One speed vs Two speed gearboxes

[asid61]

Originally Posted by z_beeblebrox

Can I visit the magical world where this is the case?

I feel like what you've been saying about extremely high gears is true if and only if power management is not a concern.

It's called a shifting gearbox.
My point is that with a shifter you don't need to worry about power managment, as long as you are either a) careful and shift or b) use encoders or other methods to automatically shift down in collisions. I prefer the latter method, as the first has much less use.

[AdamHeard]

Originally Posted by Michael Corsetto

If I understand correctly, asid61's stated hypothesis is this: limiting speed reduces competitiveness.

In the context of Aerial Assist, I agree with this. I believe our speed on the field was a component of our success this season.

All aboard the California powerhouse hype train. Next stop, Einstein 2015

-Mike

Chooo Chooo!

I think it's fair to argue for 95+% of teams the difference between a 16fps high gear and a 20+ fps high gear is not what will make them more competitive.

Gearing that high requires a higher attention to detail mechanically, as well as a good supply of batteries. Many teams don't realise their batteries are in bad shape because their robots simply don't use that much juice.

[Caleb Sykes]

Originally Posted by asid61

So you slap on a shifter.

I don't know of any teams that can just "slap on a shifter" without any substantial time commitment from their build team. Additionally, shifting necessitates both more coding time and more practice time for the drivers. Shifting is a tradeoff, just like every other part of the robot. No robot will instantly just become better if a shifter is just slapped onto it.

Maybe I am reading too much into your words asid, but it seems to me that you believe that every robot would instantly just be better if a shifter were used instead of a single-speed. On a team with infinite resources, this might be the case, but on my team at least, we always have to make tradeoffs in our designs, and we will likely not be using a shifting gearbox next year so that we can focus on other aspects of our robot.

[AdamHeard]

Originally Posted by inkling16

I don't know of any teams that can just "slap on a shifter" without any substantial time commitment from their build team. Additionally, shifting necessitates both more coding time and more practice time for the drivers. Shifting is a tradeoff, just like every other part of the robot. No robot will instantly just become better if a shifter is just slapped onto it.

Maybe I am reading too much into your words asid, but it seems to me that you believe that every robot would instantly just be better if a shifter were used instead of a single-speed. On a team with infinite resources, this might be the case, but on my team at least, we always have to make tradeoffs in our designs, and we will likely not be using a shifting gearbox next year so that we can focus on other aspects of our robot.

Arguably you can slap one of these on just as easily as slapping on a single speed.

http://www.vexrobotics.com/vexpro/ge...llshifter.html

[asid61]

Originally Posted by AdamHeard

Arguably you can slap one of these on just as easily as slapping on a single speed.

http://www.vexrobotics.com/vexpro/ge...llshifter.html

As a matter of fact, it's can be kind of hard to do that on a ballshifter. We are planning out a test WCD (again) this year, and we found it hard to integrate a ballshifter without having to make our own bearing blocks. We could add our own bearing blocks, but we wanted to try as much COTS as possible, so we're actually going with a WCP shifter for the chassis. But depending on the chassis, WCP or Vex is easy.

Originally Posted by inkling16

I don't know of any teams that can just "slap on a shifter" without any substantial time commitment from their build team. Additionally, shifting necessitates both more coding time and more practice time for the drivers. Shifting is a tradeoff, just like every other part of the robot. No robot will instantly just become better if a shifter is just slapped onto it.

Maybe I am reading too much into your words asid, but it seems to me that you believe that every robot would instantly just be better if a shifter were used instead of a single-speed. On a team with infinite resources, this might be the case, but on my team at least, we always have to make tradeoffs in our designs, and we will likely not be using a shifting gearbox next year so that we can focus on other aspects of our robot.

Shifting gearboxes do cost a lot of cash, as far as resources go. More coding, okay, more practice, okay. However, both of those can be done pre-season as long as you release the code for the shifting as open-source. Other things might be a higher priority, but if you have the money and pre-season time it definitely worth it.

[Arpan]

Originally Posted by Abhishek R

I agree, I think speed was one of the factors that separated a good team from a great team. 1678 is a perfect example of this.

As a counterpoint, though, for most teams - who don't build a practice bot and therefore don't have the time to train a driver for super-high speed manuvering or develop robust autoshifting code - super high ( I view that to be above the 16-17 fps range) speeds can be absolutely terrible.

Case in point: Us last year. While our drive was a monster and for the most part dominated the field, our driver was, for our first competition, terrified to drive in high gear because of the high risk of blowing the breaker. It took him the full competition to really get the hang of obstacle avoidance and shifting for plowing.

[AdamHeard]

Originally Posted by asid61

As a matter of fact, it's can be kind of hard to do that on a ballshifter. We are planning out a test WCD (again) this year, and we found it hard to integrate a ballshifter without having to make our own bearing blocks. We could add our own bearing blocks, but we wanted to try as much COTS as possible, so we're actually going with a WCP shifter for the chassis. But depending on the chassis, WCP or Vex is easy.

The 3 cim ballshifter is designed to mount with half of a versablock set.

[asid61]

Originally Posted by AdamHeard

The 3 cim ballshifter is designed to mount with half of a versablock set.

... <facepalm>

Well, we already ordered everything, and it's not like we're any less off for that. WCP is still fine.
We thought we were supposed to use the "gearbox bearing block" like on a WCP box.

[R.C.]

Originally Posted by asid61

... <facepalm>

Well, we already ordered everything, and it's not like we're any less off for that. WCP is still fine.
We thought we were supposed to use the "gearbox bearing block" like on a WCP box.

You can use the versablock on both actually.The "Gearbox Bearing Block" is just for teams that can CNC a hole/want a smaller bearing block.

http://www.buildblitz.com/final-cad-files/

The Copioli robot should show the setup with Ball Shifters + VersaBlock.

[Michael Corsetto]

Originally Posted by AdamHeard

Chooo Chooo!

I think it's fair to argue for 95+% of teams the difference between a 16fps high gear and a 20+ fps high gear is not what will make them more competitive.

Gearing that high requires a higher attention to detail mechanically, as well as a good supply of batteries. Many teams don't realise their batteries are in bad shape because their robots simply don't use that much juice.

Absolutely agree.

-Mike

[Travis Schuh]

Originally Posted by Michael Corsetto

If I understand correctly, asid61's stated hypothesis is this: limiting speed reduces competitiveness.

In the context of Aerial Assist, I agree with this. I believe our speed on the field was a component of our success this season.

All aboard the California powerhouse hype train. Next stop, Einstein 2015

-Mike

Originally Posted by AdamHeard

Chooo Chooo!

I think it's fair to argue for 95+% of teams the difference between a 16fps high gear and a 20+ fps high gear is not what will make them more competitive.

Gearing that high requires a higher attention to detail mechanically, as well as a good supply of batteries. Many teams don't realise their batteries are in bad shape because their robots simply don't use that much juice.

I would challenge the notion that a smaller gear ratio will always result in a quicker robot on the field. We did some modeling a couple years ago and found that because of the tradeoff of top speed and acceleration, 16 to 16.5fps free speed with two CIMS and full weight robot is fastest for a half field distance. Our thinking is that for most games, it is rare that you can have an uninterrupted run across the length of the field, and that having top speed here is not worth trading off performance during shorter runs.

We ran this speed last year and were keeping up gamewise with those geared faster. This could be because 2014 game play ended up being much further from an open field than it looked like, but I think another reason to not lose sight of is that once your robot is fast enough, it is more important how it is driven than what the exact speed. Our driver gets a lot of practice, and a fundamental of what we practice is how to accomplish whatever needed in the most efficient means possible.

[themccannman]

Originally Posted by AdamHeard

Chooo Chooo!

[Thad House]

This year, we had 6 Cims on our drive. During competition season, we were geared for 10.5 fps. With the wheels we had, this was traction limited, and we never got even close to popping breakers. But once we got to the highest level of competition at worlds, we were too slow to be competetive. We switched to 16FPS at IRI. The speeds with this actually allowed us to be competitive with the rest of the field. But in eliminations when our driver was playing rougher defense, we did pop breakers.

