABYSTEPPING #if ENABLED(BABYSTEPPING)   //#define INTEGRATED_BABYSTEPPING         // EXPERIMENTAL integration of babystepping into the Stepper ISR   //#define BABYSTEP_WITHOUT_HOMING   //#define BABYSTEP_XY                     // Also enable X/Y Babystepping. Not supported on DELTA! #define BABYSTEP_INVERT_Z false           // Change if Z babysteps should go the other way   //#define BABYSTEP_MILLIMETER_UNITS       // Specify BABYSTEP_MULTIPLICATOR_(XY|Z) in mm instead of micro-steps #define BABYSTEP_MULTIPLICATOR_Z 10       // (steps or mm) Steps or millimeter distance for each Z babystep #define BABYSTEP_MULTIPLICATOR_XY 1       // (steps or mm) Steps or millimeter distance for each XY babystep #define DOUBLECLICK_FOR_Z_BABYSTEPPING  // Double-click on the Status Screen for Z Babystepping. #if ENABLED(DOUBLECLICK_FOR_Z_BABYSTEPPING) #define DOUBLECLICK_MAX_INTERVAL 1250   // Maximum interval between clicks, in milliseconds.                                             // Note: Extra time may be added to mitigate controller latency.     //#define BABYSTEP_ALWAYS_AVAILABLE     // Allow babystepping at all times (not just during movement).     //#define MOVE_Z_WHEN_IDLE              // Jump to the move Z menu on doubleclick when printer is idle. #if ENABLED(MOVE_Z_WHEN_IDLE) #define MOVE_Z_IDLE_MULTIPLICATOR 1   // Multiply 1mm by this factor for the move step size.     #endif   #endif   //#define BABYSTEP_DISPLAY_TOTAL          // Display total babysteps since last G28 #define BABYSTEP_ZPROBE_OFFSET          // Combine M851 Z and Babystepping #if ENABLED(BABYSTEP_ZPROBE_OFFSET)     //#define BABYSTEP_HOTEND_Z_OFFSET      // For multiple hotends, babystep relative Z offsets     //#define BABYSTEP_ZPROBE_GFX_OVERLAY   // Enable graphical overlay on Z-offset editor   #endif #endif Unified Bed Leveling - UBL There exists a more advanced method of leveling the bed, UBL. Unified bed leveling is described in the Marlin firmware as "a comprehensive bed leveling system combining the features and benefits of other systems. UBL also includes integrated Mesh Generation, Mesh Validation and Mesh Editing systems." The idea of UBL is to make a very, very detailed mesh of the bed. By default it will try to probe 100 points. This mesh can then be manually adjusted and then validated through a special print. When the printer goes to print it loads this mesh and can then be instructed to pick a few points for validation. This will help adjust the mesh if minor changes have occurred. The steps for enabling and configuring UBL are much more involved than the bilinear setup that is enabled by default. The first thing which is necessary is to change the configuration in the firmware. Enable UBL in Configuration.h by uncommenting "#define AUTO_BED_LEVELING_UBL" and commenting out bilinear "//#define AUTO_BED_LEVELING_LINEAR". Next you'll want to enable the G26 mesh validation tool by uncommenting "#define G26_MESH_VALIDATION". For an ultra-precise grid you should validate that "#define GRID_MAX_POINTS_X 10" is set. If you have an LCD I highly suggest configuring babystepping as shown above. Once you've compiled the firmware and sent it to the printer it is then necessary to perform the initial calibration of the mesh. If you have an LCD you will likely stumble onto the UBL configuration section. There is a procedure in the menu for creating, editing, and validating the mesh. I have never been able to successfully use these menu options. Beware that it is possible for the printer to be put into a state where the nozzle will collide with the bed. I do not know if this is because of how z-offset is factored in or if there are bugs. I suggest using Pronterface to send the commands manually. I have found the best way to configure UBL is through following the commands in Marlin's topic on UBL: https://marlinfw.org/docs/features/unified_bed_leveling.html#synopsis In the Synopsis section is a box showing the g-code for initializing, fine tuning, and validating the mesh. It basically works like this: After sending the firmware to your printer you should reset things. M502 ; reset EEPROM M500 ; save EEPROM M501 ; load EEPROM Prepare the printer for initial mesh creation. The printer should be brought to typical printing temperature. M190 S65 ; heat the bed Home the printer G28 ; must be performed before other commands are issued Perform the initial mesh creation G29 P1 ; Probe the bed After the automated probing is completed it will likely be the case that there are points in the mesh that couldn't be probed. G29 T ; print the current map, there will likely be a bunch of 0,0 entries G29 P3 T ; smart fill empty points and print the map. if there a bunch of 0,0 entries this should be run until all are filled in Save the mesh to slot one. There are multiple slots if you have multiple build surfaces you wanted to level. G29 S1 Set the fade height to 4mm (4mm is from the Makertech configuration) G29 F 4.0 Activate the UBL system G29 A Save the configuration to EEPROM M500 ; Save current setup. UBL will be active at power up, before any G28. (Which is what we want) Before you print the validation mesh get your z-offset set (remember you cleared your settings after the firmware was flashed) Print the validation mesh G26 L0.2 B60 H200 S0.3 ; Prints mesh with a layer height of 0.2mm, bed heated to 60, hotend to 200, and is using a 0.3mm nozzle. If after you print the validation mesh you have bad areas you'll need to adjust them. I haven't had to do this, but following the guide in the link should get you there. In your slicer you'll want to edit the printer so that you can change the g-code that's executed before the printer starts printing the model. This is referred to as the "start g-code" in many slicers. After the G28 command we no longer need the M420 command, it can be removed. During our setup we activated the mesh and in the firmware the mesh should be loaded after G28, but it is good to add in "G29 L1 ; Load the mesh stored in slot 1 (from G29 S1)" after the G28. Next, it is a good practice to quickly check 3 points on the build surface to help modify the mesh for small deviations. Follow the previous command with "G29 J ; Probe 3 points for mesh alteration". You can modify the command by appending an integer after "J" to define more than 3 points if you wish. Generally, 3 should be enough but if you wanted to do more you're only adding to the time before the actual printing starts. This should get your UBL system up and running. Hopefully the overall mesh, having more points, is more accurate. Adding the 3-point check before printing should help with any small deviations. Revisiting creating the mesh is necessary if any large changes to your build surface occur. Where Do We Go From Here? Hopefully this gives us a great starting point for discussing bed leveling for not only the AXIS printer but all 3D printers running on the Marlin (or derivative) firmware. I tried to best describe what the AXIS printer is actually configured to do and what it does. I've also tried to provide an example of a configuration change to help people with leveling their beds and talked about a different method of leveling the bed. The section on UBL should provide you with enough to get up and running with a much more comprehensive bed leveling technology. I want to open up a discussion regarding the aspects of leveling. Please feel free to correct my explanations if they're incorrect. Also help clarify if things stills seem murky. I don't know everything but I am hoping that by talking among ourselves that we can all gain a better understanding of how this process works.

