Chief Delphi
Page 5 of 6 < 1234 5 6 >
Show 100 post(s) from this thread on one page

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)
-   Technical Discussion (http://www.chiefdelphi.com/forums/forumdisplay.php?f=22)
-   -   Is a shifting transmission really necessary? (http://www.chiefdelphi.com/forums/showthread.php?t=30797)

Hutch 26-07-2005 23:03
Re: Is a shifting transmission really necessary?
 
Couldn't agree more. I see a lot of teams use the same drive train (or very similar) over and over every year because it's good/powerful/whatever - but they don't learn anything. Their students don't get anything out of that. And I think that if you want to do what FIRST is really about (inspiring people), you have to let them learn.

Andy Baker 27-07-2005 10:27
Re: Is a shifting transmission really necessary?
 
Quote:
Originally Posted by Veselin Kolev
Here is why I make crazy drive trains.
I am not the person that particularly cares about the game. I make the drive train. It is proven that most drive trains can play every game in the existence of FIRST. One could have easily used a swerve drive every single year. The reason why I make what I make is because I want to learn more. However, there always is the rest of the team to negotiate drive trains with. So, when the team choses a particluar style of power train, (ex: swerve, 6WD, omni, etc...) I agree. And then I go the distance and make it as outrageous and advanced as I can with the time and resources that I have. In my case, I have a fully capable CNC shop that can make a complete drive train in 12 working hours. So what do I do? I design up the craziest thing I can possibly make. For example, if I ever do swerve drive again, its gonna be two speed, with a shifter on each module. Big? Heavy? No. It all depends on how you design it. But why add the extra 2 or 3 pounds, and the time, instead of making 1 speed swerve? Because I want to gain experiance, gain knowledge. Cory, you said that FIRST is not about innovation. It probably isnt. But I'm not one that can settle for making the same drive every year. I take risks, chances, and in the end make drive trains that make people's jaws drop.

Remember, people have a different mindset. I respect 8th year veteran teams that decide to use the kit of parts chassis and drive train for the lack of resources. I would do the same. But if I have the resources, I'm gonna have fun with them. And please, never scorn a team that makes something outrageous, like a 4 speed 6 motor transmission. They learned how to make a 4 speed 6 motor transmission, while you didnt.

On a side note: it is my opinion that learning from what you made is more valuable that winning with what you made.

These are some outstanding words. Veselin has hit the nail on the head, in my opinion. FIRST is defined by inspiration, not winning. Sure, everyone likes to win matches, but the point of the program is to be inspired and succeed in life.

Thank you, Veselin, for this wisdom.

Andy B.

mechanicalbrain 27-07-2005 17:12
Re: Is a shifting transmission really necessary?
 
Quote:
Originally Posted by Veselin Kolev
Here is why I make crazy drive trains.
I am not the person that particularly cares about the game. I make the drive train. It is proven that most drive trains can play every game in the existence of FIRST. One could have easily used a swerve drive every single year. The reason why I make what I make is because I want to learn more. However, there always is the rest of the team to negotiate drive trains with. So, when the team choses a particluar style of power train, (ex: swerve, 6WD, omni, etc...) I agree. And then I go the distance and make it as outrageous and advanced as I can with the time and resources that I have. In my case, I have a fully capable CNC shop that can make a complete drive train in 12 working hours. So what do I do? I design up the craziest thing I can possibly make. For example, if I ever do swerve drive again, its gonna be two speed, with a shifter on each module. Big? Heavy? No. It all depends on how you design it. But why add the extra 2 or 3 pounds, and the time, instead of making 1 speed swerve? Because I want to gain experiance, gain knowledge. Cory, you said that FIRST is not about innovation. It probably isnt. But I'm not one that can settle for making the same drive every year. I take risks, chances, and in the end make drive trains that make people's jaws drop.
he's my FIRST idol everything he said is exactly what i strive for! i agree and though you can't hear it im claping.

JVN 27-07-2005 23:08
Re: Is a shifting transmission really necessary?
 
