Team Update #18

[Matt Krass]

Originally Posted by Ether

You're forgetting about interrupts. If the switch is connected to an IO port which can be configured to generate an interrupt then there is no sample rate involved.

I considered interrupts, but even interrupts have a minimum latch time on most systems I've used, it may be on the order of microseconds, but it's there, and the point is, it was omitted, so we don't know if it's a fast interrupt or a slow poller, or a shift register that only gets clock in every 300ms, etc. The system must have some time component to it, which was my point. To assume it can behave instantaneously was erroneous, especially since no time component was specified. If you told me it took 100us for the interrupt to latch and take hold, then I'd say it's a negligible amount of time, but at least I'd have concrete evidence to work with.

EDIT: As also pointed out before, even interrupts cannot guarantee a clean signal without some decent debounce time on the signal, which still brings you back to a time component. Which we weren't given, and FIRST may not (yet) have.

Matt

[AdamHeard]

Originally Posted by Ether

16.8 watts max power output w/o gearhead

= 12.4 ft-lb/sec

x2 motors = 24.8 ft-lb/sec max power output

2.3 lb estimated robot weight

(24.8 ft-lb/sec) / (2.3 lb) = 10.8 ft/sec (neglecting friction)

Beat me to it. We ran the math in metric units, but it's the same nonetheless.

We then assumed instant acceleration, and used distance = velocity x time to solve for an approximate "perfect" time.

[Karthik]

Originally Posted by Paul Copioli

I guess I just found the entrance music for Karthik at Waterloo! I wonder how long it will take before someone shuts down the webcast at Waterloo?

It'd get shut down a lot faster if we went with my own personal choice of entrance music, Shots by LMFAO.

[Ether]

Originally Posted by Matt Krass

even interrupts cannot guarantee a clean signal without some decent debounce time on the signal, which still brings you back to a time component.

For reference I'm linking to a post made a couple hours earlier here in this thread.

[Matt Krass]

Originally Posted by Ether

For reference I'm linking to a post made a couple hours earlier here in this thread.

I recall your post, though I don't think my post was unqualified, merely incomplete. I should have mentioned interrupts in my original post, I was thinking more conceptually (against the notion of instantaneous detection taking zero time) and didn't see much relevance at the time.

Matt

[Ether]

Originally Posted by Ether

16.8 watts max power output w/o gearhead

= 12.4 ft-lb/sec

x2 motors = 24.8 ft-lb/sec max power output

2.3 lb estimated robot weight

(24.8 ft-lb/sec) / (2.3 lb) = 10.8 ft/sec (neglecting friction)

Continuing the calculation:

(16.8 watts @ max power)/(0.0475 Nm torque @ max power) = 353.7 radians/sec shaft speed @ max power

(10.8 ft/sec)/(353.7 radians/sec) = 0.03045 ft shaft radius = 0.73 inch shaft diameter (for direct drive)

Probably want to reduce that diameter by 15% or so for losses and margin

someone please check my math

[Ether]

Originally Posted by Matt Krass

I recall your post, though I don't think my post was unqualified

Poor word choice on my part; the word has multiple meanings which could be taken the wrong way. I intended definition #2:

un·qual·i·fied (n-kwl-fd)
adj.
2. Not modified by conditions or reservations; absolute

I referenced the previous post mostly for the remarks about debouncing interrupt-serviced switches.

[gblake]

Folks,

If I were to carefully read this entire thread, instead of only quickly scanning the last 100 or so posts, what useful information/conclusions would I acquire?

I have gathered that the tower targets are close to FUBAR status and many mini-bot designs are consequently pseudo-randomly finding themselves up the proverbial creek. It that all we have here?

My goodness, does it take 200+ posts of grouching at each other, and at FIRST, to convey that clearly?

I know everyone is tired, but this could be a fun topic.

Who has tested some work-arounds and can reports their results to teams that need to modify their mini-bots???? Is slower better? Would putting a broad squishy nose on a mini-bot help? Would leaving the mini-bot motors energized an extra few fractions of a second after target-contact help? Anything else?

Blake

[Matt Krass]

Originally Posted by gblake

Folks,

If I were to carefully read this entire thread, instead of only quickly scanning the last 100 or so posts, what useful information/conclusions would I acquire?