I would argue that with 6 Cims, the optimal points are about 10FPS for the low speed, and 16+fps for the high speed. With a 6 Cim drive, 10FPS will allow you to get around your half of the field basically as fast as possible, yet still allow you to not get in trouble with defense. Then if you need to cross the field, or the defender is faster then you, you can shift up to 16+fps and get across field very quickly.

If the field is split like 2010 or 2012, I would argue that a single speed 6 Cim 10 fps robot is the best you can get, because there is not enough room to use any higher of speed. For years like 2011, 2013 and 2014, you want to be able to shift up for extra speed to cross the field quickly. Those are the types of years shifter are useful, because they allow cross field travel faster.

[Abhishek R]

Originally Posted by asid61

Yes, they were good. But could they have been better? That's not said as a challenge, but I am legitimately interested in why they go with a single speed over a two speed.
As a counterexample, 254 has been shifting for about a decade now, and they have two championship wins.
I would like somebody from 118's opinon on this.

I don't understand this, there are numerous factors as to whether a team wins or not, and I doubt shifting is a major one.

I would hesitate to use wins/losses as evidence to back up certain points, mainly because correlation is not causation.

As a counterexample to your counterexample, 610 in 2013 was extremely competitive to the point that they took home the Championship. They also only used one speed - I recommend looking up their philosophy on that, I believe Mr. Lim and Mr. Rob Stehlik have some in-depth posts about it. Back to my point, personally while I thought they had a very good robot, I would not attribute the win to them having the single best robot on the field, but rather to 1241, 1477, and 610's capability to play as an alliance instead of playing as three teams.

I still believe one choice is not inherently better than another, both have benefits and drawbacks, and it just depends on the game or team preference.

[asid61]

Originally Posted by Thad House

This year, we had 6 Cims on our drive. During competition season, we were geared for 10.5 fps. With the wheels we had, this was traction limited, and we never got even close to popping breakers. But once we got to the highest level of competition at worlds, we were too slow to be competetive. We switched to 16FPS at IRI. The speeds with this actually allowed us to be competitive with the rest of the field. But in eliminations when our driver was playing rougher defense, we did pop breakers.

I would argue that with 6 Cims, the optimal points are about 10FPS for the low speed, and 16+fps for the high speed. With a 6 Cim drive, 10FPS will allow you to get around your half of the field basically as fast as possible, yet still allow you to not get in trouble with defense. Then if you need to cross the field, or the defender is faster then you, you can shift up to 16+fps and get across field very quickly.

If the field is split like 2010 or 2012, I would argue that a single speed 6 Cim 10 fps robot is the best you can get, because there is not enough room to use any higher of speed. For years like 2011, 2013 and 2014, you want to be able to shift up for extra speed to cross the field quickly. Those are the types of years shifter are useful, because they allow cross field travel faster.

I have to disagree with 6 cim single speed 10fps. You get a maximum of a few hundreths of a second less time to get to a point versus 4 cims, regardless of the distance. If you're gearing for 10fps, I would definitely just use a 4 cim drive.

Originally Posted by Abhishek R

I don't understand this, there are numerous factors as to whether a team wins or not, and I doubt shifting is a major one.

I would hesitate to use wins/losses as evidence to back up certain points, mainly because correlation is not causation.

As a counterexample to your counterexample, 610 in 2013 was extremely competitive to the point that they took home the Championship. They also only used one speed - I recommend looking up their philosophy on that, I believe Mr. Lim and Mr. Rob Stehlik have some in-depth posts about it. Back to my point, personally while I thought they had a very good robot, I would not attribute the win to them having the single best robot on the field, but rather to 1241, 1477, and 610's capability to play as an alliance instead of playing as three teams.

I still believe one choice is not inherently better than another, both have benefits and drawbacks, and it just depends on the game or team preference.

My example was actually to show that having a shifter isn't uncompetitive, although it's rather rather unnecesary. I wonder how many shifters were on Einstein these past years? That would provide more usable data.