Quote:
Originally Posted by Andy Baker
FIRST is defined by inspiration, not winning. Sure, everyone likes to win matches, but the point of the program is to be inspired and succeed in life.
Striving your hardest to win, trying your best to design a competitive elegant solution is DEFINITELY what FIRST is about.
This is the medium provided to us to inspire HS students. Otherwise FIRST wouldn't give us a game, they'd say "build a cool robot". (Wildstang would probably win. ;))

Learning how to distill a problem, and find a simple solution is critical to the FIRST process (as it is to the engineering design process).


Which design is worthy of more praise?
  • An overly complex system which innovates for the sake of innovation. A system which (while technically incredible) is overkill for a FIRST robot, and unneeded for the current game.
  • A system which is laughable in it's simplicity, yet incredible in it's functionality. One which fulfills the needs of the game with a minimum of "effort".
Personally...
I prefer the second option.

Here is another question, which is more education/inspirational to the students who design and build it?

I argue that they are equally beneficial.
I would even go so far as to venture, that option 2 is tougher to do, and much more beneficial to the students who accomplished it, in the long run.



Many in this thread have been quick to scorn those who do not innovate, which REALLY pisses me off. Having elegance in simplicity is often times more virtuous than being innovative.

Arefin Bari 27-07-2005 23:19
Re: Is a shifting transmission really necessary?
 
Quote:
Originally Posted by JVN
Striving your hardest to win, trying your best to design a competitive elegant solution is DEFINITELY what FIRST is about.
This is the medium provided to us to inspire HS students. Otherwise FIRST wouldn't give us a game, they'd say "build a cool robot". (Wildstang would probably win. ;))

Learning how to distill a problem, and find a simple solution is critical to the FIRST process (as it is to the engineering design process).


Which design is worthy of more praise?
  • An overly complex system which innovates for the sake of innovation. A system which (while technically incredible) is overkill for a FIRST robot, and unneeded for the current game.
  • A system which is laughable in it's simplicity, yet incredible in it's functionality. One which fulfills the needs of the game with a minimum of "effort".
Personally...
I prefer the second option.

Here is another question, which is more education/inspirational to the students who design and build it?

I argue that they are equally beneficial.
I would even go so far as to venture, that option 2 is tougher to do, and much more beneficial to the students who accomplished it, in the long run.



Many in this thread have been quick to scorn those who do not innovate, which REALLY pisses me off. Having elegance in simplicity is often times more virtuous than being innovative.
First off, I must say very nice post john.

When I used to be a freshman in high school, I used to think about drive trains, specially the ones which has a lot of complexity. I used to say "Woooo, it's so cool, let's build them." But over years I have learned (Thanks John) that simplicity is the answer to solution. It is much more easier to solve a problem when you think simple.

Is two speed necessary? If you ask me, I wouldn't do it for a first or second year team. I would build a robot that is fast enough, build a very simple mechanism which serves the purpose of the game. But at the same time, if my team built few prototypes in the pre-season and knows EXACTLY how it works, I would go with it. I would go with what I know works. The team would only have 6 weeks to work on the robot and then ship it off. Its not much time to experiment with new drive trains and such. In another word, I would go with the 2nd option that John pointed out in the previous post.

sanddrag 27-07-2005 23:27
Re: Is a shifting transmission really necessary?
 
I agree with all the things being said but I never look down upon a team (and I'm not saying any of you do) for pushing the limits, challenging themseleves, and striving for more, even if it is without reason.

JVN 27-07-2005 23:28
Re: Is a shifting transmission really necessary?
 
JVN's "How to determine if a shifting transmission is really necessary?" Cookbook:

First you need to figure out a few things:
(We will ignore things like efficiency, in this exercise. Sorry Patton.)
  1. Determine: Wheel Friction, Robot Maximum Weight (with lifted weight, i.e. balls/goals) These factors can all be calculated based on overall robot decisions. (What tread do we use? What is our robot going to do?)
  2. Determine: What motors will go into the gearbox?
  3. Determine: What max speed do you want your robot to go?
Now Run some Calculations:
  1. Based on the breaker limits (i.e. 40 Amps) how much torque load can each of your motors have on it without blowing the breakers?
  2. Based on this max torque load, what reduction is necessary to "slip" the wheels (using the robot weight and wheel friction values listed above), if the robot is pushing against something, in the max loading condition.
  3. Based on this required "slip" reduction, what is the speed of the robot?
Now... we know what our "pushing speed" is. This is critical, because every robot should be able to slip wheels while pushing a wall, and not pop the breaker.

Here is the big question:
Is this calculated robot speed at least as fast as the max speed we want?

If yes:
  1. Woo Hoo! We're done. Gear the robot at the max speed, and live happily every after with a 1-speed gearbox.
If no, we have some options:
  1. The Tim Allen approch (add more power). Adding additional motors to the drivetrain will allow the robot to be geared faster, while still slipping the wheels.
  2. Decrease Load: Reduce Wheel Friction, or Decrease Robot Weight. By decreasing the load on the motor, it allows a faster speed, while still slipping the wheel.
  3. Pop breakers: "Math, shmath... we'll be fine". Gear the robot as fast as you'd like, and not worry about it.
  4. Add an additional speed. This allows for a "high gear" for moving around the field, and a "low gear" for pushing without popping breakers.
If none of the first 3 options are available... you'd better try a shifter.
This isn't the end of our trouble, because now we actually have to DESIGN the shifter, then we have to BUILD the shifter but that is a challenge for another time. Whether or not a team is capable of doing it, certainly doesn't determine if it is necessary. If they can't do it, well they'd better try options 1-3.



In short, the 2 main determining factors for whether a shifter is needed or not are:
  1. What the team "wants" for Physical Robot Characteristics
  2. What the team "wants" for a robot top speed.
  3. Nothing else...
For more information on drivetrain gearing calculations, check out the following: http://www.chiefdelphi.com/forums/pa...le&paperid=498




If you posted an opinion in this thread and aren't familiar with the process I outlined above. (The actual mathematical justification behind a shifter.) I directly question the validity of your opinion. Shouting random platitudes about drivetrains, without understanding the physics behind them is meaningless.

...yes, there are several people this is directed to.

JVN 27-07-2005 23:33
Re: Is a shifting transmission really necessary?
 
Quote:
Originally Posted by sanddrag
I agree with all the things being said but I never look down upon a team (and I'm not saying any of you do) for pushing the limits, challenging themseleves, and striving for more, even if it is without reason.
I CERTAINLY would not do this.. live and let live...
So... let us "simple-elegant" fools live too.

Hutch 28-07-2005 00:36
Re: Is a shifting transmission really necessary?
 
I guess the thing is, any way you swing it, a two speed thing isn't exactly innovating "just to innovate". It might be innovating just to teach kids, or just to stop drawing ridiculous amounts of current/blowing breakers, or just to make it so you can go that much faster when you want to while still having good torque when you need it. But it's certainly not what I would call innovating just to innovate.

Something crazy like a shifting crab (can you say CVT crab? :D ), yeah, that's ridiculous and even assuming it work be reasonably possible to do within the contstraints of FIRST, that would be innovating just to innovate. But up to that point, I have a hard time believing that doing something beyond the bare minimum is doing too much.

NOTE: It's fine with me if you choose to do a simple tank drive - I won't look down on you. I will just think that you are missing out on a great opportunity to learn and improve.

sanddrag 28-07-2005 00:40
Re: Is a shifting transmission really necessary?
 
Yep, we are all crazy fools in our own ways. :D The reason we built a shifter was because we had never had one before, and we always wanted one, and we decided before the season that no matter what the game was, we would have a shifter. We were plenty capable of fabricating it, and it would not put us at any disadvantage in game play, and we were bound to learn a great deal from it, so we decided before the game even came out that we were going to have it. After the game came out, the reason we had six motors was "the Tim Allen Approach." Also, we knew a few teams might have 6 motor drives, we knew many teams would have shifting transmissions, and we knew nobody would have both. So, we did! We wanted to be able to face and win any and every pushing match or speed race that came our way. But unfortunately, the game didn't turn out to be played like that (speed yes, pushing no). But we still had loads of fun (and learning) designing, building, and driving our two speed transmission. It is loads of fun to upshift on the fly and then almost instantly be on the other side of the field. Downshifting on the fly is fun too, but at full speed it can almost bring the robot to a dead stop and send it onto it's nose! (2.66:1 difference between high and low). And the best part is that it has been the most reliable drive system we have had thus far (one minor incident due to a loose press fit) and it's very efficient too :)

Veselin Kolev 28-07-2005 18:33
Re: Is a shifting transmission really necessary?
 
A quick one...

Shifters are useful depending on what kind of driver you are. Just like in cars, you can keep it in "higher" gears and have a transmission that does not burn up battery power, but has small acceleration, or you can use your gearing aggressively to achieve maximum acceleration. JVN, while I agree with your post, I will mention that while your efficient single speed gearing style is good for having a top speed and big pushing power in one, having a gearbox geared "high" will result in slower acceleration. Having a shifter always increases your acceleration, at least by something bearly noticeable.

I am an aggressive driver. I'm the type that always, ALWAYS, starts the robot in low gear and accelerates to high, just like a car. Even if it only takes half a second to reach shifting speed. In FIRST, half a second can easily mean winning or losing. I think that for my stlye of driving, having a shifting transmission does give a bit of an advantage.

And finally, to adress the original topic.
Shifters are in no way necessary. You can have a winning robot, with a powerful transmission, without a shifter. Just do some math, optimize everything. Even with the KOP transmission you can get crazy power by just doing some math. If you think a shifter will help your robot's performance, enough so that it's worth making, do it! Always make the best of what you have.

Example: "If you have a 3 footh by three foot, inch thick slab of aluminum, don't let it sit and oxidize! Call up a watercutter and make your chassis out of a solid block this year"
*hint hint*

Paul Copioli 28-07-2005 18:56
Re: Is a shifting transmission really necessary?
 
I am in the JVN "keep it simple" camp.

I have not always been in that camp. Here is the history:

Pre-Paul (pretty simple ...)

1999 - Van door motors directly coupled to wheels (2WD)

2000 - Drill motors directly attached to wheels (2WD)

(Paul sticks his nose in ...)

2001 - Andy Baker designed TechnoKats shifting transmission built right from his prints with 6 wheel drive (initially).

2002 - 4 wheel swerve drive using the CCT (see white papers section) shifter.

2003 - 4 wheel drive with rear steering using the Baker style shifter.

2004 - Single speed 6 wheel drive using a shifter for power take-off for the winch.

2005 - Single speed 6 wheel drive using the Tim Allen approach (4 CIMs and 2 FP).

2005 was, by far, our simplest robot yet and simplest drive base. Our acceleration was awesome, our top speed was good enough (9 fps). And, the best part of all ... we had zero problems at our first regional. I have found, at least for my students, that working right from the start is one of the most inspirational parts of the season. When what you worked on for so long works ... right out of the box.

I used to encourage my students to innovate just cause it is cool, but that just justed work for us anymore. I can tell you this: I don't innovate just because it is cool at my real engineering job so why should I in FIRST? Making an elegantly simple robot is definitely the way I will encouraging my team to go in the future.

As far as the game goes .. the drive base, once working, just isn't that big a part of the game (exception: 2002). Making an inventive manipulator is much harder than making an inventive drive base and as much fun. If you get inspired by making cool, inventive, and complicated drive trains then great; but I am staying on the side of the simple folk ........

-Paul

techtiger1 28-07-2005 21:11
Re: Is a shifting transmission really necessary?
 
Wow Paul is certainly a tough act to follow, I will attempt to do my best though. :) There is no necessity to be able to shift first of all, shifting is a nice ability to have though. Second of all you should design to meet the requirements of the game. All I know is that this year some of the most successful robots didn't shift 67,71,217,108 and 33 I think didn't shift there are many others but just to throw some out there. So think about it is this really needed for our application and for the game. Finally designing transmissions may be fun, but it's not fun when a bad transmission leads to a whole bad FIRST season. 1251 (my team) built a fairly simple robot this year, we used kit transmissions and kit chassis and got two regional finalist awards. Just some food for thought when thinking about this subject.

My two cents,
Drew

BTW Paul the three motor transmission is crazy !!!!! :D

Raul 01-08-2005 00:03
Re: Is a shifting transmission really necessary?
 
I don't know about you guys, but I'm thinking 6 wheel, 3-speed swerve next year! :ahh: ;)

Raul 01-08-2005 00:05
Re: Is a shifting transmission really necessary?
 
Oh yeah, and I forgot to add - with independent suspension.


All times are GMT -5. The time now is 23:59.
Page 5 of 6 < 1234 5 6 >
Show 100 post(s) from this thread on one page

Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi