Re: Team Update #18
 

Could you be a bit more specific? What "implied assumption" are you referring to?

The only assumptions used in the calculation were to ignore friction (which was explicit, not implied) and to ignore additional stored kinetic energy (rotational). These tend to cancel each other out. Over very short distances (such as 1/4"), the stored rotational KE easily overcomes friction, so it's actually a conservative assumption. You can demonstrate this for yourself, if you like: with the wheels raised, power your minibot with 12V and then remove the power. Observe how much the wheels turn after the power is removed.

If you have questions about the mecanum analysis I would be glad to discuss those, but not in this thread.

Re: Team Update #18
 

1.) Proximity sensors may work, yet are subject to calibration and vibration for a single data point. If you're using the difference of two data points, you still have to either increase sampling rate or # of sample to prevent false positives. Thus, changing to proximity sensors could pose the same issues/risks of the current system, which means there's no reason to change to proximity sensors.

2.) Light sensors are subject to noise and calibration. These may work better than proximity sensors since the noise is most likely easy to isolate, yet the tolerances in spacing would have to be tight to ensure even the slightest movement of the plate triggers the sensor. The designers simply may not have the time to implement something so quickly across all of the regional events for this week or even next week. This one seems the most feasible to me from a technical perspective though.

3.) Any instant replay sets a dangerous precedent. There will always be those who make assumptions based upon what's happened before, regardless of what's stated in the manual and regardless of what's talked about at driver meetings. I think FIRST just doesn't want to go there unless there's more thought put into it. That desire is reasonable in the longer term just to simply avoid stress during competition. Yet assuming a photo finish for 1 specific aspect of every game could probably also open up the types of activities we'd perform in future games. So I'm with you on this one ... subject to that caveat.

Re: Team Update #18
 

Alright, I did say implied and you were explicit about the friction. Yet it was still ignored in the conclusion without any explanation in the original posts. Ok, so there's the explanation, which I still don't think allows for a valid conclusion based upon the calculation. Your latest statement makes the implied assumption that the unloaded motors act identical to the real system during impact. Impact forces induce shock into the gearing which momentarily increases the friction in the gearing/bushings with a positive correlation to the amount of force applied to the plate. Impact forces induce friction in the bolts on the triggering mechanism, which may be bind on the plate if the impact causes a torque moement. Thus it may take 474 newtons at a specific instance in time to slow a minibot down without taking friction into account -- yet until we characterize the friction in the system we're making a dangerous* conclusion that "more than enough force is applied to moving the plate".

*dangerous because the process of coming to the conclusions would then be acceptable to use elsewhere.

I can't characterize the friction with numbers or equations, which I will concede. Interestingly, that's the very specific reason we [industry] do shock testing for our real world complex systems. The equations simply don't match the results of the real implementation.

Re: Team Update #18
 

I'm counting the days down until Waterloo. Get your popcorn ready.

Re: Team Update #18
 

I think the calculation allows a valid conclusion. Even if the gearbox were to completely lock up, the only effect it would have for purposes of this calculation would be the resulting friction between the (locked) wheels and the pole. Compared to 474 Newtons, such friction can reasonably be ignored for purposes of the conclusion, which is simply that a 3lb minibot traveling at 7fps which is decelerated to zero speed in 1/4" is exerting much more than 4 Newtons on the plate.

Re: Team Update #18
 

I don't feel the specification was unclear. There is a force given in the rules. The original tower used limit switches.

Logically one would think that a setup using limit switches would be momentary-ie not having to be triggered for some finite time period.

We should have foreseen that there was not enough real world testing done to show how robots hitting the towers would affect the response of the triggers?

In the future I guess when we see things like this you're saying we should assume FIRST will not be able to implement them properly and as such we should ask way more questions than necessary to ensure that everything works right?

Re: Team Update #18
 

...I've been chastised for championing this point for years.

