My current BuildTak surface has lived for long enough. By now, I have multiple tears and holes in the middle of the bed, along with an air bubble in one corner. It is time to replace it.
It is awesome how cheap BuildTak surfaces are. Here is a new one I had in reserve for when the current one dies. Now I can finally use it.
The total width of it is 240 mm, which is too much for my bed. If I want to be able to fit it, I will need to cut it down a bit. To keep it centered, I decided to trim 10 mm from both left and right sides, bringing it down to 220 mm.
I marked multiple spots and then drew a line. This is the line which needs to be cut.
Donât be stupid like me. Cut it while looking at the line instead.
Alright, the surface is trimmed, but now we need to make clearance for the hotbed screws.
I found that marking 15 mm on both sides of the corner and then cutting the triangle out worked quite well. It provided a nice clearance for the screws, but didnât lose too much surface.
Corners trimmed! Now comes the difficult part.
Join me, as we take the old BuildTak off. There will be much suffering.
Got a bit more...
And even more! Okay, this really is hell. And okay, it is no longer going, letâs try the other side.
Using a knife to separate it here...
AAAAAAAAAA!!!!!
After a lot of sweat, smell of the glue they used, sticky fingers and pain, I finally got it off. This one is going to the trash!
Before proceeding, I cleaned up the aluminium bed with some alcohol. Just to be sure there is no contamination which could ruin adhesion of the new BuildTak.
The essential tool for installing. Seriously, you need this. A simple card works. The thing is, while installing it, you really want to squeeze all of the air out BEFORE you glue it down. This means you need to go slowly, bit by bit, using a card to push the air out as you advance.
Behold, a fresh new BuildTak, ready for sticking!
And installed! This time, i managed to perfectly position it to cover the bed as optimal as possible.
A bit of re-leveling, just to make sure that bad stuff doesnât happen straight off the bat.
Then, I cleaned up the surface with alcohol. I found that I can use even 96% alcohol without damaging BuildTak. Does an awesome job at cleaning it.
Aaaaaaand, insane success! Not only am I able to use the WHOLE surface now, but I was also able to double the first layer printing speed. In fact, I found that the faster first layer speed makes up for an even better bottom layer!
Nightmare Rarity project is well under way already. The frame itself is complete, and I am currently in the process of printing the smaller components. But, lets take a look at the first week or so of building her...
I have to admit that the build was not too easy, at least not when you go full OCD on it and want to assemble it absolutely perfectly. The first thing I did for every section was preparing the screws and nuts in all the brackets.
I would make sure that when installed, the brackets are completely flush with the endings of the aluminium profiles.
Here is the left side assembled.
I then connected the front and back profiles. This part was hell. It is incredibly annoying to slot the nuts in. You adjust one, then the other one unadjusts, repeat.
Eventually, I succeeded, making sure that everything is completely flush.
The corner pieces accept these little rubber feet. You need to use quite a bit of force to push in the screw all the way in.
Here is the corner piece with a rubber foot added.
And here is the frame so far! While it is a bit annoying to build, it is also incredibly satisfying to see her grow.
Next up, right side profile. This one features a mounting point for the bed cable chain. I remixed another one into my own design and I will release it eventually.
Here is the right side fully assembled.
Aaaaand....
Boom, the bottom is assembled... ... am I missing something? ...I am missing the motor mount on the back, arenât I... Well then, time to disassemble...
The motor mount I will be using is from the AM8 Reinforced set. Awesome little thingy, much stronger than the original mount.
NOW it is assembled.
Time to move on to the vertical profiles. I used one of the printed corner pieces to make sure that it is under a right angle.
And I also used my trusty spacers I talked about earlier to correctly align it.
These also accept rubber feet, but along with another screw, they screw into the profileâs side.
I repeated the same with the other side.
Next up: top profile.
Here is the whole part assembled and ready for insertion.
...aaaand done! The whole frame is done for now.
In the meantime, I replaced the blower duct with a different color, as per Nightmare Rarityâs design. Goes really well with the E3D v6 silicone sock, I have to say!
Fast forward a day or so, I printed out the Y guide rod holders and the belt tensioner block, also from the AM8 Reinforced set. Here are all the parts with all the screws already installed, ready for mounting.
The front side, using my trusty spacers.
And here is the back side as well. Everything aligned perfectly. I hope.
Here is how she looks like after this step.
Fast forward even more time... The Z motor holders and rod holders printed, all from the reinforced set. I also printed the Z endstop.
Right side parts with installed screws.
Here I used the spacer to make sure the height is correct.
And here is how she looks like after these parts. Better and better, right?