I hooked the board up to flash the firmware and flashed it. I must have done something wrong in the config because I get nothing when I turn on the printer and connecting it back to the USB shows no port in Windows. I've pulled out my Arduino UNO and attempted to flash a bootloader via the Arduino as ISP method but I'm getting nothing. Right now when I plug in the board I'll get a red and blue light with the blue light turning out after a moment. There are two green lights next to the USB that I don't recall seeing illuminated when I first connected it. Is it a lost cause?

I've been building the latest version of the Marlin firmware for my Ender 3 and SKR 1.4 Turbo for a while. I've created two branches in my fork on GitHub, one for a base configuration as close to what Makertech has provided in their AXIS repo, and the other for "enhancements" that I think will make things easier for users. At the moment I don't have a MKS GEN L V2 board handy so I cannot test these settings until I get my printer. PLEASE BEWARE I TAKE NO RESPONSIBILITY FOR DAMAGED PRINTERS OR BROKEN BOARDS. EVERY ATTEMPT HAS BEEN MADE TO VERIFY THAT EVERYTHING WORKS BUT IT IS UP TO YOU TO INSURE THAT ENDSTOPS AND LIMITS WORK CORRECTLY. Here's the link to the "base" branch: https://github.com/mikejr83/Marlin/tree/feature/AXIS/base The enhanced branch: https://github.com/mikejr83/Marlin/tree/feature/AXIS/enhanced Features Enabled in "Enhanced" Branch: Babystepping Advanced Pause - M600 filament change support In the future I may turn on volumetric features and linear advance. Please let me know if there is anything you may be interested in enabling or testing. Compiling and Flashing For either of these branches you can download the source as a zip file or you can clone the source and switch between any of my branches (including the work I've done on the Ender 3). I've set them up to build using PlatformIO and VSCode. There are tutorials you can find on YouTube. The nice thing is that VSCode is a much, much better IDE compared to the Arduino IDE. Notes on configuring before building: Stage 09 - Steps 3 - 6 show how to configure the firmware for upgrades you got with your printer. You will still need to make these changes on the base or enhanced branches. The Makertech configuration had the heated bed enabled by default. Both the base and the enhanced branch match this but made no changes with respect to enabling the TMC drivers. If you purchased that upgrade you will need to make those changes. The changes should be the same as what is outlined in the instructions. If there is interest I would be willing to putting together a version of the Marlin 2.x firmware compiled for the SKR 1.4 Turbo as well. I have a spare board and set of TMC2209 drivers. This will not be a priority until I get my printer, set it up, and have it printing satisfactory. Once I get things working like they should I break them and make them not work so that I can experience the joy of making them work again!

Let everyone know if you have received a shipping notice or have received your printer. It might be helpful to let us know where you're located as well. I'm still waiting for mine in FL, USA. I haven't got a shipping notice yet but heard that other international backers have received their printer. Hopefully I'll hear something soon!