So the refs, judges, and staff at the events are VOLUNTEERS. According to the people who don't see it my way, they should be forgiven any and all shortcomings, because they're VOLUNTEERS. Everyone claims that a cornerstone of the FIRST program is the tenet of GRACIOUS PROFESSIONALISM. To me, doing your job poorly, or worse, outright incorrectly (as in 2008 SVR Finals Match 2), is UNPROFESSIONAL. Not only that, but each and every team has mentors. All of whom are, themselves, VOLUNTEERS. The mentors are expected to know and follow the rules, because if they don't, and their teams robot doesn't comply with the rules, they don't get to play.

This isn't even what this argument is about, though. FIRST HQ is NOT run by volunteers. Many of the staff at FIRST HQ take a salary, paid for by our entry fees, and the sponsorships provided by industry. They are being paid to provide the teams with a product, a competition.

While I might be able to be convinced to forgive the shortcomings of a ref based on the "volunteer" argument, I will never be able to be convinced to forgive FIRST HQ for shamelessly doing things at the expense of their customers, the teams.

Re: Team Update #18
 

Does a standard scale, such as the ones used to inspect the robots at competition, give us the weight or mass of the robot? In other words, is the number we see equal to mass*[accel due to gravity] or equal to only mass?

Using a direct weight-equals-mass calculation, I get 0.586 feet stopping distance and 488N for stopping @ 1/4".

Dividing by the acceleration of gravity, I get a stopping distance of 0.060 feet (~3/4") and ~50N to stopping @ 1/4".

Re: Team Update #18
 

The scale measures force.

If you measure something on Earth, then you can use that force to calculate the mass, as follows: kilograms = pounds * 0.4535924

If you use the same scale on the Moon, you will get a force reading about 1/6 of what you got on Earth, even though the mass is the same.

Re: Team Update #18
 

In imperial units, a scale measuring pounds gives the weight of an object, not the mass. The imperial unit of mass is slugs, which is equivelent to approx. 32 lb-mass. When doing Newtonian calculations in imperial units, you must use the units slugs, lb-force, feet, and seconds. This yields the results you mentioned first, .586 ft and ~480 N (107 lbs).

Re: Team Update #18
 

I wouldn't say I'm a whiner, but I did propose a solution: http://www.chiefdelphi.com/forums/sh...57#post1040957

Re: Team Update #18
 

I am paid a coaching stipend, so technically I am not a volunteer.

(I was asked to itemize it out this year because we're doing budget negotiations and it looks like the admins are trying to cut compensation for extracurricular advisors, so I did the math and discovered that when it comes down to brass tacks I get paid about $0.07/hour if we keep it to only one summer project, but still...)

Re: Team Update #18
 

@pfreivald: ok, so maybe not ALL the mentors on FRC teams are 100% volunteers, but my point is still valid since you're in the overwhelming minority.

Re: Team Update #18
 

I EARN that seven cents an hour, dag-nabbit!

(I hope you realize I was just playing with you.)

Re: Team Update #18
 

For those 7 cents I expected more from you!

Just kidding. :)


The biggest problem I see with a 'photo finish' is something still needs to trip the camera. Taking constant video of the poles at a frame rate of 29.97fps may work, but close minibots will still be very close there, it may be hard to determine accurately. I'm not saying it isn't possible, but it's certainly something to think about.

Logically a computer system can only sense a condition that exists for a finite period of time related to the sampling rate. Since no specification was given regarding how long the force must be sustained, nor the sampling configuration of the system, it was an incomplete specification. Also, the force specification given appeared to be a rough estimate of the minimum required, not a clear definition of what we had to attain. I believe, as other posters have stated, you're hung up on that number.

Real engineering specifications are pretty explicit, I'm working on a project here at work that has an entire page explaining that the power LED must turn on when I apply power to the system. It even states how fast the LED must turn on when power is applied. The 'spec' we got from FIRST, if you can even call it that, was most definitely incomplete, and yes we should have foreseen complications from a lack of real world test data. I stand by my original argument, both the teams and FIRST are responsible for this.

Matt