The biggest print I ever did. 13 hours, 200 grams of plastic. It ate through 1/5 of my filament roll. My custom redesigned case for the RAMPS 1.4 board and MOSFETs, all slightly reduced and compacted in size. I printed it at 0.32 mm layer height.
Boom! Here is the whole case assembled. Note that I only assembled it to make sure that the parts fit. The electronics are just thrown in there for now.
And here come the issues! First issue, it looks like my printer is not completely square, and that it prints in a slight parallelogram. I will fix this by sanding for now, but it is rather annoying.
Second issue, by design, the fan was expected to be a bit curved for the lid screw to fit...
Real life shows a different story however... Looks like I will need to bring out my sander and grind a bit of plastic off the fan.
I am quite proud of this. The custom front face of my LCD case, complete with her name. I had to re-print the button because the tolerances were just too tight, making it not fit well.
And finally, all parts I have printed and prepared for now. There is still more to go! I ordered a whole new roll of violet PLA because I ran out. I ran out by literally 30 grams.
While designing Nightmare Rarity, I came to an interesting conclusion: How will I ensure that all of the parts on my build are properly spaced from each other and that all distances are correct? To combat this, I designed spacers.
Here is a nice shot from above which shows the alignment of all spacers. There are also more spacers located below the Z motors.
Here are the spacers already printed. Nothing special really, just a set of pieces which can be used to correctly align all parts of the printer. They can be inserted into the aluminium extrusion channels and pushed firmly against the edge. A part that should be aligned can then be further pushed against these spacers, putting it in the exact location it has to be on. I believe I covered all of the most critical parts with this, but that is yet to be seen when I start assembling her. I will release it on my Thingiverse at a later date.
Oh and here is her very first part I ever printed! The print stop button. Gotta use that red filament while it is still loaded!
The Logitech C270 webcam. Quite a nice piece of hardware. I use it with my Octoprint so I can get a nice live video whenever I am printing. It has one major flaw however: It is impossible to focus with it. Time to fix that.
Believe it or not, but this camera CAN actually focus. Itâs just impossible to focus from the outside of the case, so we have to disassemble it. First, I had to take off the front panel. I used a small flat screwdriver for this.
The next part involved removing the screwed in cover that was below the previous one. For this, I had to find the smallest cross screwdriver I had. Once removed, the electronics were exposed. Nothing special. But, see that ribbed ring around the lens?
Sadly, one canât just turn it to focus... at least not before the glue is removed. So, I removed the small piece of hot glue and now the lens focus was adjustable!
Finally, I mounted it back on my printer and adjusted the focus. Once I was happy with it, I reassembled the camera. My theory why they did this is that they didnât want it to unfocus by itself as people are using it, but they were also too lazy to add some sort of a focusing knob on the outside. Oh well.
In the last post, I was assembling my new direct extruder carriage, but ran into some minor issues. Time to fix them all up.
First up, the filament guide. As promised, I have uploaded it to Thingiverse, so you can find it there. The guide accepts a 1/8'' BSPT thread for E3Dâs Threaded Bowden Coupling and provides a nice mount for a bowden tube which guides the filament to a direct extruder.
It mounts neatly behind the LCD display on already existing screws. You can choose whether to mount it on the left or right side as well.
The second problem I encountered was that the Z endstop had to be positioned on a lower height than possible with stock. To solve this, I designed this printable replacement endstop which is incredibly simple to install and directly replaces the stock one.
Removing the stock.
The switch installed on the new endstop.
And finally, the new endstop installed. I did have to do some slight adjustments just so it triggers exactly on the height it should, without having to re-level too much. Iâm lazy.
And the final problem I had was the inadequate height of the chain mount, so time to get rid of the old one.
The new mount being installed! Now that everything is in place, here are some glamour shots of this bad girl...
And that pretty much concludes this whole post! I have also finally made my carriage publicly available few days ago so you can go download it and print it from my Thingiverse!Â
I have to say, Iâve never had prints as beautiful as this before. Going back to direct extrusion was one of the best choices I ever did.
I would like to introduce a whole new project I am working on: Nightmare Rarity. A heavily customized AM8 build with tons of parts I have designed myself.
She is a true beauty, isnât she? I have already designed the whole printer virtually in Blender, using all of the components that I intend printing, along with the color scheme. The beautiful combination of black and purple should be stunning once complete. The printerâs design is heavily inspired by the character Nightmare Rarity from My Little Pony: Friendship Is Magic comic book series.
Considering how custom this whole build is, I engraved the name of the printer onto the LCD case as well.
