I am working on an arm similar to the arm used by team 44 for their elevation robot, which consisted of what seems to be two sets of parallel bars (I believe this is commonly known as a four-bar) supporting their cube intake, which can rotate more than 180 degrees from its starting position when the bars are pointing downward, and i was wondering if there is any issue with using this type of parallel bar setup, due to weight on the back two bars. I am finding through prototyping with this arm type that the rear two bars, when at the same height as the front two bars, as in when they reach out the most and highest amount of force is applied to the arms that they will fall a small amount which will cause the arm to require a large amount of force to move any higher and also at this point prevent the bars from being parallel. If anybody has any thoughts, has had the issue or has a solution to this issue it would be very appreciated.
I’m having trouble following your rambling description.
It would help if you can post a picture of your prototype.
Even better is a cartoon drawing with the bars labeled.
Four bars with small transmission angles can be a problem.
What are you using for pivots?
If using Vex screws in Vex metal holes, the tolerances are loose.
Try using Vex (or mcmaster or whatever) shoulder bolts, which will have tighter tolerances. Vex Hinges work well too.
Well what I’m trying to make is a four bar lift which rotates about 250 degrees, with a starting position of 0 degrees, what happens is that at about 90 degrees, the back bars of the lift fall below the front two bars tilting my manipulator back about 20 degrees. I apologize for my vague description, team 44’s elevation robot is a great reference for what i am doing so i could put a link to a picture of their robot to display what i am trying to accomplish. The link shows the exact point in which i am having an issue, imagine that intake/grabber for the cubes being offset by about 20 degrees as if someone is pushing down on the back of it. my pivots are vex shafts in bearing flats.
I know that for this style of arm, unlike a conventional four bar, both arms should be powered. Otherwise, the linkage will have a tendancy to want to “cross,” once it reaches this point.
yeah we have noticed this while working with it, and i guess that’s where the issue comes from, because it wants to “cross” at that point and i need a way to prevent that from happening, and keep the motion smooth… any suggestions?
He just gave you a solution. He’s saying if you power both arms (by gearing them together or otherwise), they will not cross.
well all of them already were powered, I’m using four motors, one on each bar, and at that point where they want to cross there is still a lot more friction than usual and it causes the manipulator, or cage or whatever you want to call it to not be parallel with the floor, it creates about a 15-20 degree angle. That is my issue, I am trying to be rid of this small angle and keep it parallel like that because the arm has difficulty lifting itself at this spot, because of all the friction caused by this
Try linking the front/back motors together with a 1:1 run of chain; this should mechanically lock the motion of the links together. With separate motors, a whole bunch of factors could be coming into play. For example, the weight of your manipulator is probably not evenly distributed between the front and back links, forcing one pair of motors to do more work than the other. Or, you may have just stumbled upon a motor or 2 with a touch more speed or torque than the others. A mechanical connection that forces the links to rotate at the same rate will eliminate many of these potential trouble spots.
If that doesn’t work, post pictures. I think it would help us understand the problem better.
Beat me to it, if you lock the bars this problem will probably fix itself. Power off a central gearbox and run chain or gears to the linkages, or power one ( front or back) and then run chain to the other bar at 1:1. Repetitious I know but just throwing out ideas.
-Andrew
thanks for the suggestions guys, and we did find that this helps the issue a lot, but the problem is very small at this point, so what i am going to try is using either latex tubing or rubber bands to creating a lifting force for the rear bars, which will be attached to the rest of the system. Maybe the extra force will lift the rear bars just enough, since chain gives a little bit of slack on the two sprockets.
Team 3018 used a similar design and posted this video: http://www.youtube.com/watch?v=XQkIg_6oslU
But it has the bars aligned vertically rather than horizontally.
yeah now what i have found is using something to counter gravity (i.e. a counter-force) helps to remove the issue… now to add a little bit more torque 
a bit more? you doubled the ratio lol
Just like a 4 bar linkage, if the gap between the back and front bars are different at the top from the bottom, it makes the manipulator rotate from parallel a certain amount of degrees when it lifts.
(At AURA, we call this a “swing lift”, like team 44’s elevation robot)
(ignore the dots)
Back . . Front
bar . . . bar
| . . . . .\ <-- top, smaller gap between bars (powered from here)
| . . . . .
| . . . . . .
| . . . . . .
| . . . . . . .\ <-- bottom, bigger gap between bars (where manipulator is)
Note you can use this tilting feature in your favor, if you want to tilt the manipulator forwards as you lift instead of backwards.
If not, make sure the gap between the top is exactly the same as the bottom by counting the holes.
The gaps were the same, but i am definitely going to take advantage of this idea as it will help us with what we are doing, thank you very much for the insight