NinjaFlex

This page will be updated as we make more test prints

Test shape
27/9/2016
IMG_20161013_180109

  • Layer height: 0.2
  • Temperature: 240
  • Print speed: 30 mm/s
  • Travel speed: 100%
  • Wall thickness: 1.2
  • Width of nozzle: 0.4
  • Infill: 15%
  • Buildplate: 40 °C
  • Material flow: 145%
  • Fan speed: 100%
  • Support: No
  • Brim: Yes

Comments:
During the printing of the test cube we tried to optimise the setting of the printer as much as possible. We found out the material feed should be a lot more then the standard 100%. We also found out that it was too hard for the feeder to unspool the elastomer. There were clearly marks in on the test cube where not enough material had been deposited due to the resistance from the spool. What we learned from the test cube was that an overhang until 65 degrees was possible. The smallest possible diameter of vertical holes is 2mm that of horizontal holes is 1 mm. The horizontal holes however have a rough finish. We also found out small details are hard to print due to the high viscosity of the material.

Blue finger
29/9/2016
IMG_20161013_180130

  • Layer height: 0.2
  • Temperature: 240
  • Print speed: 20 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 1.2
  • Width of nozzle: 0.4
  • Infill: 100% (It was hollow)
  • Buildplate: 40 °C
  • Material flow: 145%
  • Fan speed: 100%
  • Support: No
  • Brim: Yes

Comments:
In the blue finger the layers were more consistent than in the test cube the only noticeable surface defect was the webbing due to excess material coming from the nozzle. In the next prints we will try to find a solution for this problem. The finger however seemed to be airtight. The only problem we have is finding a way to attach the pressurized air system to the finger. This finger only has walls and no solid parts so there is no drillable surface to attach the threaded nozzle to. Creating an airtight seal between the flexible material and the nozzle will probably be the most challenging.

Finger – #1
3/10/2016
IMG_20161013_113524

  • Layer height: 0.2
  • Temperature: 240 °C
  • Print speed: 20 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 0.8
  • Width of nozzle: 0.4
  • Infill: 100 %
  • Buildplate: 40 °C
  • Material flow: 140%
  • Fan speed: 100%
  • Support: No
  • Brim: Yes

Comments:
In this finger we printed a solid part to attach the nozzle to. We tried drilling a hole in this solid end to insert the nozzle into. The material is however not suitable to be drilled, it rips instead of cuts, this means the drilling leaves a terrible surface which makes the connection leaky. Possible solution is to print a hole, and afterwards heat it up to make it smooth and tight. Not airtight, but it is possible to get a movement when we blow air into it. The rest of the finger seems to be airtight. Testing with pressurized air also show the finger displays the expected behaviour. We expect that it won’t work with to compressor (to control the movement) because it is leaking too much.It did require a little bit of work after the print, since we had to cut away the filament between the air chambers, and fix the brim. The challenge remains to attach the nozzle in an airtight way. According to the literature the closer the air chambers are together the easier moment will occur. we will try to optimise this by decreasing the distance between the air chambers.

Finger – #2
3/10/2016
IMG_20161010_093618IMG_20161010_093724

  • Layer height: 0.25
  • Temperature: 240 °C
  • Print speed: 25 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 1.2
  • Width of nozzle: 0.4
  • Infill: 100%
  • Buildplate: 45 °C
  • Material flow: 135%
  • Fan speed: 75%
  • Support: No
  • Brim: No

Comments:
To try to get the weebing inside of the finger to stop we tried enabling retraction with a small retraction prime ammount. this however resulted in a lot of additional webbing. halfway the print the ultimaker crashed and burned the finger so we aborted the print.

Finger – #3
6/10/2016
IMG_20161007_121758

  • Layer height: 0.25
  • Temperature: 230 °C
  • Print speed: 25 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 1.2
  • Width of nozzle: 0.4
  • Infill: 100%
  • Buildplate: 40 °C
  • Material flow: 120%
  • Fan speed: 75%
  • Support: No
  • Brim: No
  • Retraction disabled

Comments:
The print is almost perfect except for the fact that it is not airtight, especially the sidewalls. the webbing is almost gone, because the retraction was disabled, also the 10.5 mm printed hole seems to line up with the nozzle perfectly. We now have to find settings in which the sidewalls are airtight.

After using a hot air gun to sinter the sidewalls the were relatively airtight. the only problem is that because the heat made the material softer that the material attached to the nozzle ruptured.