While a large number of parts will simply be moved over to the new printer, I decided to replace the stock mainboard with the RAMPS 1.4 board. I will use two TMC2100 drivers to drive the X and Y axes, while the extruder and Z axis will be driven by two A4988 drivers. I will keep using the MOSFETs however. I donât trust the connectors on the RAMPS board with this job.
The printer is not complete without a custom bootscreen!
Overall, this is still a heavy work in progress. I ordered the black AM8 upgrade kit from RatRig and I am currently waiting for it to arrive. I will probably be posting more updates about building this printer in the near future instead of my main printer. The blog might even receive a color conversion as well once she is complete. And yes, I WILL release all of my custom parts once the build is complete and running.
I am a member of a rather large Facebook group for the Anet printers, and people often rave about having to solder the hotbed connectors and such and blabla, and it is all absolute bull****. It is time to explain and debunk WHY one should NOT be soldering the cables to the hotbed, and what is the ACTUAL proper solution for the connectors burning out.
All of these posts and comments stem from the fact that peopleâs connectors burn out. They burn out for two reasons actually, related to the way how Anet cheaped out on them and the fact that they didnât include a way to secure the cables.
The first problem is with the current rating of the connector. The connectors in question are JST VHR-6N. Notice that the rating in the official spec sheet is 10A, which is around equal to the current that the bed eats when preheating. It is important to note that this rating is PER PIN, not the whole connector.
Now, if you look at the hotbedâs pins, you will see that it has TWO plus and TWO minus connectors. Anet decided to use only one of both. If two were used, the current would be split up between the two pins and each would happily run with 5A of current.  Â
EDIT: The official specs state it is not a good idea to spread the load on two cables if the current exceeds the maximum rating. However, if you are doing things correctly and safely, it should not matter. The current limit is 10A, and that is generally the maxmimum amount that the bed will pull during preheating. Two cables might deliver current asymmetrically, but it should not matter as each will still be well below the current rating.
So, the first possible solution is to buy a pre-crimped connector like this one. Note that you should probably find a longer one somewhere on the internet. this one is too short.
The other solution is that you could crimp the connectors yourself. In this case, you will need the connector housing and the individual pins. Please check the links for relevant parts. You also donât need cables as thick as everyone says you do. These pins accept a cable of maximum 0.83 mm2, and I used 0.75 mm2. For those who donât understand the metric system, that is AWG 18. Anything bigger than that wonât be securely crimpable.
EDIT: On the image above, I crimped the connectors manually myself without using an adequate tool. I would highly suggest against this. The cables will fray no matter how much you try over time. Get a proper tool. I suggest getting the cheap SN-28B off eBay.
The other issue stems from the fact that the wires are not secured and strain relieved. This causes arcing on the contact surface between the hotbed pins and connectors in the plug each time the bed moves. We are dealing with very high currents here, so this is a real problem. The arcing slightly oxidizes the connector each time it happens, increasing the contact resistance. A higher contact resistance results in heating up, and eventually, the connector burning out. The best solution is to use a printable strain relief. You could also use a cable chain, and they are completely fine if assembled correctly. You can read more about my failure to assemble it correctly and how I fixed it in one of my previous blog posts.
So, why should you NOT solder?Â
Because chances are that your soldering iron is too underpowered for the task of heating up this massive surface. Most soldering irons are too weak, and will not be able to heat up the connectors without losing all the heat to the large aluminium plate. Some people used a hair dryer to heat up the whole surface.
Because you probably donât have enough experience with soldering to do a proper job. Just look at the image above. I am not sure if it was done satirically or not, but Iâve seen some solder jobs so bad that my head hurt.
Do you understand what will happen when these connectors snap off? You have a set of wires carrying 10A of current soldered to the bottom of your bed and all of the strain is now re-routed to both the solder joints and the copper traces. When one of them breaks off, they WILL cause a short circuit. In a best case scenario, this will result in a fried MOSFET. In the worst case scenario, with your house on fire. People who solder often believe that they suddenly donât have to add strain relief, which is absolutely incorrect. Unlike soldered connections, when using the connector the cables are securely insulated in the connectorâs housing and there is no risk of a short circuit.
What happens when you need to disassemble your hotbed? Will you unsolder again? Will you unscrew the terminals at the MOSFET and then pull the cables through the whole length of your printer just to remove the bed? This causes so many issues it is hard to even count.
Iâve also seen people wire their hotbeds like this. It does âseemâ fine and it will work, but it poses a different set of risks as well. First of all, if one does it like this, using a heat shrink is absolutely necessary to insulate the crimps. Notice how exposed those crimps are. If one gets pulled off, the same happens as with the soldered hotbed. This also means that a strain relief is still a must.
