Team Update #18

[JB987]

Originally Posted by Ether

Using the nominal dimensions on the field drawings for the pole OD, the threaded bolts, and the hole diameters and location in the polycarbonate plate, plus the apparent location of the sensors, it appears that if the plate is lifted off-center so that it pivots on the bolts it will reach an interference condition with the bolt threads and the pole before it travels 1/4" at the switch. FWIW.

So...impacts within 2" of the pole (which is where every one of the "fast" minibots that occasionally didn't trigger hit) are less likely to lift the plate in an exaggerated off-centered manner right? This suggests an off-centered lift isn't the culprit in those cases in my mind.

[jspatz1]

It would be great if the trigger design itself provided whatever dwell was necessary for the system sample rate to detect, was not sensitive to bumping the tower, did not matter where it was contacted, did not have any bolts or obstructions that kept it from working every time, and required only the specified force with no other mystery requirements.

Such as:

[Ether]

Originally Posted by JB987

So...impacts within 2" of the pole (which is where every one of the "fast" minibots that occasionally didn't trigger hit) are less likely to lift the plate in an exaggerated off-centered manner right? This suggests an off-centered lift isn't the culprit in those cases in my mind.

I don't follow your logic.

[Mike Copioli]

Originally Posted by colt527

So even if you can prove that something unfair happened, please lets all keep a calm and professional composure and remember our true greater purpose here.

I am well aware of the greater purpose here....are you? It's to inspire. It is not inspiring to work as part of a team, to inspire others to strive for excellence and success only to have that under minded by a poorly documented field condition that changes the outcome. As far as unprofessional and uncalm at what point did you think my post was any of those things?

You may want to consider these points before you appoint yourself mediator in a battle that has not even happened yet.

Originally Posted by colt527

I am not saying that this conversation has turned into this.

Then what are you saying?

I'll tell you what, when and if it happens to your team you can pretend it did not affect the outcome and then later face the disappointing faces your student members who worked so hard on solving a problem only to have the question change after answering it correctly. Oh if first can't find 4 dudes with a button I know three that would be happy to help with this problem.

[colt527]

Originally Posted by Mike Copioli

As far as unprofessional and uncalm at what point did you think my post was any of those things?

The simple answer is: at no point.

I wasn't referencing your post. It was a different one where the author deleted after reconsidering its GPness. Without that post for context, I think mine seems overdramatic.

The intention of the post was not really meant to chastise a single individual or set of individuals, but to remove some of the tension in the air, which seemed to be coming up after the now deleted post. So if it came off that way, I apologize, like I said, wasn't the intent.

[JB987]

Originally Posted by Ether

I don't follow your logic.

Given the amount of play/spacing between the floating plate and the bolts as well as the pole itself as you referenced, wouldn't a push on one side far out at the perimeter cause more of a "tilting" effect than a push near the center of the plate?

[Matt Krass]

I for one am on the bandwagon of wanting a final official specification from FIRST. I want to know exactly the force I'm expected to exert, for exactly how long, and exactly how clean the signal needs to be. I want to know what kind of sample rates, sensor wiring, and hysteresis is in place here. FIRST owes us that.

Before anyone jumps on me for blaming FIRST, I want to make it clear that I'm not mad at them. They gave us a very vague specification, we designed to it, we did our best, and it appears we have fallen short.

FIRST also designed to it, they designed a system to react to that vague specification, it seems they also have fallen short. They're now taking steps, within reasonable bounds, to correct that. I agree it's not wonderful, and it is unfortunate that teams in prior events are not going to benefit from this system, but I think we can all agree we'd rather improve it, then have everyone suffer in the name of fairness, especially on a playing field that will never be entirely level anyway.

Also, I agree with the people here that are concerned the system still isn't reliable, my inner engineers hunch is telling me it's not going to work as well as they want, and I don't agree with their approach of "We fixed it, trust us, and deal with it.".

But wait! I hold the teams responsible as well! We were given a vague specification, and everyone made their own interpretation of it, and many of us were wrong, and this is our mistake. Questions should have been asked weeks ago, more research done by teams, we all dropped the ball. I'm not saying people should assume failure, or injustice, but just one team asking FIRST for more information could have made a world of difference.

As I said before, I'm glad FIRST is stepping up and trying to fix their foul-up, now it's our turn to return that professionalism, ask questions and see what we have to work with, and do our best to meet the most accurate specification we can wrangle out of them.

Matt

[colt527]

I know this has been suggested already, but I think the right place to pose questions is directly to the Q&A:

http://forums.usfirst.org/forumdisplay.php?f=1465

You need to have the team leader account to post there, but I am sure that FIRST will be more than receptive to answer all the questions posed here. If people do get an answer to some of the common questions, can they please post them here as well?

The main questions that seem to need answering is:

How long must the contacts be connected for FMS to trigger / what other technical constraints of the triggering system should be kept in mind by the teams?

I do not have an account for this, otherwise I would pose the question myself. I do think teams have full right to ask these questions though.

[Chris Hibner]

Originally Posted by Schnabel

Also another note for the fast minibots. When you go up so fast that you hit the plat and skew it from horizontal you actually bind it against the threads of the bolts holding it together. This means the plate won't move past a certain point, which just may be before the trigger point.

Thank you Eric, and Ether for your analysis. I was starting to doubt my eyes and memory. I remember when the refs tested the plate with a stick at Kettering, the first couple of pushes bound the plate.

As one of the teams with the light, fast, tiny minibots, here's hoping the new solution works.

[Ether]

Originally Posted by JB987

Given the amount of play/spacing between the floating plate and the bolts as well as the pole itself as you referenced, wouldn't a push on one side far out at the perimeter cause more of a "tilting" effect than a push near the center of the plate?

All else being equal it would seem reasonable to suppose that. But there are other dynamic factors at play whose effect on jamming due to tilting is difficult to quantify.

[Dad1279]

Originally Posted by Matt Krass

..... Questions should have been asked weeks ago, more research done by teams, we all dropped the ball.......

Questions were asked. Here's one:http://forums.usfirst.org/showthread.php?t=16174

The answers were vague. We didn't drop the ball. We built a pole, with lexan plates, three limit switches & timer. Worked 100% with our minibot during development.

At NY, our minibot triggered the pole most of the time, pole didn't trigger once, but it was counted. I can only hope they are all counted at our next competition in DC.

If a time specification has been introduced, that has changed the specs and the game. We expected and designed to "Only one limit switch needs to be actuated" as per the Q&A Jan 16th.

[jspatz1]

I have submitted the following question to the FIRST Forum Q&A:

"With any electro-mechanical contact closure that is sensed by a programmable control system, there is both a force component required to perform the closure, and a time/dwell component for the control system to register the closure. The Competition Manual provided the force component necessary to activate the minibot trigger contacts, but did not provide the time/dwell required for this contact to register with the FMS. What is the minimum length of time that the specified force (2N-4N) must be applied to the trigger plate in order to be recorded/registered by the FMS?"

[Ether]

Just for fun (for those who are so inclined):

A 3 pound minibot is traveling up a pole at 7 feet per second.

It is depowered and a downward force of 4 Newtons (plus gravity) is immediately applied to it.

1) Assuming no friction, how far will the robot continue to rise until it reverses direction?

2) Instead of 4 Newtons, how much (constant) force would need to be applied to limit the rise to 1/4" ?


Note: The purpose of this exercise is to show that the force exerted on such a minibot by the plate assembly (and thus the force exerted by the minibot on the plate assembly) far exceeds 4 Newtons.

[MrForbes]

It looks like we're making progress on figuring this out.

[jspatz1]

Originally Posted by Ether

Just for fun (for those who are so inclined):

A 3 pound minibot is traveling up a pole at 7 feet per second.

It is depowered and a downward force of 4 Newtons (plus gravity) is immediately applied to it.

1) Assuming no friction, how far will the robot continue to rise until it reverses direction?

2) Instead of 4 Newtons, how much (constant) force would need to be applied to limit the rise to 1/4" ?


Note: The purpose of this exercise is to show that the force exerted on such a minibot by the plate assembly (and thus the force exerted by the minibot on the plate assembly) far exceeds 4 Newtons.

1. 0.586 ft.

2. 474 N