# Drivetrain Speeds

You don’t. Not in 2019, anyway. If you take a quick look at some of Zebra’s Dart data, this year no teams are hitting those speeds in matches. Most cap out at 12-13, some at 14-15. 95th percentile speeds are 10-11 ft/s. This is one such example thread: 2019 Chezy Champs Zebra's Dart Data Analysis.

There’s got to be a way to have fast acceleration. The 18ft/s is just a random high number

There are two ways to get a high acceleration (both of which are talked about in this thread already, iirc).

1. More motors – means more force, means more acceleration
2. More mechanical advantage ion the motors you have – this means a lower top speed

Short sprints like this year mean teams want high acceleration, resulting in lower top speeds than other years with more open fields (2017 and somewhat 2018).

You can also add shifting in to give you some more freedom in your top speed / acceleration trade-off. I’d recommend looking at what gear ratios top-teams use each year, along with the field and game task to get a feel for what’s right. 254 puts their gear ratios in their technical binder each year. Lots of CD threads (like this one) have good info as well. Play around with JVN’s drivetrain calculator and/or the various sprint time calculators.

1 Like

Traction is an intrinsic limiter on acceleration as well. You can only apply so much torque at the wheel before it slips. You’re going to be constrained by the fact that the robot weights 120lbs + battery + bumpers, and that you’re typically running on carpet.

The size of the field is also a constraint, you only have so much room to accelerate.

This is why we try to gear for a sprint time, rather than a top speed. We identify the most commonly traversed distance based on strategy, and gear to minimize the time to drive that distance.

1 Like

Do you use a particular design calculator for that?

2 Likes

If you make a robot that can pull 1 g of acceleration and still have a somewhat decent top speed I’d love to see it.

Not trying to be rude but I just haven’t been able to figure out a way without having a muti shift gearbox or a ton of motors.

1. Reduce the mass/weight being accelerated.
F = M * A
3 Likes

So if I make my bot out of aerogel and put 2 four NEO gearboxes on it I could hit that 20 FPS mark?

You are confusing top speed and acceleration.

There are factors such as the available traction that may limit the maximum acceleration that can be achieved.

How rigid is aerogel? The performance might not be very good if the wheels don’t all point in the same direction.

2 Likes

If I made my bot out of aerogel it also would cost \$32,000 so I’m not going to do that.

1 Like

Maybe you joke, but here is a demonstration of possible vs practical. It is possible to hit 20ft/s actual speed on a FRC field with a full-weight bot, but I’d argue that it isn’t practical without a game like 2008.

8-NEOs limited to 40A each with a 4.3:1 gear ratio on a 120-lb bot will net a very reliable 20ft/s in about 19 feet of acceleration. Unfortunately, the robot has to start braking about 0.5 seconds after hitting top speed in order to prevent robot splatter on the opposite side of the field.

Of course, one might say “well just throw the sticks in reverse, and we can stop any time we want!”. Yes, sort of. The pit crew may have some consequences to deal with, since the robot will skid for about 9 feet if it doesn’t topple over first.

An 80-lb bot does not fare much better (all else equal to the scenario above). While it is able to maintain full speed for almost a full second, it still takes about 6 feet to stop when thrown in reverse.

4 Likes

What is going on under the hood in all these lovely acceleration simulations?

Assuming `1g = 9.8m/s^2 = 32.15ft/s^2`, this can be done with an 80-lb bot and 4-NEOs @ 60A (or 120lbs with 6 NEOs @ 40A) and a 6:1 ratio on 4" wheels. The hard limit here is the traction. This graph was created with a 1.1 coefficient of friction; 1.0 or less dipped below 1g of acceleration. At 1.1CF, even as I switch the gearing around the acceleration itself remained the same - only the top speed changed.

Interestingly, with 4 NEOs the current limit must be raised to 45A on a 80-lb bot or 60A on a 120-lb bot before the current draw is no longer the limiting factor to getting to 1g of acceleration. So it is possible with just 4 NEOs, but probably not pragmatic due to several live-match factors. For example, the individual 40A breakers could pop if the bot got into the slightest of pushing matches using this configuration.

Cheers!

3 Likes

Many thanks.

Three thoughts continuing this experiment:

Though steady-state the current limit will be about 40A, for short bursts you can go well above that. Probably not super helpful as you already showed this isn’t the limiting factor.

At a certain point, even with enough traction, your center of mass location becomes a limiting factor. With enough acceleration you start to pop wheelies and eventually flip.

Finally, if you design things solid enough and don’t mind scaring the daylights out of the opposing alliance, the robot doesn’t have to hit zero velocity prior to traveling the full length of the field.

4 Likes

We were close this year, accelerating at 9 m/s^2 with a top speed of 3.5 m/s, on a single speed gearbox, with a 125lbs robot

How did you measure those numbers?

At first using the encoders, then comparing the theoretical times for a 20 ft dash to the actual ones and seeing that those match, to account for slipping.
4 775’s per side seem to do the trick

1 Like

Because of our teams budget last year we used the KOP drive again. We made the Ontario provincial championships. The drive-train speed is around 10-11 ft/s with cims and standard gearing. We found this speed was not ideal because it could not push other robots. If a defense robot blocked one tower we would drive to the other tower and often their drive-train speed was the same for faster and we could not reach the other tower before them. A drive-train at 12-14 ft/s will do much better. Too fast it may become hard to control, the limit I remember seeing is around 20 ft/s. Shifting comes down to budget and often better mechanisms should be priority. Also you may want to consider getting brush-less motors for your drive-train instead of shifting.

1 Like