FDM – Test blocks with different wall thicknesses
7/10/2016
IMG_20161010_093828IMG_20161010_093856

  • Layer height: 0.25
  • Temperature: 235 °C
  • Print speed: 30 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 2mm, 1.6mm, 1.2mm
  • Width of nozzle: 0.4
  • Infill: 100%
  • Buildplate: 50 °C
  • Material flow: 120%
  • Fan speed: 75%
  • Support: No
  • Brim: No

Comments:
The blocks showed that the walls are only airtight from 2.0 mm. the 1.6 mm block was close to being airtight and could be made airtight by heating the part with a hot air gun. we however would prefer to print something that would be airtight immediately after printing because not all shapes are as easily uniformly heated as a block.

Finger – #4
7/10/2016
IMG_20161010_094129IMG_20161010_093925

  • Layer height: 0.25
  • Temperature: 235 °C
  • Print speed: 30 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 2
  • Width of nozzle: 0.4
  • Infill: 100%
  • Buildplate: 45 °C
  • Material flow: 120%
  • Fan speed: 75%
  • Support: No
  • Brim: No

Comments: 
The printed finger is almost completely airtight but show marks of burnt plastic. We suspect a spooling problem to be the cause of this and will investigate further. Due to the relatively low air chamber hight the stiffness of the finger is quite high compared to our previous prints. Fitting the nozzle did again leave small ruptures around the fitting hole. We will investigate the proper geometry of the nozzle hole.

Finger – #5
10/10/2016
IMG_20161013_175938 IMG_20161013_175858

  • Layer height: 0.25
  • Temperature: 235 °C
  • Print speed: 30 mm/s
  • Travel speed: 75 mm/s
  • Wall thickness: 2
  • Width of nozzle: 0.4
  • Infill: 100%
  • Buildplate: 45 °C
  • Material flow: 120%
  • Fan speed: 75%
  • Support: No
  • Brim: No

Comments:
The exact same settings as in the previous print were used. This time the relative height of the air chambers was increased in comparison to finger #4. The result is a more flexible finger. We have uploaded a video of this finger moving that you can find here.

Finger – #6 (Valve fitting)
14/10/2016
IMG_20161017_163032

Same settings as #5

Comments:
This was printed to test the new valve fitting; a plastic push-in air fitting, which was smaller and had 4 mm unthreaded plastic pipe, going into an equally small hole in the printed finger. We glued the valve to the finger using LOCTITE 4850 for Flexible materials. This worked surprising well and was completely airtight. Therefore it was very satisfying since it also reduces the size needed for the fitting.
The valve can be found here.

Double finger – #7
14/10/2016
IMG_20161017_163052

Same settings as #5

Comments:
We tried toi print 2/3 of a tentacle, but the design proved to be less flexible, hence less functional due to a very rigid ‘core’ between the two rows of air chambers.

Finger – #8
IMG_20161017_163138 IMG_20161017_163214

Same settings as #5

Comments:
We printed the first version of the finger again with the new valve fittings, and saw a really nice flexibility and functionality when applying air pressure. We were able to apply a lot of pressure, and make it go all the way around to create a circular shape, and at the same time exert quite a lot of force to the in the movement.

 

Double-sided finger – #9
17/10/2016
IMG_20161018_153258

Same settings as #5

Comments:
We tried again with the double-sided finger (to be able to bend in two directions). This was unfortunately not very succesful either, since the small space between the air chambers again caused a very rigid finger, since the to of the air chambers would touch even at a very small angle.

 

Finger – #10
17/10/2016
IMG_20161018_153236

Settings were taken from article “High-Force Soft Printable Pneumatics for Soft Robotics Applications”

  • Layer height: 0.1
  • Temperature: 245 °C
  • Print speed: 40 mm/s
  • Travel speed: 120 mm/s
  • Wall thickness: 1.2 mm
  • Width of nozzle: 0.4 mm
  • Infill: 100% (hollow)
  • Buildplate: 60 °C
  • Material flow: 100%
  • Fan speed: 75%
  • Support: No
  • Brim: No
  • Retraction: No

Comments:
We tried to print with the settings from the above mentioned article. The main difference was the temperature setting, which resulted in a more transparent and glass-like finger. Unfortunately, it left a big hole in the top layer, so we weren’t able to test this design under air pressure.