New Project -- 3D printed truss frame

[Matthew]

Hi everyone, hope all is well!

I thought I would jump in here and share the progress on one of my new projects. I’m not sure if any of you know the youtube channel Proper Printing, but he recently made a video about 3D Printed lightweight gantries. Is this system the next step in 3D printing? - YouTube

I thought it could be interesting to design a whole printer around this concept. So that’s exactly what I have been doing! I found out that you can print multiple of the segments attached together without supports which I think will be much easier than trying to clamp them all together. Anyways here is a screenshot of the current F360 project:

Still early stages of design, but if I can even get this thing to print I will call it a success

It will be a 220x220 build surface (just using an old ender 3 bed), not sure about what speeds it will be capable of yet. I will have to be careful however, as it will be very light, so I will have to make sure that I make it bottom heavy so that it won’t knock itself over. I have 2 motors dedicated to the bed, and another 2 motors at the bottom for the x and y movement (corexy as of now). I will also keep the power supply and the mainboard at the bottom of the printer, and possibly the 5kg lead brick I have sitting in my basement (not sure about that one yet). I will also definitely need to add some cross bars/ supports to the frame to keep it rigid.

If you have any ideas or suggestions I am all ears, the model is far from finished, I haven’t been able to really model anything for around a year now with school, so it’s nice to finally get back into the swing of things!

Thanks,
Matthew

[mykepredko]

I tried something like this years ago with limited success. I did not make the rails from small units like shown in the video, instead I made 150mm long pieces that had (two) 1/4" threaded rods (from the hardware store) going up through the centre (rather than click into the sides as with these pieces). I can’t find my pictures of the damn thing - I took it apart a while ago to make some space available. Sorry about that.

I was never able achieve anything with the stiffness you get from a piece of V-Slot 2020 rail despite several iterations and that was a real problem. One of the big problems was that the threaded rods could bend their ABS (in my case, not PLA like the video) base and top supports. Rather than providing rigidity to the structure, their main contribution was tension (like running a cable through the centre), holding everything together with the mechanical properties of the plastic providing the main characteristics of the rail.

As the threaded rods act like cables, you’ll probably need diagonal cross pieces to make the structure acceptably rigid. I did this and, while I got an apparently acceptably rigid result, access to the model and the bed was “problematic” (ie just about impossible).

In any case, you’ll need a Bowden set up as having the mass of the extrusion stepper is going to be problematic.

The final issue is, getting the gantry and hot end to run acceptably smoothly to get a good quality print. I made a Cartesian motion system with Delrin V Wheels as the runners (the rails were designed to take the insides of the wheels).

Delrin V Wheel Kit

This didn’t work at all as the tension needed to keep the wheels on the rail edges distorted the rails at different parts - maybe using three rods, like in the video on each corner, rather than two running up the centre would be better. I ended up going with standard MGH12 linear rails and that worked better but not great.

I haven’t thought about it for a while but, based on what I’ve learned over the past couple of years, I would do a few things differently with a big one being the controller and how the bed was handled as well as being a lot more adept at using CAD software. I used an Arduino and RAMPS 1.4 with a basic Marlin Cartesian firmware for control with a cantilevered bed. Because of the diagonal side rails, the model was accessed by moving the gantry to a corner position and moving the bed up to access things from the top.

In terms of final weight, it ended up being around 25 pounds and, to print the printer pieces, I was using a Sapphire Pro which weighs 22 pounds. I also found that things ran best at 20mm/s (even with the diagonal supports, there was still a lot of flexing when moving the hotend around) which is definitely not optimal. So, I did not gain the expected benefits of a lower mass printer which could run at higher speeds.

The positives of the project was a) costs were quite reasonable as I got the threaded rods from Home Depot and 1/4" pieces are a LOT cheaper than say an M3 threaded rod and all the pieces were recycled from other printers and b) it’s an easy structure to put together and take apart (much easier than using 2020 V-Slot rails).

I’ll be interested in hearing what your experiences are - keep us updated as to your progress.

Good luck!

[Matthew]

Thanks!

Yeah I am going to be running bowden for sure, I am using smooth rods with the curved bearings (I can’t for the life of me remember what the curved bearings are called) and from some initial testing with smaller pieces it seems to run quite smooth and consistent. But we will have to see how it ends up going on the finished print. I am using a manta board with klipper, and it will have bed leveling with an old BLTouch I have laying around. I am also planning to print in ABS which should hopefully help out with the rigidity.

All in all I have ~$100 in parts put into this printer, so even if it’s a complete failure I wont be too disappointed! Even just getting back into F360 was worth it to me.

Matt

[mykepredko]

FYI. Here are some of the pieces I came up with and tested.

[Davethegearhead]

I’ll be watching this . Very interested on how the gantry turns out.