Quadrirotor made in 3D printer

jujucam70

Member
Helo guys,

I'm french and my new passion is the "3d printer"

I have one dji phantom,one F450 and now, i want to create a new one only using 3D printer.

You can visit my website ( you can traduce in english) and see my project.


www.rc-impression3d.fr


Thanks and see you...
 



SoCal Blur

Member
I have thought about building some components using a 3D printer and ABS but I don't know how strong it would be and if it will be too heavy.
 



SoCal Blur

Member
I would say having a solid frame/parts is very important haha just my $.02.

Agreed. You want a ridgid frame with very little or no flex. Some have attributed the S800 EVO "flip of death" issues to the first gen arms that were very flexible. I don't know if that really is the primary cause but arms flexing was probably at least a factor.

That's why I asked the question - it seems to me that in order to get the same or similar regidity as CF you would have to make thicker parts out of ABS and therefore add additional weight. I'm not saying it's not worth trying...in fact I hope it works as I will be getting a 3D printer next month as well.
 

jujucam70

Member
Agreed. You want a ridgid frame with very little or no flex. Some have attributed the S800 EVO "flip of death" issues to the first gen arms that were very flexible. I don't know if that really is the primary cause but arms flexing was probably at least a factor.

That's why I asked the question - it seems to me that in order to get the same or similar regidity as CF you would have to make thicker parts out of ABS and therefore add additional weight. I'm not saying it's not worth trying...in fact I hope it works as I will be getting a 3D printer next month as well.


I'm agree when you say "rigid frame with no flex" but you certainly know that parts of frame are not flexible if you have a big volume.

The structure is like a "bee socket" empty less than 10% so : it's strong and not heavy...

(sorry for my poor enlish)
 

RTRyder

Merlin of Multirotors
I have thought about building some components using a 3D printer and ABS but I don't know how strong it would be and if it will be too heavy.

No need to worry, I've been making entire multirotors out of 3D parts using both ABS and PLA...





And also making parts to mod existing frame plates...



All of these are currently being flown and they fly quite well. The Y6 is great for windy days, the little orange one is my low and slow FPV quad, I like to slalom through trees with it.

A peek at what's currently on the bench...



I think this one is going to be the best one yet!

Ken

P.S. if you print a part using a rectilinear or honeycomb infill of 25 to 30% it winds up quite light and as far as I can tell just as strong as if you printed it solid
 
Last edited by a moderator:


RTRyder

Merlin of Multirotors
That's nice work and very encouraging. What about on larger multis in the 1000mm class or larger?

With carbon fiber arm tubes you could probably make a large portion of it using 3D printed parts. On something that big I'd probably opt for 2 or 3mm C/F center plates, especially if the intended use is carrying heavy APV gear, at the very least some decent G10 if C/F isn't an option.

The Y6 center plates are ABS, the top has a web of stiffening ribs around the edges and through the center, the bottom is 4mm thick with 3 solid layers top and bottom and the middle layers are 30% honeycomb infill. When the arm blocks are bolted in place the entire center structure is very solid yet light, it doesn't flex at all when assembled and flys amazingly smooth, the view through the FPV pilot cam is as steady as you could want from a multirotor.

It all depends really on a couple things, one the quality of the prints produced by the printer, and your ability to design what you want in a CAD program and then translate that through the steps to make the design into gcode that the printer can understand.

Ken
 
Last edited by a moderator:


deluge2

Member
Thanks for the earlier pics of your MRs! Your comments, below, reflect important factors for all of use would-be 3D printers to consider carefully. Can you share what printer(s) and CAD sw you would recommend?

Steve

<snip>

It all depends really on a couple things, one the quality of the prints produced by the printer, and your ability to design what you want in a CAD program and then translate that through the steps to make the design into gcode that the printer can understand.

Ken
 

RTRyder

Merlin of Multirotors
Oh nice job RTRyder!

What about the weight of them?

