Why does the front roller in a "chinatake" run faster?

So, I’ve seen many people make ‘chinatakes’ : intakes that have a flap roller in the front and a non-flap roller on the back, with the front roller running faster than the back roller.

This is counter-intuitive to me.
I’m inclined to think that acceleration, which is inversely related to RPM, should be more important to the front roller than the back roller.
If the front roller were to run slower than the back roller, we would be able to quickly accelerate the game object at the front intake, and then, further ramp up the speed at the back roller.

So, why is it that, in a chinatake, the front roller is faster than the back roller?
Am I missing something here?

One possible reason is because the back one needs more torque to pull the ball through the wall goals that have a smaller space than the balls. Which the goals needs to fix, thus needing more torque to pull it through while in the front those rollers will not need to do that making it so you can speed up the front ones.


I’m pretty sure that the front rollers would also need that torque, as it’s the one that first establishes contact with the ball.
And also, I saw the ‘chinatake’ first appear during TT, and back then, there wasn’t a wall goal that we had to pull through.

Having a faster front roller back then in TT would have allowed for them to have a reasonably fast fielding speed, while still being able to push through the weight of the blocks on the tray.
With the existence of uptakes in most snailbot designs, this aspect of having a slower back roller is probably negated.

Following this line of logic, it seems reasonable to derive that having a chinatake-style intake in Change Up wouldn’t really have a significant advantage, and if anything, the advantage on acceleration that comes from mechanical advantage should make it so that ‘reverse-chinake’ intakes be more popular.

Why, then, are people still building chinatakes? Where am I getting this wrong?

Do you mean not using two sprockets connected by flaps and rubber chain?

1 Like



Sorry just found an image of my old intakes to see if this is what you mean. And yes the lift was up when I took the picture.

1 Like

Intake from here, we can see how the OP mentions how the front roller runs faster.

We can also see how the higher torque in the back roller helps with fielding balls in the wall goals, and now I think I get what you meant by pulling the ball through the wall goals.

I guess I should have read the field spec more carefully. Now I see why chinatakes are still viable.

1 Like

this year you want the most torque in the front rollers, the ones touching the ball while it’s still in the goal

1 Like

tbh tho having the same torque throughout the rollers is fine too; if you’ve got a ramp to lead up to your uptakes you’re gonna need some torque to get the ball up that too

1 Like

all about finding a balance between speed torque and efficiency


I don’t agree with you here. Think about what you just said. Acceleration is a change in velocity in this case angular velocity. So the greater the acceleration of the intakes, the greater the angular velocity –– in this case RPM.

Which acceleration were you referring to? The acceleration of the ball into the robot is also proportionally related to the RPM of the intakes because the tangential speed of the intakes is also proportional to the RPM and the force acting on the ball is tangential to the intakes.

I mean this should all be intuitive –– the faster your intakes are, the faster the raw intaking cycle time is (so long as you overcome the torque threshold).


I was referring to acceleration of the front roller.
Now that I think of it, though, this acceleration wouldn’t really matter, as long as the torque threshold is overcame.
I’m guessing that past me got confused between acceleration of the roller and acceleration of the balls.

Anyways, I agree with you here.

torque is angular acceleration * inertia, so…

1 Like

i never had any formal education with this stuff, leave me alone

In all seriousness, bruh

So somewhat related to the discussion of torque threshold, what do you guys think the most viable and efficient speed would be this year? I am thinking since last year the most commonly used (and fast) speed was approx. 200 rpm (18t sprockets) this year it might be around 300-400 rpm due to the lower stress on the intakes. That might only be possible on the front roller though because of having to pull out the ball.


I’d say a bit faster than that since we won’t have the weight of the stacked cubes going against the intakes as we did in Tower Takeover.
Most reveals I found used 600RPM if my memory serves me right.
Of course, RPM doesn’t necessarily mean anything of value unless we also know the radius of the spinny thing.

This is coming from a guy that didn’t even make a robot yet, so take this with a grain of salt.


depends on what sort of competitions you’re planning on attending imo.

If you’re doing in person events, I would say stick to around 300 rpm or lower to be safe. you don’t have control over the goals, and tolerances might make some goals require a bit of extra torque. Better to have slightly slower but consistant descoring then fast but unreliable.

If you’re competing in virtual skills however, you can probably up the speed to 300-600 rpm. Since it’s your field, you can make sure all your goals are spaced perfectly so that it doesn’t require more effort than normal to descore.

And if you’re doing remote events, there’s no point at all in having side rollers.


Why is this?
20 characters

you don’t need to descore.

1 Like