Erratic Chain Lift problems!

VEX V5 General Discussion General Forum
1

So, I researched about building a chain lift using the 1103 idea and this animation by Jpearman: http://jpearman.smugmug.com/Robotics/Misc/17225808_v2Hjfz#1322058158_xq42dQ7-A-LB

I have built this chain lift, but with less spacing between the 2 stages of linear slides. Also, the driven sprocket is 12 teeth and the other 4 are 6 tooth sprockets. Another difference (this could be the reason for my problem) is that when the motor turns to lift the arm, the middle stage goes up first (with the bottom center sprocket not turning). Next, the final linear stage goes up with the same sprocket turning in place. Now, the actual problem is that when the final stage is going up or going down, it moves in an erratic/jerking motion. I think that is somewhat related to when the sprocket turns in place but I want to hear from you guys. I would appreciate any help, Thanks.

2

First let me say that the animation was just to demonstrate the principle and is not a good example of how to actually implement this idea. Also, what the animation does not show is that the order in which the two stages rise is not fixed and depends on weight distribution and friction.

It sounds like something is binding on your final stage or it is far heavier than the center stage (which is quite likely), does everything move smoothly with the motors and/or chain disconnected? Perhaps post some pics for us to look at.

3

On the final stage, we have our intake attached so it is definitely weighing it down. I have read that the part with the least amount of friction will move first and I guess that is what is happening here. So is there any way to put more friction on the middle stage? I say this because I can’t really make the intake lighter. Maybe latex tubing for tension would work? I will try to post pics later as I do not have the robot with me at the moment, but it is almost exactly like your animation except for the two differences I posted above.

On another note, I would really like to thank you for your animation as it really helped me a lot in understanding the chain lift idea!

4

I wouldn’t recommend adding friction to any section. Generally with arms and lifting mechanisms, adding friction on the upstroke makes it slower and less effective. Most uses for latex elasticity help counterbalance the weight, in essence lessening the friction. Maybe you could use tubing on the segment with more friction to try to counteract it?

5

Ok, I do have some Latex Tubing, so how would I go about attaching it so it counterbalances the weight? Do I just attach it to the intake and the final linear stage?

Also do you guys think you could explain how the stage with the least friction moves first? I’m not really fully understanding this part and it might help in attaching the latex tubing in the right places. Thanks

6

I’ve worked a bit with continuous chain lift systems. Might I suggest you try putting a block/protrusion on the bottom of the second stage extending towards the third stage such that the third stage must remain above that second stage block? It’s similar to how Jpearman’s animation depicts the first and second stages, where the first stage [stationary] has an extended element under the second stage so that the second stage won’t go under the robot. Just extend that geometry to the second and third stages, so that the third stage will never be under the second stage. A standoff should work fine. You will lose that increment (one hole for the standoff) of maximum vertical length in the 18 inch size limit on your third stage however.

7

By third stage do you mean the intake connected to the second stage of linear slides? And the standoff that prevents the intake/third stage from going below the second stage; Does it remove the extra friction/weight that the intake causes on the second stage by putting it on the standoff instead? If so, would latex tubing work in the same way? I think I see what you are saying but please correct me if I misunderstood.

8

Attached is a diagram that should explain the fix. Highlighted in yellow are the suggested downward stops in the two stage linear lift. It is a physical stop, a very simple attachment for multiple stage linear lifts. (No tricks with friction/latex tubing, which are over-thinking the described problem, and will probably not work as well.)

I agree with mediumdave that adding friction/opposing force should be avoided as it is a waste of your lift power. I also doubt the ease of figuring out the specific elastic tension equal to the force gravity of your feed system.

Comment: wow image quality is frightening. reason to stop using jpg?
Chain Lift Possible Solution.jpg

9

Ok, I looked at the picture and I do have a C-Channel attached to the 1st stage (the first stage in your picture, not the first stage). I do not have the smaller yellow section as in your picture, but I will attach it. Would the big yellow part be attached to the 2nd stage and would the small yellow part be attached to the 3rd stage? Or would it be just be spaced out like in the picture? The X’s lead me to think they would be attached but i just want to confirm it.

10

The yellow highlighted portions are attached directly to the stages they are on. We seem to be using different definitions of “1st, 2nd, 3rd Stage”. I think I understand your confusion, related to how my preferential naming conventions are not intuitive logical. I will translate them over. From now on ignore the stage numbering of the diagram and previous statements.

The bottom section, ZERO stage, is non-moving. It is attached to the robot. The first yellow protrusion is the base of the lift and robot, attached to this bottom section. It keeps the upper stages from ever running down into the ground etc.

The next section (going up and to the left) might be called the FIRST moving stage. It also has a yellow protrusion, which is only connected to this stage. This protrusion keeps the SECOND moving stage, the next section up and to the left in the diagram from being lower than the FIRST moving stage, (i.e, a lower stage going up before the top feed-attached stage). I believe that is the problem you want to solve. The SECOND stage is at the top, which may be attached to your feed system.

Described is how blocks work for a 2 stage linear lift. For a 3 or more stage linear lift, just repeat the instructions until at your top feed-attached stage.