Thanks! As for the weight, I'd say comparable to a multi built from "normal" materials, possibly a bit lighter depending on choice of material and how it's printed. I tend to over-engineer at first until I see how the pieces hold up to some hard test flights and then refine the design to parts that are as light as possible without compromising strength. I have access to a very large open field that is private property and the first flights are done there where there is no one around and no possibility of damaging someone else's property if something fails in flight. I've only had one part fail in flight in the very early days when I didn't fully understand the printing process and what to do and more importantly what not to do.


Thanks for the earlier pics of your MRs! Your comments, below, reflect important factors for all of use would-be 3D printers to consider carefully. Can you share what printer(s) and CAD sw you would recommend?

Steve

My primary printer is a highly modified Solidoodle3 that has been upgraded to an enclosed and heated build chamber (a MUST for ABS IMO...), and a more efficient platform heater topped with a layer of borosilicate glass for the flattest build surface possible, plus about a dozen other things to make it more reliable and a lot more accurate. The printer runs a version of open source Marlin firmware on a 2560 Arduino processor (same as you find on a Multiwii Pro board :) ) and uses Repetier Host software for controlling printer functions. As for CAD programs, there are many to choose from from free to way more $$$$ than I'll ever spend. Currently I have Autodesk 123D and Sketchup for design work which then gets exported to STL format which I then run through Netfabb to clean up any holes and other problems with the mesh, then export back into a clean STL file that gets run through Slicer to produce gcode that the printer can use.

It sounds a lot more complicated than it is, once you get used to the workflow it only takes a few minutes to go from finished design to having it printing on the printer. The most important part is that whatever printer you choose, and there are a LOT of them out there, be as accurate as possible to be able to repeatedly produce parts that do not vary dimensionally from first to last, that narrows the field considerably. IMO only printers that have build plates that are fixed in the X/Y plane are worth considering. It's OK for the platform to move vertically on the Z axis while only the printhead moves in X and Y, but if you have the platform moving horizontally back and forth on the Y while the printhead is moving horizontally back forth on X plus vertically on Z there is far too much motion going on and places where things can loosen and lose calibration to be able to maintain tight tolerances and layer alignment print after print, after print.

Bottom line is do LOTS of research before you buy, it can save a lot of aggravation later. Also get the largest printable area that your budget will allow, if you buy a low end unit with a 5 X 5 X 5 workable area you'll soon be wishing you had something larger and upgrading to get that extra few inches on each axis means either a new printer or about half the original cost of what you have to upgrade if it can be upgraded to larger print volume.

Ken
 


RTRyder

Merlin of Multirotors
Hello ken,

Do you finish ABS parts with acetone ?

Or do you use more PLA?

regards.

I use the ABS parts right off the printer, never used the acetone smoothing. 99% of the parts I make for multirotors are done in ABS, I only use PLA where I want the part to be less flexible but generally not in critical areas like motor mounts. So far my use of PLA on frames has been very limited, it's stiffer but breaks easier than ABS and is more affected by temperature being more brittle in the cold and softening at a much lower temperature than ABS.

Ken
 

Simspeed

Member
Hi Ken,

I'm looking into having frame parts 3D printed for the design I've posted on the forum here..http://www.multirotorforums.com/showthread.php?18278-X8-Design-Study . Below are a couple of components I've created for consideration. The hinge post is a structural member of the frame where the folding arms pivot. I first planned to have it machined in aluminum...but I've been wondering if ABS or Nylon would be strong enough to handle the stresses. The right angle posts will be threaded on the outer ends for the locking nuts and the center hole will be threaded as well for the lower battery box mount to screw into. What has been your experience with those materials considering the compact density of the part? It is about 11.9 cm3 in volume.

The second part is an ESC mount that clips into the frame and bolts down. The ESC slides in under the top strap and snaps into place with a tight fit. The volume there is 19.1 cm3. No real stress on this piece but it does have some odd geometry to seat it for printing. Ever printed anything like that? Thanks...TP.