Secondly, Iâve drawn a simple diagram above. This is how the connection between these crimps and the pins on the hotbed look like. The contact areas are circled with red. Does this look like enough surface contact for 5 amps to you?
I hope this post was informative enough to deter you away from soldering to your hotbed or using some weird DIY methods which are sub-optimal. Please, donât listen to those who tell you to solder to the hotbed.
What a beast this direct extruder is. I finally came around to assembling the direct extruder design that I came up with in Blender. Take a journey with me as I assemble it!
A sad sight, but a needed one. The first thing I did was that I removed the bowden tube. I wonât be needing it for some time now.
Unwrapping the wires...
And pulling the wires out of the cable chain... Messy...
The old hotend in pieces. It will soon have a better life.
Iâve completely pulled the E3D v6 out of the assembly so I can use it easier later.
Considering I will be changing the cable chain mounts, it temporarily had to go as well.
And that is the whole original carriage gone, with only the Igus Drylin bearings and the belt holder left.
Considering I am changing the cable chain mount and the carriage itself, the motor temporarily had to go as well. This will allow me to test whether X moves smoothly easier as well.
The new carriage mounted! Along with me testing the smoothness of movement. I had to do a few minor adjustments to the screws to get it working.
Motor and the new cable chain mounted as well.
Aaaand we have a problem. The cable chain is too low and will hit the top of the extruder.
Quite a drity temporary fix, but it worked. I guess I will redesign the motor cable chain mount and make it higher.
Time to get rid of the old extruder!
The motor as it fits into my new bracket and the Titan Extruderâs casing.
Just look at that absolute beast. Titan Extruder and E3D v6 hotend in one extremely powerful package. I have to say that it is pretty heavy too. I didnât post pics of assembly because E3D already has that covered in their official guide.
The extruder assembly mounted on the carriage.
Heatsink cooling fan attached, and the blower fan mount screwed into the assembly.
Everything completely assembled.
Now comes the part where I fix up the wiring a bit. First step: Exposing the mainboard.
Kinda ugly, but that will soon change.
First step is to secure all of the cables at the head to prevent any unwanted movement. I used a ziptie.
Then, all the wires inserted neatly into the chain. This is the important part. You donât want any cables too tight or stretching in there. This is how the chains mutilate your cables.
And for the final touch, tying all the cables in there nicely! The âlooseâ hotend cable up there shouldnât matter too much as the printer doesnât make any major movements on the Z axis.
Picture of the finished assembly!
I admit, I forgot the print the bowden tube holder. Again, I am using it here only as a guide for the filament. It should help prevent filament hitting the chain and/or other parts of the printer. That said, I used a zip tie to hold the tube in place for now.
Itâs alive! And honestly, I absolutely adore the quick extrusion response. With bowden, I used to have to wait a few seconds for the filament to start coming out. Here, it is instant.
One problem I had was that the new head puts the nozzle tip so high up that the endstop couldnât go that low. To compensate, I had to use only one screw and hope for the best. I will design a customized part for this and add it to my printing head design.
The first thing I did was calibrating the Linear Advance feature in Marlin. This basically lets the firmware handle all pressure changes in the nozzle and thus gets rid of bulging corners, stringing and some other artifacts. You can read more and perform the test for it in the link provided.
Was the decision to move back to direct extrusion a good decision? Absolutely in every way and form. The cube above is the most beautiful one I ever printed. It is simply pristine.
This is only the Part 1 of this little âstoryâ. I will be posting a Part 2 soon with all of the issues I encountered here addressed, like printing the new parts. I will also upload my head design at that time, once I am sure that everything works well.
Ever since I upgraded to a bowden extruder, Iâve always had constant issues of various sorts. Lets look at an overview of all the issues Iâve had so far and whether a bowden configuration is really worth it. (Image taken from http://www.fabbaloo.com/)
Iâve indeed had various issues with my bowden extruder. Here is a complete list:
Clogs, clogs and more clogs. I admit, Iâve never had a complete clog where the extruder would just completely stop extruding, but Iâve had semi-random skips on the motor as it doesnât have enough torque to push the filament through. This is exactly why I upgraded to the E3D Titan Extruder, but that felt like it didnât help all that much at all, as now I have the hob drive gear grinding into the filament instead. Iâve always had to oil my filament to combat this issue, and while it does work, I feel like it is just avoiding the main problem.
