Pneumatic System / Vacuum

The following vaccum system was realized. To avoid influencing the vacuum level while picking and placing in the two nozzles, they were separated and two pumps used. To switch the vacuum, directly actuated valves are used.

The pumps are mounted to the machine each by a flexible 3d printed part. Its made from flexismart so the vibrations don't spread over the machine and its little more quiet. Additionally a top cover was printed to have even less noise. The holder is designed for the vacuum pump from Robotdigg which is about 37mm in diameter. To finally fix the pump 3 cable ties are guided through the printed holder.

Next to the mandatory diode, some caps were installed to the vacuum pump from Robotdigg. Had EMC issues (microcontrollers crashed irregularly, etc.). Used 3 ceramic condensators to get away with that, like presented here. Only one condensator wired to + and - didn't do the trick.

  • Vacuum pump EMC: 1x100nF from + to Case, 1x100nF from - to Case, 1x100nF from + to - (had issues before, recommended
  • Vacuum reservoir: 1l Laken Alu Bottle with 3d printed screwed cap
  • Nozzle: Juki 506
  • Nozzle Adapter: Tool changer from Peter Betz
  • Vacuum Sensor: NXP MPXV6115VC6U
  • Motor Driver: A4988 on CNCshield with GRBL
  • Motor Current: 0.65A (already slightly overpowered, motors getting really hot)
Images / Files Description vac. nozzle open [kPa] vac. nozzle closed [kPa] vacuum sufficient? loosing steps? Evaluation
reference: vaccum tube only. value for “nozzle open” is tube just open ended, value for “nozzle closed” is with a finger on the end -3.2 -49 - reference only -
reference: vacuum tube directly mounted to nipple end of nema8 hollow shaft stepper motor -10.9 -36 - reference only -

see stl-file

3d printed adapter with o-ring ID: 7,8mm pressed against motors rear end (left side in photo) -7.3 -13.5 no no vacuum so low that parts got lost sometimes
n/a 3d printed hose with r-ring ID 4,5mm on nipple end of motor n/a n/a ? ? not tested, because DIY store had no sealings with ID4.5mm available
rotary-joint-robotdigg.jpeg adapter_vacuum2hollowshaft_20170827_203052.jpg
rotary joint from robotdigg connected via flexible tubing to hollow shaft (right side in photo) -12 -35 yes sometimes, especially after long still standing machine torque load of these adapters applied to the nema motor seem to be very near at their limit
rotary-joint-sanga.jpg rotary joint from local supplier, manufacturer Sang A -12 -35 yes yes too expensive compared to robotdigg adapters and not less torque as was expected
3d printed rotary joint with 2 rubber sealed bearings -10 -20 to -27 (actual value depends on how tight the printed adapters fit the bearing) yes(?) no vacuum OK, quite leaky, nearly no additional torque
3d printed adapter with o-ring 5mmx2mm pressed slightly against motor rear shaft, same principle as with original OpenPnP part but adapted to my machine -9.5 -25 to -30 yes no realized that, little grease reduces friction significantly
  • maschine/pickandplace/vacuumsystem.txt
  • Last modified: 2017/11/10 17:48
  • by michael