We have an encoder integrated into the climber. We plan on monitoring the current to detect when the climb starts, and set a stop distance based on that start point. The operator will have the ability to jog higher if we stop too soon, but hoping this approach leads to automatic climbing.
Still untested theory at this point. Fingers crossed. Would like to have working repeatable solution by week 1, we are going to need it.
I think the easiest solution is to simply use the motor controllers to sense when the motor is stall. This is what 6101 is doing. The Davit has a C-channel piece to protect it in brief high-stress situation and the 40A breakers wouldn’t pop immediately. I don’t see any reason why this wouldn’t work.
To prevent gliding, we are using a ratchet head on the winch similar to Ri3D 1.0.
This shouldnt be too hard.
Once you see the light go on, dont drive the lift anymore.
We plan to use a simple ratchet to prevent it from backdriving.
Will you have the ability to disengage the ratchet under load?
If not, what keeps you from effectively clamping onto the field when waiting the 1 second?
Sorry, I’m slow. I dont quite follow?
The light does not go on until a full second after the plate is pushed.
Hope your knot holds.
Yes, that part we knew and it will hold.
I’m missing something here - isn’t this something everyone has to worry about?
Yes, but if you wait to stop your winch until AFTER the light comes on you will be putting the full power of your winch + the weight of your robot pulling on the rope for that second.
That’s probably a decent amount of force (at least 300 lbs)
Hence the need for sensors.
When you hit the wall on top, your motor will go to stall (unless you’re current limiting) and push very hard on the top.
Now imagine everything in the system is a spring, you’re going to compress/stretch parts of the robot/field and load everything up.
Now there is a ratchet on there preventing you from backdriving (and you can’t go forward any more as you’re already loaded pretty firm into the field).
Now that’s how you lock yourself onto the field.
In 2010, we created a simple ratchet pawl system that worked for our climber. It should release if we at least carry the robot, releasing weight tension.
You won’t be able to release weight tension as you’re literally clamping the field. Can’t go the other way because of the ratchet as well.
I’m expecting at least one case where a robot yanks its stopper knot out of the davit this year, especially with the crazy ratios that people are putting on their climbers.
Well, I guess we shall see.
The current measuring method works very well. We’ve used it in other instances in the past and plan on using it this year for the climber mechanism.
Interesting point. How is 973 dealing with this?
If you have a ratchet and are planning on stalling against the top for the full second, it needs to be able to disengage under load OR you need some way to remove the load.
Prepare to be disappointed. There’s nothing special about the scenario Adam has outlined. Any system with a ratchet and pawl is going to behave exactly as he’s explained. We had this happen multiple times in 2010 with our kicker and our hanger. It’s a HUGE PITA when it happens. We had to use a giant flat head screwdriver and apply major force to the pawl to lever it to disengage. This is assuming you have a pawl that’s actually accessible. If you’re using a ratcheting wrench you’re totally screwed.
Imagine trying to release a ratchet strap that has been ratcheted to significant tension. If you try to pull the release with your fingers without taking up some of the load with the handle, you will never get it undone. You have to release tension off the release bar in order to move it. This is exactly what will happen in the scenario outlined above.
It is legal to remove the rope from the field without removing the robot from the rope. So couldn’t you have two team members support the robot, while another team member releases the rope from the davit?