Very lengthy retractions. The E3D v6 hotend is designed to have a very short retraction distance, usually less than 1 mm. However, I need to run retractions at 2 mm or more to combat the backlash caused by the empty space inside of the bowden tube. This is why I upgraded to the E3D Capricorn tube which has a smaller inner diameter, but this only helped by 0.5 mm. I used to use a 2.5 mm retraction, I am now using 2.0 mm instead.
Filament wasting. The bowden tube is relatively long, and there is no efficient way to swap the filament without having it string and stretch inside of the tube. Because of this, my filament swapping script simply pushes out the whole filament out of the hotend until the new one arrives. This is extremely inefficient as it wastes filament length of the whole bowden tube.
Lighter head provides no true benefit on a Prusa-style printer. The main thing to consider here is the weight of the hotbed. It is heavy. Much heavier than the head. Reducing the weight of the head will not help with increasing the acceleration and jerk as the bed is still a heavy limiting factor here.
So, all in all, what can we do here?
Move back to a direct extruder, obviously. The bowden setup simply has no benefits for my printer, and I can still easily use both the E3D v6 hotend and the Titan Extruder together to design one awesome direct extruder. (Image taken from http://www.fabbaloo.com/)
So, I came up with my own direct extruder carriage. It was designed to be as rigid as possible while being modular AND reusing all of the parts that I had. The Titan bracket is mounted using four screws screwed directly into the carriage, and the blower fan is mounted onto this bracket using two screws. Like this, the blower fan should be easily swappable in case you want to use something like a pen instead.
The whole design still uses a cable chain, customized so everything can fit properly. The only drawback will be that a few cm of height will be lost.
Now, why wouldnât a direct extruder still use a bowden tube for guiding the filament? For this exact reason, I designed this little piece which can be mounted behind of the LCD screen on the Anet A8 and an E3D bowden plug can be screwed into it. The Titan Extruder was already designed to accept a bowden tube on the input side, so why not use this feature and couple a tube on both sides?
I will soon be printing these parts and trying them out. If everything goes right, I will release them on my Thingiverse.
Because I still had an awful clicking sound coming from my Titan Extruder, I contacted E3D support and asked them for help. Awesome as they are, they sent me a set of new bearings for the extruder!
The main source of the sound appeared to be this bearing located in the front cover where the large wheel is. The bearing would keep stalling and it would spin âfreelyâ inside of the plastic, causing audible clicks.
I removed the old bearing and installed the new one. Guess what, the clicks are gone! I would like to thank the E3D support again for helping me fix this incredibly annoying issue!
Few days ago, I noticed a sudden shift in one of the layers while printing a Benchy. Not a good sign, right?
Here is a picture of that shift, specifically on the X axis. I did notice for a while that moving the X carriage by hand felt a bit... difficult. I shrugged it off and let it print again, not knowing just how bad it would get.
Yeah.... At this point, the X motor completely stalled. Itâs was dead. Sort of.
I removed the motor and tried to spin it by hand. It spun freely... until I applied sideways pressure against the shaft. It appears that the bearings inside the motor died. It was possible to spin it for some time, and then it would hit something inside and just lock up. I tried to add some oil to the bearings as well, without much success.
Luckily, e-radionica.com sells some awesome NEMA17 stepper motors, so I got them as a replacement for both X and Y. Paranoia told me that the Y motor might suffer the same faith eventually unless I replace it right away.
I first just tested the motor to check if it runs normally...
And then I installed it. I had to use a bit of a different configuration of washers as these motors have shallower screw holes.
Just as a test, the X carriage moved smoothly now!
The Y motor was a bit more difficult to remove because of a lack of clearance to the screws...
But, I did manage to install the new motor eventually!
A number of reasons could have caused this whole thing. First is that these Anet motors are incredibly cheap, and so is probably their build. The other is that I might have over-tightened the belts. Whatever the case is, I would still replace them as they can only cause issues.
So, Iâve bought the E3D Titan Extruder, along with some extra parts which should help increase the quality of my prints. Titan should in theory be able to fix my skipping issue without adding oil. But does it? Time to find out...
All parts of the Titan Extruder. The packaging and everything looked quite premium.
Before I tried to do anything else, I checked if the motor works well. Luckily, it did, and it spun in the right direction as well!
Time to dismantle the old extruder. Pretty straightforward process.
There are a few things I wanted to do before installing the Titan, starting with the left side. First, I cleaned up the smooth rod from grease...
...and added two layers of kapton tape around the both ends of the rod. I noticed that these rods have a slight wiggle in their holes. Adding kapton will help fixate them in there.
Next up, I bought new Z couplers from E3D, as I read that the stock ones are rather bad. This should further increase the print quality.