View attachment 17183View attachment 17184View attachment 17185View attachment 17186View attachment 17187View attachment 17188View attachment 17189
 

Attachments

  • FF.69.JPG
    FF.69.JPG
    67.3 KB · Views: 418
  • FF.70.JPG
    FF.70.JPG
    59.7 KB · Views: 382
  • FF.71.JPG
    FF.71.JPG
    73.2 KB · Views: 287
  • FF.72.JPG
    FF.72.JPG
    51.9 KB · Views: 323
  • FF.66.JPG
    FF.66.JPG
    58.1 KB · Views: 266
  • FF.67.JPG
    FF.67.JPG
    44.1 KB · Views: 263
  • FF.68.JPG
    FF.68.JPG
    59.2 KB · Views: 283
Last edited by a moderator:

RTRyder

Merlin of Multirotors
Hi Ken,

I'm looking into having frame parts 3D printed for the design I've posted on the forum here..http://www.multirotorforums.com/showthread.php?18278-X8-Design-Study . Below are a couple of components I've created for consideration. The hinge post is a structural member of the frame where the folding arms pivot. I first planned to have it machined in aluminum...but I've been wondering if ABS or Nylon would be strong enough to handle the stresses. The right angle posts will be threaded on the outer ends for the locking nuts and the center hole will be threaded as well for the lower battery box mount to screw into. What has been your experience with those materials considering the compact density of the part? It is about 11.9 cm3 in volume.

The second part is an ESC mount that clips into the frame and bolts down. The ESC slides in under the top strap and snaps into place with a tight fit. The volume there is 19.1 cm3. No real stress on this piece but it does have some odd geometry to seat it for printing. Ever printed anything like that? Thanks...TP.

View attachment 21251View attachment 21252View attachment 21253View attachment 21254View attachment 21255View attachment 21256View attachment 21257


On the hinge post an ABS part printed using filament deposition would not hold up to the stress plus while it is possible to cut threads into and on the surface, they don't have nearly the strength that threads cut into metal do. It works OK for parts that aren't subject to being stressed but I wouldn't do it anyplace that is a critical part of the structure. A better method when using 3D printed parts is to use a through bolt with nut putting the part into compression against a harder material such as aluminum or carbon fiber. Even then the clamping force will cause the plastic to deform over time leading to loosening of the part if too much force is applied, there's a fine line between just enough and too much. Parts made using an SLS (Selective Laser Sintering) process are stronger than parts made using filament deposition but the machines to do that type of printing are quite expensive and consequently parts made on them also tend to be quite expensive. Shapeways gets something like $1.75 per cubic cm of printed volume so if you do the math you can see it adds up quickly. I've done some parts in nylon but it's difficult to work with unless you're using SLS type machines, on FFD (Fused Filament Deposition) printers the trick is trying to get the nylon to stay stuck to the print surface long enough to fully print the part.

Here's a couple ESC holders I'm experimenting with now...

View attachment 17191View attachment 17190

The issue with the design you have is that it would be difficult to print in its current form, would be easily done as two pieces though.

I'm out of time, have to head out the door for a meeting in downtown Boston...

Ken
 

Attachments

  • ESC_Holder.jpg
    ESC_Holder.jpg
    113.5 KB · Views: 340
  • ESC_Holder_V2.jpg
    ESC_Holder_V2.jpg
    90.7 KB · Views: 328

SoCal Blur

Member
One solution could be to 3D print the parts and then make silicon molds and cast the parts in high strength mineral filled resin. Its a relatively inexpensive process you can do yourself (with a little practice). Mineral filled resins are very strong and are routinely used to prototype parts that will be exposed to high stresses. The downside is that the resin is heavier than ABS but depending on the size of the part, the increase in weight could be negligible.
 

Simspeed

Member
Thanks Ken. Good advice. I've looked into the SLS process previously and there are a number of companies offering the service. Because of the complexity of design for the upper and lower corner post, I've been leaning that way. There is a local Denver company I'll contact for pricing once the design phase is finished and all the parts have been dimension finalized. Thanks...TP
 

Top