I have gathered that the tower targets are close to FUBAR status and many mini-bot designs are consequently pseudo-randomly finding themselves up the proverbial creek. It that all we have here?

My goodness, does it take 200+ posts of grouching at each other, and at FIRST, to convey that clearly?

I know everyone is tired, but this could be a fun topic.

Who has tested some work-arounds and can reports their results to teams that need to modify their mini-bots???? Is slower better? Would putting a broad squishy nose on a mini-bot help? Would leaving the mini-bot motors energized an extra few fractions of a second after target-contact help? Anything else?

Blake

I don't believe this has been 200 posts of people grouching at each other, instead it's been a very lively, interesting debate of the technical details of the situation. And since this is just coming out now, nobody has publishable results yet, just a lot of hypotheses on what may or may not be occurring, and how to possibly improve the towers or minibots.

Nobody has a definitely answer for you, other than expect to have to work at this some more.

Personally I think this group has done a fantastic job dissecting the situation and working to figure out the information FIRST has woefully failed to provide once again, but give them a chance, this isn't old news yet.

I thought it was a fun topic, though I wish I could contribute more effectively to it.

EDIT: I wanted to answer to this as well.

Originally Posted by Ether

Poor word choice on my part; the word has multiple meanings which could be taken the wrong way. I intended definition #2:



I referenced the previous post mostly for the remarks about debouncing interrupt-serviced switches.

Sorry to misunderstand you, your intended wording makes a lot more sense!

Matt

[jspatz1]

Originally Posted by gblake

My goodness, does it take 200+ posts of grouching at each other, and at FIRST, to convey that clearly?

There is a large number of posts because it is a topic of interest and importance to everyone, not because folks are looking to grouch at each other. The volume, to which you have contributed, is an accumulation of many individuals, of which you are now one. It is the result of many people sharing their 2 cents, like you did. Nearly every post has dealt with the topic, the only one I've read that really grouched at others is yours.

[boomergeek]

Originally Posted by Ether

16.8 watts max power output w/o gearhead

= 12.4 ft-lb/sec

x2 motors = 24.8 ft-lb/sec max power output

2.3 lb estimated robot weight

(24.8 ft-lb/sec) / (2.3 lb) = 10.8 ft/sec (neglecting friction)

Here's a different way to get at approximately the same point.

At kick off, the video of the FIRST built minibot showed about 1.3 ft/sec for a robot that looked to be more than 5 lbs and using a pair of Tetrix motors and gearboxes.

Its not hard to cypher that a 2 lb robot can be 2.5 times faster than a 5 lb robot. By the same token, if you throw away a near 50% inefficient transmission, you should be able to obtain another doubling of speed. That gets you in the 7-8 ft/sec range. Add additional lowering of friction in the attachment to the pole mechanism and you approach the magically 10.8 ft/sec. Add topping off the battery to get to a better motor curve is gravy on top.

[Chris Hibner]

Just as a note:

We had 3 minibots up the pole today at the Detroit district, all three triggered just fine. I'll keep my eye on it this weekend.

[Lil' Lavery]

Originally Posted by Chris Hibner

Just as a note:

We had 3 minibots up the pole today at the Detroit district, all three triggered just fine. I'll keep my eye on it this weekend.

Were any additional modifications made to the towers since the last time this field was used? What was the last event this field was used at?

[sanddrag]

Originally Posted by Ether

16.8 watts max power output w/o gearhead

= 12.4 ft-lb/sec

x2 motors = 24.8 ft-lb/sec max power output

2.3 lb estimated robot weight

(24.8 ft-lb/sec) / (2.3 lb) = 10.8 ft/sec (neglecting friction)

Thank you Ether! This is the first post I've seen all season to do a proper (yet quite simple) calculation of how fast a minibot of given weight can get up the pole. I've been meaning to do this with my students all season long, but was too busy with the arm and other things. I think now is the time to revisit it.

Trial and error is one thing. Knowing what you're shooting for, because you did some calculation first, is quite a different (and much better) thing.

[GaryVoshol]

Originally Posted by Lil' Lavery

Were any additional modifications made to the towers since the last time this field was used? What was the last event this field was used at?

The field from Waterford went to Detroit.