First I removed the old coupler...
...and installed the new one. Pretty simple process. I suggest using something below it to add decent spacing from the motor so the coupler can flex properly.
Another thing I wanted to add were these anti-backlash nuts.
The image above illustrates what what backlash is. There is essentially a little bit of play between the nut and thread. To combat this, anti-backlash nuts use a spring to push up and down against the thread, effectively eliminating it. This will allow me to use Z-Hop, which means that the printer will lift its nozzle when moving between parts of the print.
Iâve removed the old nuts and installed the new ones. It is important to install them facing upwards.
The upper nut should be squished against the lower nut so it can achieve proper tension. The drawback of these nuts is that the motors need considerably more power to move the carriage up/down.
And finally, the rod was secured in the coupler.
Next up, I repeated the whole process on the right side! Fun!
I took this time to re-oil the rods as they needed some pretty badly by now, especially with the new nuts.
It is important to note here that I had to increase the current going to the Z motors on the mainboard. You can do this by turning the little potentiometer clockwise. I had to turn it around 90° clockwise so it has enough power.
Now, on to mounting the Titan Extruder! The bracket I will use to mount the Titan is of my own design and available on Thingiverse. I designed it so it can easily be mounted on the Anet A8 frame. A good friend of mine printed it for me as my printer had bad skipping issues (I havenât oiled it for over a month).
It is important to check beforehand how the Titan will be mounted. Seems good to me.
I wonât be going into a tutorial how to assemble the Titan Extruder, as you can find a comprehensive guide on the official website.
I mounted the Titan onto the frame using only one screw. I found that to be easier.
While doing this upgrade, I decided to swap out the bowden tube as well. I went with the new Capricorn by E3D. I was convinced reading their explanation why it is so much better, so why not.
The final touch... Plugging the motor in.
Similar like before on my blog, I recalibrated the steps of the extruder. I used a 6 times larger step number as a starting point as the new motor had 0.9° per step, instead of 1.8° like my old motor. There is also a 3x gear reduction within the Titan Extruder.
All assembled! Yet... My first experience was not so good at all. The new motor had absolutely not enough power to push the filament. It skipped way worse, even with the new 3x increase in torque. If I had a RAMPS 1.4 board, I would just up the current on the extruderâs driver, but that wonât be happening here.
To combat this, I simply returned the old motor. Looks like it has enough power to push normally now, even though the precision wonât be as high.
Not everything is fun and games though. I ran into a ton of issues. Looks like oiling the filament was still absolutely necessary. The new Titan Extruder did have enough power to push the filament, but the filament will still come out in bursts from the E3D v6 hotend instead of smoothly.
This worked as a nice morbid temporary solution. I used a stand usually used for holding PCB boards while soldering to hold my old oiler in place while I design and print a new one.
I designed a new oiler which is mountable directly on the Titan using one of the longer screws supplied with the kit. You can find my design on Thingiverse.
As a final touch, I moved the whole assembly so it is positioned diagonally. Helps a lot with smooth filament unrolling.
Here is one of the first serious prints I did. I have to say that I am very happy with the new printing quality. The wavyness I used to have on layers is completely gone. This D&D figurine was printed at 0.08 mm layer height and tons of supports. Rather impressive!
In conclusion though, is the Titan Extruder really worth it? Honestly, I am not sure. While the increase in quality is there, it did not solve the issue I was originally having: having to oil the filament. It did however increase the resolution of my extruder so I may print 0.08 mm layers better. The Titan also has this bad clicking sound on retraction. Adding some machine oil against the main gear helped reduce it, but I feel a bit let down that it is not perfect out of the box.
One beautiful sunny morning (Was it really sunny? I canât remember...) I discovered that my printer is reporting 0 °C on the hot bed sensor. Little did I know that this will turn into a multi-day saga of repairing my printer over and over again...
The Straining Wire: The Dirty Non-Fix
That definitely doesnât look good. It appears that the hot bed wires got somehow... eaten. Iâve read that cable chains ruin wires, and Iâve always found lots of wire dust (small red and black particles of the wireâs insulation) below the chain. Everything has been fine for ages, but it appears that it finally got to this point.
Quite terrible indeed. I am actually surprised how bad this is. Yes, these strands got pulled OUT of the insulation. One could say that the quality of these cables is pretty bad if they got THIS ruined.
At this point, I did a quick fix of ânot fixing itâ. I simply cut off the ruined wire and assembled everything again.
It made me hopeful to see it work again. Naturally, it stopped working again after moving the bed around a bit. I honestly thought that the ruined wire was simply shorting somewhere.Â
The Straining Wire: The Chain Amputation
At this point, I was getting a little bit pissed. I decided to completely remove the whole cable chain while re-crimping the cables.
Cable chain removed from the bed carriage and wires pulled out of the connectorâs housing...
Aaaand the cable chain completely removed from the printer. At this point, I had to put my printer on the floor to do these adjustments.
I re-crimped all of the current carrying wires. Incorrectly. I didnât know about this at this point, but this is most definitely the wrong way to do this. The cables should be held by both the insulation AND the wire strands, not just the strands.
The crimps have been re-inserted into the connector...
...and I added some zip ties to keep it in place. Yet again, the sensor was losing temperature info and reporting 0 °C. Honestly, the only reason why I didnât look at the sensorâs wires until now was because the bed would randomly turn on when I would wiggle the wires. This made me think that the current carrying wires were somehow shorting with the thermistor.
The Straining Wire: The Nerve Transplant
Considering I was now convinced that the issue was in the thermistor wires, I decided to take some two new blue and green wires and splice them with the old ones. I removed the insulation on both and prepared them for soldering.
I recently bought a fantastic new soldering iron. It is called TS100, and I got it off Amazon here. The soldering iron is as tiny as a pen, and runs a full blown firmware which controls the temperature digitally. It is a thing of beauty and I would recommend it to everyone who needs a good iron.
Considering that my printer was on the floor, I have set up a small soldering station like this, using filament boxes. I have to say that kneeling and soldering was quite annoying but TS100 made it a lot easier.
I soldered the wires again using the 3D printed soldering fingers I mentioned earlier on my blog.
Some nice ugly insulation...
...and I assembled the wire again. This time I paid more attention to properly zip tie the wires.
I also used some of the braiding plastic that came with the original Anet A8 to hide the wires.
While doing all of these modifications, I also decided to finally fix the floating front frame. For a while, I noticed that the front frame on my printer is floating by a milimeter off the table. This made leveling the bed a lot more difficult, and it was the actual source of what I thought was a warped bed. To fix it, I cut small pieces of rubber from the unused leftovers of rubber legs I usually use on all of my prints. It worked like a charm
I also put an ugly piece of plastic foil over the LCDâs push buttons. The bowden tube would always hit them and make loud clanking noise. It was annoying and unnerving, making me thing there is something wrong with my printer.
Back to business!!! Or so I thought... It lasted for about 28 hours.
The Straining Wire: The Blood Vessel Transplant
Lets face it: The current carrying wires are simply way too thick. I was using 1.5 mm2 wire and the connector crimps are rated up to 0.83 mm2. This made it impossible to properly crimp them. The thickness of the wires was also the cause of the cable chain âwire eatingâ.Â
Because I had no real strain relief now, all went to hell.The wires kept twisting bit by bit, and all of the strands snapped off. Both the current carrying wires and the thermistor wires as I didnât crimp them correctly.
I Removed all of the wires out of the crimps. Some crimps I couldnât even properly get out of the connector.
Removing the wires was annoying too. I had to remove the frame brace and pretty much gut my printerâs nervous and blood system.
The new wire I will be using is only 0.75 mm2. This wire is just right for the specs of the JST connector crimps. As usual, be sure to crimp the MOSFET end as well with proper fork crimps!
A single pair of the new wire was routed through the printer and measured for length. I also split up the pair for the length of the cable chain, primarily to avoid what happened before.
Inserting both pairs of the cables into the cable chain to make sure they fit...
THIS is how the wires should be crimped in the JST connector crimps. The larger metal piece needs to hold onto the insulation and the smaller onto the wire strands. And no, I didnât have a proper tool to do this, so it was all DIY with pliers.
A picture of all six wires properly crimped. I used new crimps for all of them, just to be safe. Luckily, my friend had a lot extra so I didnât have to wait on shipping.
All wires pulled through the cable chain...
...and inserted into the connector housing.
Everything reassembled with the wires braided. Notice how I left enough leftover length for the wire, in case of another re-crimping.
If you ask me, that now looks absolutely beautiful. And now, it actually works. It has been working for a while and I keep monitoring to see if everything still looks correct. The wires now have enough room to âbreatheâ in the cable chain, and all are properly crimped.
So, what is the morale of this story? I guess there are multiple.
Always use a strain relief. Not using strain relief WILL ruin your wires.
Cable chains are not as bad as people will make you believe. They are only bad if there is no room for the wire to breathe in them. In fact, they are very much needed for proper strain relief.
Donât use a thicker than needed wire. It wonât help you in any way and will only lead to more trouble.
Always properly crimp the wires. If you donât crimp them properly, they will definitely get ruined with time as the wire strands will twist around until they snap. The twisting part should be the insulated one instead.
So, last night I upgraded my head to this Frankenstein monstrosity. A completely new 40 mm fan for cooling the E3D heat sink. Why? Because it is quieter. So quieter that it is completely silent.
The fan I installed is SUNON HA40201V4-000U-999. I bought it from TMEâs website. The fan is exceptionally quiet as it uses some special MagLev motor drive, yet it can still puss through same volume of air as the old one. It is quite massive though. I printed a shroud for it and a hexagonal cover as well, just to keep it safe.
As an added bonus, I finally released the bowden carriage that I remixed! You can get it here! Live isnât all fairy tales and magic though...
I was printing some replacements in case something goes wrong one day.... and this happened. Same old story, motor skipping like crazy, ruining the print. So, my first thought was, is it the new fan? Bad shroud design? I tried to place the old fan again and it was all the same. I also tried to add more oil (I havenât oiled for a few weeks now), and still same result. But then I noticed that the filament is curling like crazy from the tip of the nozzle. There was only one thing left to try: Replace the nozzle.
Aaaand boom! That did it! I have a few theories what might have happened, but the biggest one is one simple word: dirt.
You see, Iâve had this sponge in my filament filter/oiler for a while. And I also used a lot of oil for some time on it. It attracted dirt. Tons of dirt. In fact, I think a piece of it broke off, entered the bowden tube and deposited itself in the nozzle. The filter did exactly the opposite. It messed up my nozzle.
Well, that is ÂŁ4.50 down the drain. I only have one more replacement nozzle, so I should probably order more.
I have completely updated the âMy Printerâ page on this blog with newest photos, list of upgrades and settings for my printer, as it is right now. You can go check it out HERE.
People talk about these now and then in the 3D printing community. Beautiful semi-transparent plates which reveal a monochrome image when viewed against a backlight. Letâs learn how to make them.
First thing you want to do is to acquire an image you wish to print. It can be a photo, some drawing or really anything. The image I chose for this was Luna look by RacoonKun.
it is important to know that the finished product will be monochrome. That said, it is a good idea to convert it to monochrome in a controlled environment before doing anything else. Use something like Photoshop or Gimp for this. This is now exactly how the finished image will look like after printing.
The tool we will be using for the lithophane conversion is called Image to Lithophane. It is a rather simple online tool, but exactly what we need. To begin, simply drag and drop the image to it.
Then, you will want to go to Settings and adjust the Image Settings like shown above. The only thing you need to change here is to set the image type to Positive Image.
In Model Settings, there are plenty of things you may want to adjust. In here, it is important to increase the Vectors Per Pixel. I am using the maximum value, which produced an STL of about 100 MB. A lower value will produce a pixelated image as shown below. The right image was made using a lower value.Â
Other settings include the Maximum Size, which controls the maximum height or width of the image. If you want anything larger than 10 cm, this is the value you wish to change.
You can also increase the Thickness, but I prefer to use the default 3 mm. A thicker plate will produce more contrast (darker blacks), but the printer will struggle with overhangs more.
You also definitely want to set a Border. The border will help with bed adhesion and will produce a nice frame around the final image.
Thinnest Layer designates how thin will the lightest part of the image be. The default 0.8 works if you are using a 0.4 nozzle. I havenât tried, but I would assume it is best to set this to use an exactly 2x larger value than the nozzle size.
Once you have adjusted everything, go back to the Model tab and hit the Refresh button (not the one in your browser!). This process will now take a while, especially with a larger number of vectors. Once done, press the Download button to download your STL file.
For a good print, there are only a few things you need to do:
Print the lithophane using white filament. It seems to work best.
Orient the lithophane vertically like shown above. This will ensure that the X-Y motor movements design the surface, not the Z motor, resulting in a higher resolution.
Turn the lithophane parallel with the hot bed. This will ensure that the bedâs movement doesnât make the lithophane shake violently, progressively resulting in a worse and worse print.
Use a large brim to prevent separation from the bed. It will most definitely separate otherwise.
Use a high layer resolution, something like 0.08 mm.
By following these steps, you will get some of the most beautiful lithophanes imaginable. Of course, it is implied that your printer is calibrated well beforehand. Lithophanes are extremely precise prints with little to no room for error.
titsay
pixelskylines
gracieabrams
jimmyeatworldofficial
stormborns
s-u-f-j-a-n-s-t-e-v-e-n-s
gummygunk