The green arrow and x illustrate the fact that you cannot force a section/stage to go lower than the preceding (lower level) section/stage, what you seem to desire. The green arrow without the x above each moving section/stage illustrate how the sections/stages can move up without impediment.

11

Ok, I understand what you’re saying now. This should solve the stages moving in the wrong order I think. I will implement this change and post back. Thanks for your help Edward!

12

You’re welcome! Sorry for confusing/illogical explanations at first. I think I write too much…

13

Post back when it’s running. I know everyone if focused on the single chain triple lift but there are other ways of achieving the same result, perhaps not as compact but there can be other advantages. Here’s an early preview of one of our teams ideas, it uses multiple chains.

https://vexforum.com/attachment.php?attachmentid=4459&stc=1&d=1315847617

slide_3.jpg
slide_3.jpg400×562 22.5 KB

14

K, so I’ve done some testing/troubleshooting with the lift and I learned that when my intake is not attached to the last stage, there is less jittering when going up or down on that stage. I brought up the last stage so it’s higher than the middle stage and I am still getting the same problem as before (jittering). I think that the cause of this problem is the fact that the center sprocket (center out of the bottom 3) is turning in place when it should not be and is not turning when it should be. Meaning that the order that the slides go up is backwards, which in turn would mean that the weight on the last slide is too heavy. As I said, I already brought the last stage above the middle stage so that’s not really the problem, I think, but I’m not sure. Any suggestions as to how to fix the sprocket problem and/or jittering? Thanks.

15

I would take the motors off and run it by hand to find what exactly is getting stuck or is making it hard to turn.
It may be a problem of the linear slides binding, seeing as how it jitters more when there is more weight on it.
All of the sprockets should be free to turn, and the only simple way to make sure the right ones are spinning at the right time would be to make physical stops like has already been described.

16

Yeah, I’m gonna need to put some elastics on the intake on the last stage because that seems to be the only way that i can reduce the weight of the intake. The support at the bottom isn’t enough. Any ideas on how to use the elastics so they hold up the intake both going up and down? Im not really familiar with Elastics, having never used them before. However, I’ve seen them used a lot in 4-bar linkages, and I’m wondering if I could use the same sort of principle in the chain lift.

17

Elastics on vertical lifts is hard because the elastics go from being stretched all the way to being not stretched at all. With a 4-bar, you can add elastics such that the length only changes from 12 inches to 8 inches (for example), so the force stays fairly constant. On the vertical lift, having the elastics help for the bottom foot or so is the easy part, it’s the next 1.5 feet that is hard…
I don’t actually have a solution for you (or my team for that matter…)

18

So, I preface this with an admission that I have not used elastics on vertical lifts before. However, I have spent a heck of a lot of time with vertical lifts (continuous chain fed two tier) and on the vex surgical elastics individually.

If the problem is that you need more even force from your elastics over a certain stretch range, you could always try to make your elastic either stretched even at the “loose” distance, or try looping the elastic around a free spinning/low friction pulley roller, such that you get more total distance when fully stretched.

F[spring tension] = (-)(k)(x)
F[spring tension] = (reverse to stretch direction)(spring coefficient)(distance of stretch)

note: the coefficient is not important for our purposes unless you are going to calculate everything precisely. i suggest guess/check/by seat of pants on how much tension is appropriate

So, for example, your problems seems to be this: let’s say you stretch from the bottom of a higher tier to the top of a lower tier. that is around 18 inches. then you have a certain amount of upward tension of that distance times a coefficient. but when you move that tier all the way up, the distance of stretch becomes 0 and you have no tension.

Solution possible: what if you change your elastic layout to go from the bottom of a higher tier to bend around a low friction pivot point (i.e. pulley/rollers that will equalize tension on either side) at the top of a lower tier, and then stretch down to the bottom of that same tier. this way, when fully stretched (where all tiers are stowed down) you have the elastic stretch 18 in up and 18 in down for 36 inches of stretch. when your arm is fully up, it will have 18 inches of stretch. this is still a 50 percent loss of tension, but it definitely beats zero at the top.

As to mechanical implementation of this idea (making sure no tangling, losing alignment with the roller at the top etc.), i’m as lost “as the next guy”. Especially for the tight dimensions/very little free space in continuous multiple tier linear lifts, this sounds daunting. I remember having an idea similar during the clean sweep year, but the complexity of thinking how this would be implemented reliably versus the minimal effect/benefit of such a tension system on a pivot-style arm (mentioned by dontworryaboutit) made me quit that endeavor.

If you get it working, props to you and please post pics. Perhaps some who have had well tensioned linear lifts already have figured it out. I suggest you ask Titan (1103) about this.

19

What if you ran the elastics from the top of the upper-most tier down to the bottom of that tier, to the top of the middle tier down to the bottom of that middle tier, to the top of the lower-most tier and back down to the bottom (mounted on the base of your robot)? That way you have only a 40% loss of tension, and it would keep the elastics constant throughout the lift.

As for mounting the elastics, I would recommend using pulleys on each end of the axles that hold your sprockets, just because those axles are already there.

20

haha yes. just do that on both sides to maximize the ratio of stowed elastic stretch length to full height stretch length. cleverrr.