Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Help with robotic arm - fuel robot

  1. Oct 29, 2008 #1

    I have a college project running for which I have to design an automatic car refuelling cell. This includes automatic robot positioning, automatic payment, assisted car alignment et cetera.

    I like to stress that neither I nor my fellow students are mechanical or electrical engineers. We are system designers: we look at what technology is out their and we choose and combine what we think fits the design goal.

    Anyway... I started this topic to find a solution for a problem I am having concerning the robotic arm that I would like to use to position the fuelling nozzle. Since the position of fuel openings on cars is quite random (atleast in Europe it is 50/50 left/right and then some variation in orientation) I want to have one flexible robotic arm that can serve the majority of cars.
    That is why I would prefer a ceiling mounted arm that is mounted on xy linear axes to go from one side to another.
    I came across one type of robotic arm from Fanuc which has excellent specs with regard to my needs, namely the M-710iC/20L (Fanuc website). This robot is very well suited since it has an action radius of +3100 mm.

    I have a couple of concerns/questions about this type of robot:
    • Can I just mount this type of robot to a ceiling without compromising performance?
    • This robot weighs 540 kg, which is quite heavy for mounting on xy linear axes. If cost is not important would I be able to reduce mechanical weight by using f.i. Ti-alloys or Carbon fiber composites?
    • My application would only require a 5 kg work load (dry-break valve) and 5 kg load on the arms (fuel hoses) while the Fanuc can bear a 20 kg work load. What effect would half that work load have on mechanical weight and motor torque?
    • To summarize the above: How low can mechanical weight get if cost is not of importance and taking into account aspects like 5 kg work load, 3 m long action radius ...

    I also thought about mounting a smaller robotic arm on a vertically oriented axis, on which the arm can move up or down, but I cannot decide of this would be a better option or not.
    http://img48.imageshack.us/img48/4234/34427566vo2.th.jpg [Broken]http://g.imageshack.us/thpix.php [Broken]

    Any suggestions are welcome!

    P.S. I have put this topic here, since I thought it was a too practical problem to put in the Education section.

    Last edited by a moderator: May 3, 2017
  2. jcsd
  3. Oct 29, 2008 #2


    User Avatar
    Science Advisor

    Just my opinion, but I think you'd be better off with a vertically mounted cartesian robot (on a reversable mount, so it can reach out to either side of the pump), and trust your patrons to park their vehicle on the correct side of the pump.
  4. Oct 29, 2008 #3


    User Avatar
    Science Advisor

    Just my opinion, but I think you'd be better off with a vertically mounted cartesian robot (on a reversable mount, so it can reach out to either side of the pump), and trust your patrons to park their vehicle on the correct side of the pump. It would only require 4 axes (including the rotating base), and some air cylinders. Should be lighter than the articulated design, and cheaper, too.
  5. Oct 29, 2008 #4
    Thanks for your advice, LURCH.

    Independently of your comments, I am shifting away from an articulated design as well. I'm not convinced however of the benefits of a vertically mounted cartesian robot versus one that is horizontally mounted.
    With a vertical mount you would still need 2 installations for 2 lanes. With a horizontal mount, I like the flexibility you have with left/right fuelling; it could be installed as a stand-alone fuelling unit.
    A ceiling mount would also intrinsically provide better protection against weather, thus lowering needs for maintenance (although I admit this is merely speculative).

    Just want to know what your thought is on this.
  6. Oct 29, 2008 #5


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi driesel,
    Have you had a look at what Linde did at the Munich airport? They have a fully automated system for liquid hydrogen powered vehicles. There's some general information about it here, including pictures:
    http://www.ieahia.org/pdfs/munich_airport.pdf [Broken]
    Last edited by a moderator: May 3, 2017
  7. Oct 30, 2008 #6


    User Avatar
    Science Advisor

    Not sure I understand this correctly; do you mean that two lanes would be required, with one robot between them, for servicing vehicles with the refueling cap on either side (the same way a single pump services two lanes in a current fueling station)? I can see that the ceiling mounted design would permit vehicles of either configuration to use the same lane.

    BTW; have you thought of how to try and resolve the problem of vehicles that have their fueling cap in the middle (under the license plate)? That could be a serious headache.
  8. Oct 30, 2008 #7


    User Avatar
    Gold Member

    Good point about the centre-mounted filler, Lurch. The Roadrunner has one of those, with a spring that I'm pretty sure came from a bear trap. It's hard enough to gas that sucker up manually, without having a robot try to figure it out.
  9. Nov 4, 2008 #8
    That's exactly what I mean. So there's an XY linear base on the ceiling, an arm coming down (Z) and then a double-rotary nozzle.

    About the centrally positioned fuel caps: if the car can be positioned far enough to the front the robot should manage to service these kind of caps as well (the robotic arm should be able to lower itself far enough). I think I might be overlooking something, because I can't see any immediate problem with that.

    LURCH, because you mentioned air cylinders I got digging into pneumatic component catalogues and figured that moving fuel hoses in close proximity of relatively high-power, high-torque electrical motors and electrical cables might be a bad idea.
    So I got myself asking: can I make this robot (XY-base, Z-arm, double rotation nozzle) fully pneumatically actuated knowing the robot has to move about 3 m in XYZ, without using electrical motors? Are there pneumatical rotary drives with precise angle control available? (I don't seem to find such parts in the Festo product catalogue)

    Another reason for me to fully switch to pneumatic power is that I could install a multitude of fuelling cells, connected to one central pneumatic compressor.
  10. Nov 4, 2008 #9


    User Avatar
    Science Advisor

    Yeah, that is quite a long reach. I suspect that if such cylinders exist, they would have to be telescoping.

    I'll do a search for "telescoping pneumatic cylinders," bit it could be a while (couple of days) before I get back in here.
  11. Nov 4, 2008 #10
    The arm doesn't need to come down all the way 3 metres or something. Most fuel caps are at about 1 metre above ground, so I would only need an (estimated) displacement of +/- 0.5 metres. I guess this is feasible with a linear pneumatic axis.

    Can someone please confirm the existence of pneumatic rotary drives which allow for precise angular swivels? I need to get my specs laid out and would like to know if precise angular motion with pneumatics is possible (I haven't found any suitable components yet).
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook