Big Minibots

[sanddrag]

Originally Posted by dodar

Why is it that alot of teams are building these huge and complex minibots? I mean teams are building minibots that look like they are maxing out the size limits near 12"x12"x12". Why build a big one over a small one?

Potential reasons:

1. Because they don't have someone who understands Physics leading the design.
2. Because they don't read Chief Delphi
3. Because they are ignorant of the world around them.
4. Because they lack the resources, or design experience to build something custom

I don't mean this post to be harsh. I lot of teams have put a lot of work into some very successful larger minibots. But as a matter of engineering and physics principals, a small and light one is the only correct way to do it, if the goal is to get to the top the fastest.

[Richard Wallace]

As a famous engineer likes to say, design is an iterative process. I've visited the pits of a few teams* that began by building larger, heavier minibots, and evolved their designs toward smaller, lighter minibots. The physics is the same, but the numbers (weight, friction, wheel diameter, gear ratio) are different. Lighter = faster, after you get the design optimized.

*One of those teams has a leader who is also a physicist, and a teacher. I am certain she did not do the design for them, but I am also certain she did not allow them to believe their first design was good enough. The advantage of understanding the physics behind an engineering problem lies in knowing how to predict the theoretical best-case result -- after you have that knowledge, iterate until your actual result is so close that further effort is better spent on something else.

[Mike Betts]

Originally Posted by sanddrag

Potential reasons:

1. Because they don't have someone who understands Physics leading the design.
2. Because they don't read Chief Delphi
3. Because they are ignorant of the world around them.

I don't mean this post to be harsh. I lot of teams have put a lot of work into some very successful larger minibots. But as a matter of engineering and physics principals, a small and light one is the only correct way to do it, if the goal is to get to the top the fastest.

I normally respect your opinion but let's look at engineering and physics for a minute...

How many of those "small and light" minibots are using limit switches rated for AC only? Answer: All of them.

From an engineering perspective, this is reckless and stupid. AC current switches through zero 120 times a second and the arc created when one tries to open a circuit is extinguished.

This does not happen for a DC current. It is much harder to switch... The arc tries to bridge the gap resulting in contact pitting and/or welding. This arcing becomes even larger when switching inductive loads (like DC motors).

To pick an example, the Honeywell microswitch in the KOP is rated for 11 amps AC only. An electrical engineer would never use it for DC currents except at milliamp (logic signal) levels. And in the extreme cases where expected lifetime is measured in thousands of cycles, never at all...

Now, you can specify DC rated microswitches but they ain't so micro...

So... We have teams who use NXT controllers and NXT touch sensors and they end up with large, non-competitive but better engineered minibots. And then we have poorly engineered but competitive minibots whose mentors have turned a blind eye to good engineering and are just hoping that they get though the season without failures setting in.

The correct way is not necessarily the competitive way...

Just who is ignorant of physics or of the world around them?

Regards,

Mike

[sanddrag]

Originally Posted by Mike Betts

I normally respect your opinion but let's look at engineering and physics for a minute...

How many of those "small and light" minibots are using limit switches rated for AC only? Answer: All of them.

From an engineering perspective, this is reckless and stupid. AC current switches through zero 120 times a second and the arc created when one tries to open a circuit is extinguished.

This does not happen for a DC current. It is much harder to switch... The arc tries to bridge the gap resulting in contact pitting and/or welding. This arcing becomes even larger when switching inductive loads (like DC motors).

To pick an example, the Honeywell microswitch in the KOP is rated for 11 amps AC only. An electrical engineer would never use it for DC currents except at milliamp (logic signal) levels. And in the extreme cases where expected lifetime is measured in thousands of cycles, never at all...

Now, you can specify DC rated microswitches but they ain't so micro...

So... We have teams who use NXT controllers and NXT touch sensors and they end up with large, non-competitive but better engineered minibots. And then we have poorly engineered but competitive minibots whose mentors have turned a blind eye to good engineering and are just hoping that they get though the season without failures setting in.

The correct way is not necessarily the competitive way...

Just who is ignorant of physics or of the world around them?

Regards,

Mike

All valid points. Perhaps I should have been more specific in specifying engineering for the task at hand which is getting up the pole in the least time, with a life cycle of only a few times.

I completely agree on the switches. It's bad practice to use a switch for something it isn't meant for. In this case, it does seem to work, for a little while anyhow.

[theprgramerdude]

Originally Posted by Mike Betts

I normally respect your opinion but let's look at engineering and physics for a minute...

How many of those "small and light" minibots are using limit switches rated for AC only? Answer: All of them.

From an engineering perspective, this is reckless and stupid. AC current switches through zero 120 times a second and the arc created when one tries to open a circuit is extinguished.

This does not happen for a DC current. It is much harder to switch... The arc tries to bridge the gap resulting in contact pitting and/or welding. This arcing becomes even larger when switching inductive loads (like DC motors).

To pick an example, the Honeywell microswitch in the KOP is rated for 11 amps AC only. An electrical engineer would never use it for DC currents except at milliamp (logic signal) levels. And in the extreme cases where expected lifetime is measured in thousands of cycles, never at all...

Now, you can specify DC rated microswitches but they ain't so micro...

So... We have teams who use NXT controllers and NXT touch sensors and they end up with large, non-competitive but better engineered minibots. And then we have poorly engineered but competitive minibots whose mentors have turned a blind eye to good engineering and are just hoping that they get though the season without failures setting in.

The correct way is not necessarily the competitive way...

Just who is ignorant of physics or of the world around them?

Regards,

Mike

This is FIRST. People shouldn't give a hoot about lifetime ratings in this case; the part needs to last a grand total of about 1-2 minutes max in competition. I'd take something even lighter than the KoP Honeywell switches if I could get my hands on them. It is better engineering to realize that the task at hand requires performance wayyyyyy over survivability.

[wilsonmw04]

Originally Posted by theprgramerdude

It is better engineering to realize that the task at hand requires performance wayyyyyy over survivability.

i'll remember to tell that to a team who's minibot burns out half way up the poll in elims.

[theprgramerdude]

Originally Posted by wilsonmw04

i'll remember to tell that to a team who's minibot burns out half way up the poll in elims.

If it's burning out halfway up the pole, it's safe to say they didn't engineer the thing - they just built it and prayed it would work. Complete burnout's like that would occur because the motor is operating in a red zone on the power curve, something that could have easily been avoided had analysis been done on the machine.

[wilsonmw04]

Originally Posted by theprgramerdude

If it's burning out halfway up the pole, it's safe to say they didn't engineer the thing - they just built it and prayed it would work. Complete burnout's like that would occur because the motor is operating in a red zone on the power curve, something that could have easily been avoided had analysis been done on the machine.

but wait, hold on a sec. I thought you said that speed was "wayyyyy" more important than survivability. So by your example,not sacrificing too survivability for speed is a good thing. Thanks for proving my point :-)

Be aware not everyone analyzes the game the same way. Don't be so bold as to say your analysis is better than someone Else's. We all come from different places and experiences. There needs to be more middle ground around here and not all this "i'm right and you're wrong" crap-o-la. It's getting a bit tiresome.

[Mike Betts]

Originally Posted by theprgramerdude

If it's burning out halfway up the pole, it's safe to say they didn't engineer the thing - they just built it and prayed it would work. Complete burnout's like that would occur because the motor is operating in a red zone on the power curve, something that could have easily been avoided had analysis been done on the machine.

Hey... My team used the micro-switches also... But to call someone ignorant or lacking in a science background because they build a big minibot is just wrong...

The same logic, being "competitive", that permits switches to fail applies to allowing motors to burn up... That is assuming that all of you self righteous "engineers" actually considered the consequences of your actions... There are teams who "guessed" at the motor torque curves and others who used switches that "seem" to work... Is one engineering oversight less important than the other?

I prefer to think that any team who fielded a working minibot and delivery system should be praised and not looked down upon because they took a different path to the solution...

This can't be what FIRST is about...

I hope I have made my point and I will post no more on this subject.

[theprgramerdude]

My team used the switches also; I have to say that they work great, at least for the time being. I just wish that their application required them to be a bit more robust; we've already broken a few by crushing them.

Mike, a working minibot is a working minibot. No one reasonably doubts that that is a good thing to achieve. However, I think the point that people are trying to make, which may be confusing/rude some due to it's harsh language, is that it can be improved to be better at the goal it's trying to achieve. FIRST isn't about teaching kids to do it once, and then be done.

[Jin Hayashi]

In terms of physics, we would have loved to have made a lighter minibot with a better gear ratio. In terms of practicality,we worked on the KISS principle. KISS won over physics for us.

Our reason for making a big minibot was due to lack of experience and lack of shop resources.

Our team didn't have any experience with FTC before the kickoff. The minibot was built by a student who joined our team three weeks after kickoff and a parent who was a first year FLL mentor. Our minibot used the NXT, due to the mentor's knowledge of NXT programming and its sensors.

We are also working out of the garage of one of our mentors. We had to budget our usage of power tools for building the hostbot and minibot. Which limited the amount of tool time for the minibot team for making it faster.

I was amazed and pleased that we got a minibot working with the amount of experience and resources the two had before shipdate. Although we lost many minibot races with other teams, it was satisfying just to successfully launch the minibot.

[PAR_WIG1350]

Originally Posted by Mike Betts

I normally respect your opinion but let's look at engineering and physics for a minute...

How many of those "small and light" minibots are using limit switches rated for AC only? Answer: All of them.

From an engineering perspective, this is reckless and stupid. AC current switches through zero 120 times a second and the arc created when one tries to open a circuit is extinguished.

This does not happen for a DC current. It is much harder to switch... The arc tries to bridge the gap resulting in contact pitting and/or welding. This arcing becomes even larger when switching inductive loads (like DC motors).

To pick an example, the Honeywell microswitch in the KOP is rated for 11 amps AC only. An electrical engineer would never use it for DC currents except at milliamp (logic signal) levels. And in the extreme cases where expected lifetime is measured in thousands of cycles, never at all...

Now, you can specify DC rated microswitches but they ain't so micro...

So... We have teams who use NXT controllers and NXT touch sensors and they end up with large, non-competitive but better engineered minibots. And then we have poorly engineered but competitive minibots whose mentors have turned a blind eye to good engineering and are just hoping that they get though the season without failures setting in.

The correct way is not necessarily the competitive way...

Just who is ignorant of physics or of the world around them?

Regards,

Mike

1) although it may be common for teams to use AC microswitches, not all teams do. A small minibot could be engineered very well.

2)Engineers solve problems to be effective and efficient. The problem at hand is triggering the target before the other minibots. If you are too busy trying to extend the lifespan of a $1.50 (more or less) switch at the cost of abandoning the overall goal, you are not engaging in good engineering practices either. If an NXT gets smashed you are out $150.00, for just 5% of that cost, you could replace a microswitch 5 times. In addition, the initial cost for a microswitch based system is about 0.6% of the initial cost of an nxt based system if you include the touch sensor and the motor controller. Thus, the most effective system in terms of how fast the goal is reached and cost is the small minibot. The big minibot just has too many disadvantages.

[boomergeek]

Originally Posted by Mike Betts

I normally respect your opinion but let's look at engineering and physics for a minute...

...

The correct way is not necessarily the competitive way...

Just who is ignorant of physics or of the world around them?

Regards,

Mike

Destruction of material and risk taking IS part of good engineering, especially part of race engineering.

The competition is not about the team that builds the minibot with the least cost, most reliable parts that never degrade over the course of a season.

Teams spend many thousands of dollars per season and most of the parts depreciate in value very quickly most KOPs are replaced within a few years.

How much engineering maintenance does a NASCAR need compared to the family minivan?

Per mile, how much more often does a NASCAR fail as compared to a family minivan?

In reading the competition manual as a good engineer, are the requirements for the portion of the competition regarding the minibot more like NASCAR or like the family minivan?

Is it correct engineering to teach students to build a minivan for a NASCAR race?

There are plenty of relatively reliable 2.5-3 lb minibots.

If FIRST wanted the best teams to use the NXT and motor controllers, they should have make a challenge more like a use of a minivan- running all over town, picking up toddlers and groceries, making controlled stops, obeying all speed limits and signals, etc.
But then, who wants to go to a competition to watch that?
Probably not the stuff a Cirque du Soleil promoter could promote.

Has anyone had to replace a KOP limit switch from electrical overload in this challenge? What is the failure rate?
(Team 241 has used them without failure over 100 times.)

What is the perception of the other failure rates that the large minibot engineers have actual data to back up their concern?

[wilsonmw04]

Originally Posted by sanddrag

Potential reasons:

1. Because they don't have someone who understands Physics leading the design.
2. Because they don't read Chief Delphi
3. Because they are ignorant of the world around them.

Wow,
I didn't realize that you could be wrong three times with three bullets. I don't mean this to be harsh, but here is where you are wrong.

1. my kids understand physics. Since that's what I teach, I make it my mission to sneak it in when I can.

2. I read CD as well as a few other members of the team. I'm not quite sure what was meant by this statement. We should all take the best ideas and use them as our own? Wouldn't that make this whole process so VERY boring? Yeah, I pointed out the "one day minibot" to my team today. They thought it was very cool but is beyond our ability to make (the most complex machining tool we have is a drill press). They want to focus on a consistent deployment system after watching 3 weeks of events. To sum up the mood of the team: getting a 4 sec minibot up the poll every match >> 2 sec minibot up the poll 80% of the time.

3. ignorant is such a "large" word to use. I'm not sure how to respond to that, so I won't.

Some teams want the students to learn something along the way (i'm not saying you don't Sand). FIRST gave them the problem. They are the ones who are deciding how they chose to solve the problem. I can point out the mistakes and help them along the way, but I refuse to do it for them.

I think it comes down to how you want to inspire your team. I have my way, other mentors have different ones. Who knows which is best. I don't. I don't think anyone truly can. What I can say is there isn't one correct way to do anything.

[boomergeek]

Any team that first worked on their own to try to design a minibot, understand the physics and math and the engineering specifications and then built it and experimented and refined it- has done great engineering instruction.- It's quite commendable and appropriate to spend long hours brainstorming and experimenting free from the din of other teams ideas.

In real engineering, when is not appropriate to understand the approaches of your competitors? In real engineering, customers don't want engineering teams that can just copy: they want a team that create ideas and throw away ideas and sometimes learn and improve from what they see from competitors. If you are heavily inspired from some other team's work, then spend the time to understand the physics and spend considerable time to improve on it versus just copying it.

We were not impressed by the reliability of the gearboxes: especially using the hubs directly on the wheels: after only a few times up the pole using about 10 lbs of normal force, the wobble seemed obvious that the gearboxes would not last a season's worth of stressful pole climbing.
The gearbox sounded inefficient and the speed of our pole climbs were approximately the same as those shown in the Kickoff video.
We saw smoke and lost a motor/gearbox that first week after we got the kit.
We lost a second motor on our bigger minibot (4.4 lbs) that uses the motor/gearbox after we got back from our first Regional.

It was partly the quest for reliability that led us to simplify and get rid of the gearbox. Less weight means less stress on the motors. Dynamometer results posted here on CD helped.

Is it good engineering not to network and know how good the competition is?

Because of the cross fertilization of ideas between teams, the robots at week 8 of competitions can be much stronger than those at the first week.
The question is: what should your team do with this wealth of information and ideas for solving the challenge? Should your team ignore it? Should your team understand how fast the competition is but not how?

These are not necessarily easy questions to answer.

Any mentor/volunteer willing to spend time and energy facilitating any of the physics lessons, the engineering design, the fabrication, or the testing, etc. is helping students immeasurably.

Many teams end up spending TOO much time redesigning their robot and school work or other aspects of life suffer. The key is balance and mutual respect.

Big ; little ; can't we all just get along?