3D Printer


Tronxy X3S Aluminum Frame LCD Screen 3D Printer DIY Kit 

     

I have been working with 3d printing since around 2002, when I designed and printed some parts for an intake manifold for a college race car project. Then the options were binder powder parts or SLA, neither of which were strong enough for functional parts. We printed the parts using binder powder, and then pulled a silicone mold off of them and then casted urethane parts in quantity for the project. 

 

Additionally, during various work projects, we had prototypes made on 3d printers for form/fit/function testing, but function was usually pretty weak due to materials being weak. Then the FDM or fusion deposition printers came about, and more functional materials became available, and printers became more common, and also came down in price. 

Now getting functional parts was much easier and economical for projects. 

 

Then recently I started working on the ATTC project, specifically the radio control transmitter enclosure.  Radio waves do not like metal enclosures, and I wanted a lighter material to save weight and be more portable. The enclosure would be large, roughly 11 by 17”, and needed to be able to withstand high temperatures, as it could be stored inside a vehicle in the Texas sun. Vehicles can easily reach 180F in the sun, which rules out PLA material as it softens at 140F. ABS would be the choice, but it is more difficult to print.  

Designing the part was no problem, as that’s what I do, but getting it printed economically proved to be a challenge. Due to its size, it would not fit on any commercial beds, so I split it into 4 segments. Still too expensive, so I looked for alternatives. At first I looked at buying a printer, but my lack of knowledge in them steered me away from it. Fortunately, the local Makerspace had recently opened and had a TAZ Luzbot printer with a 300x300mm heated bed. After much learning and help from members, I finally started printing some usable parts for the project in ABS. First was a double 4 bar linkage to provide for a fully adjustable joystick off the enclosure.

Then I started printing the enclosure parts, each quarter a long process, with each one taking 17 to 25 hrs to print. 

Then the Makerspace member, Billy, found a reasonably priced printer kit with a large footprint. So we both bought one at the same time: a Tronxy 300x300mm bed with 400mm tall capacity. 

Driving back and forth to the Makerspace is inconvenient, especially during long prints, so I looked forward to having my own. 

Many weeks later, the printer kit finally arrived, and I started building it. The mechanical portion of the build was easy, and the instructions were relatively straightforward but not particularly well written. The biggest challenge was missing hardware in the kit. Frustrating to be stopped because of a missing bolt or nut. 

Build pictures:

I finally got it working, sort of…. 

Getting it to print was challenging, working through the software, and the interface to get it to print… Definitely not my idea of fun. 

I’m using Repetier Host for software, and Slic3r for slicing layers. 

First thing I noticed was that getting that critical first layer right was not easy to set. The TAZ had an adjuster screw for fine adjustment of the nozzle height to the bed. This one did not.. You would have to loosen a screw on the limit switch move it, then re-tighten the screw. It would never be right. So the first job was to print an adjustable z-axis assembly to make life easier. 

After that was installed, I could get that critical first layer to print. Starting out with PLA, since that material is easier to print with and does not require bed heat, I attempted to print more parts for testing. It seemed like it was not pushing out enough filament, and adhesion between layers was not good. Research showed to check actual versus theoretical filament useage, and sure enough, it was not pushing out enough. Supposedly it can be changed in the firmware, and while it would work temporarily, it changed back to default after a restart. I looked into flashing the firmware, but I could not find clear enough instructions to make that work. 

I finally resorted to changing the factor in Repetier Host and got it to work. That helped immensely with print quality. 

Then I designed a large square grid part so I could print and set the bed height easier. Getting the bed exactly flat to the nozzle was another improvement. 

Since PLA was now printing fairly decent, I needed to transition to ABS, which needs bed heat, and a lot! 110C to be precise, and so far I could only get the bed to roughly 60C, due to limitations in the stock power supply. Stock is 12v, and has to do a lot of work: nozzle heat, extruder motor, x-axis, y-axis motors and 2 motors for z-axis. Not much left for the bed, which really needs 24v. My friend recommended a mosfet and separate power supply for the bed. I had learned that an enclosure around the printer would help keep heat inside and protect the print from drafts and cool air that cause warp. He also recommended some cork to insulate the bottom of the bed, and gave me some since he had leftover from his printer project. I used silicone caulk, liberally applied and some sand bags to hold it down while curing.

This marginally helped with bed heating, but not enough.  I decided to build the enclosure around it next while waiting for parts. Aluminum tube, plastic corners and ¼” plywood makes up my enclosure. Aluminum angles, t-nuts and bolts anchor the structure to the printer frame adding much needed rigidity to the printer. I also built in a small bathroom heater fan to provide heated air inside.

     

The front plastic sheeting is velcro’d on so I can easily flip it up for the start of the print.

Before getting the separate power supply, I decided to try some printing some ABS parts. Well, the heater fan worked a little too well, and got the inside hot enough during a print that the 2 main angles that support the x-axis gantry and raise it for the z-axis that are made from PLA and they softened & sagged! The whole x-axis gantry was now loose, and the print had shifted all over the place.

X-axis gantry removed:

Rather than trying to print new pieces, I made them out of steel flat, using my vertical mill to machine slots and TIG welded them together. They are now the most rigid pieces on the printer.  

 

I had also been dealing with the y-axis shifting during prints. The belt seemed tight, so I researched and learned that adjusting the motor voltages up may help too. It helped, but only slightly…. More to come, back to the bed heat. 

 

The mosfet takes the signal from the control board and regulates the amount of power to the bed. I ordered the mosfet from amazon and got a 24v 320w power supply from China while on a work trip there.  Once home, I installed the mosfet and large power supply, and now the bed heat worked great! It could now achieve 110c on the bed, and did not need additional enclosure heat. 

Now to try printing some more parts in ABS… Printing well, except for some layer shift in Y-axis. Upon closer inspection during a print, I finally observed it skipping on the cogged gear and belt. There is no easy way to adjust tension on this belt with the stock setup, so I printed a Y-axis tensioner assembly and it fortunately printed without skipping!  Then I removed the bed, and assembled it in place. Luckily I had not trimmed the belt during initial assembly, so I had enough belt to add it. 

 

After adding it, and really tightening it up, it still was getting some shifting, so slightly loosening it and also slowing down print speed seemed to resolve that. 

 

Now I finally had a functioning printer that would do ABS parts! 

 

So I printed more parts for the ATTC project, like a new handle for the heater switch, handle protrusion for the control enclosure, another bracket for the enclosure and a few other parts. It’s nice to have another tool for custom parts and I have certainly learned a lot myself! Much more to learn, and many more parts to come! 

All Terrain Track Chair (ATTC) parts that I’ve printed recently:

Share