New Bumper Rule makes West Coast drive Illegal?

[Alan Ing]

I was reading through the robot rules and came across this.

Rule <R08> M says:

"The entire length of the BUMPER backing must be supported by the structure/frame of the ROBOT (i.e.) the backing material must not be in "free space" between or beyond attachment points) (see Figure 8-3)"

Wouldn't this make most cantalevered live axle drive robots Illegal? We have basically used our own version of the so called west coast drive chassis similar to 254, 60, 968 and others and have floated our bumpers on the outside of the wheels using brackets attached at various points at the end of the frame and between the wheels. If I understand this correctly, I now have to have a full frame member along the complete length of the wood bumper backing which would really defeat the purpose of having cantelevered wheels.

Anyone else come up with the same interpretation? Might as well just use the Kit Chassis.

[MrForbes]

We figured the same thing. I wonder how substantial of a support piece you need to have it considered as a frame rail?

[Tristan Lall]

That's correct. However, FIRST was not specific as to what constitutes the robot's frame or structure. You could conceivably lay a strip of some sort of lightweight material behind the bumper, and call it part of the frame.

Personally, and unofficially for the moment, I'd say that it's up to the team to decide what is and is not part of their structure or frame. (They designed it: they ought to have a better idea than anyone else.)

As a matter of policy, it would be foolishness to ask inspectors to second-guess the teams as to what is, or isn't part of the frame, based on subjective measures like how substantial the support piece is. It would just lead to different interpretations at different events, as teams appear with all manner of plywood, metal channel, plastic sheets, etc. as structural backings for their bumpers. Even if there was a universal standard applied everywhere, what would be the point—is that really a useful thing to have, among all of the other rules that teams need to deal with?

[Kevin Sevcik]

The intent here is to prevent bumpers from possibly snapping in half, etc. under repeated impacts. I submit that if the "frame member" behind the bumper isn't going to protect your robot in this event, then it probably doesn't count for this rule. Per usual, try to avoid lawyering to gain an advantage. If I were an inspector, I don't think I'd be passing any robot claiming at something classified as a "strip" was somehow a structural frame member.

[AdamHeard]

Originally Posted by Kevin Sevcik

The intent here is to prevent bumpers from possibly snapping in half, etc. under repeated impacts. I submit that if the "frame member" behind the bumper isn't going to protect your robot in this event, then it probably doesn't count for this rule. Per usual, try to avoid lawyering to gain an advantage. If I were an inspector, I don't think I'd be passing any robot claiming at something classified as a "strip" was somehow a structural frame member.

This is lame. Really lame. 973 has built a bumpered west coast from 06-08 without a single failure. Why punish us and force us to awkwardly reinforce bumpers that don't need it?

We'll do it, I'm just annoyed.

[Kevin Sevcik]

Originally Posted by AdamHeard

This is lame. Really lame. 973 has built a bumpered west coast from 06-08 without a single failure. Why punish us and force us to awkwardly reinforce bumpers that don't need it?

We'll do it, I'm just annoyed.

I'm not claiming to know if it's necessary. I'm just saying that if you're claiming that a 1" wide strip of 1/16" aluminum around the outside of your robot is part of you robot frame/structure.... Well it's silly. And anyone doing it KNOWS it's silly.

As for necessity... This is a completely new environment, and there's going to be a lot more collisions than in years past. Including your trailer colliding with your bumpers with all that nice leverage behind it. I think FIRST is erring on the side of safety and caution per usual.

[Tristan Lall]

Originally Posted by Kevin Sevcik

If I were an inspector, I don't think I'd be passing any robot claiming at something classified as a "strip" was somehow a structural frame member.

That's exactly why it's hazy, though: let's say I use a 0.19 in thick × 1.00 in tall strip of polycarbonate as the backing material, with several screws into the bumper plywood. Will it be strong enough across a 20 in span? My gut feeling is that under most gameplay, it would not break, or even deflect appreciably. But what if the team used (brittle) acrylic instead? Or if they turned the strip 90° to buttress the plywood? I don't think it makes much sense for inspectors to be trying to make that particular determination—it will just be their best guess, and it's ripe for debate that serves no useful purpose.

Originally Posted by Kevin Sevcik

The intent here is to prevent bumpers from possibly snapping in half, etc. under repeated impacts. I submit that if the "frame member" behind the bumper isn't going to protect your robot in this event, then it probably doesn't count for this rule.

If your bumper plywood ever breaks, even with no backing (and especially with the aluminum angle depicted in Fig. 8-1), you're probably doing it wrong. And if you use high-quality hardwood marine plywood (as opposed to the regular softwood stuff), you could probably span over 30 in without any backing, and suffer no ill effects.

I just don't think it's valuable to add a specification here, given that it's one more thing for teams to worry about, one more thing that needs to be checked (consistently) at inspection, and is probably superfluous anyway, given moderately-well-constructed bumpers.

[Kevin Sevcik]

Originally Posted by Tristan Lall

That's exactly why it's hazy, though: let's say I use a 0.19 in thick × 1.00 in tall strip of polycarbonate as the backing material, with several screws into the bumper plywood. Will it be strong enough across a 20 in span? My gut feeling is that under most gameplay, it would not break, or even deflect appreciably. But what if the team used (brittle) acrylic instead? Or if they turned the strip 90° to buttress the plywood? I don't think it makes much sense for inspectors to be trying to make that particular determination—it will just be their best guess, and it's ripe for debate that serves no useful purpose.
If your bumper plywood ever breaks, even with no backing (and especially with the aluminum angle depicted in Fig. 8-1), you're probably doing it wrong. And if you use high-quality hardwood marine plywood (as opposed to the regular softwood stuff), you could probably span over 30 in without any backing, and suffer no ill effects.

I just don't think it's valuable to add a specification here, given that it's one more thing for teams to worry about, one more thing that needs to be checked (consistently) at inspection, and is probably superfluous anyway, given moderately-well-constructed bumpers.

I submit the following two photos into the evidence:
http://www.chiefdelphi.com/media/photos/31159
http://www.chiefdelphi.com/media/photos/23644

A 38" span of bumper isn't likely to survive many of those.

As you say, high quality marine hardwood bumpers with reasonable length spans would be more than fine, and we both know that. But the bumper specifications don't call for high quality plywood, nor maximum span lengths. Instead, they call for the bumpers to be fully supported along their length, which serves a similar function of protecting rookies from underestimating impacts and ending up with snapped bumpers. The rule is, after all, ultimately there to protect less experienced teams from themselves. As are a large number of the rules. Which is why I don't spend time railing against having to electrically isolate everything from the robot frame, the various fusing rules, and other things that I know enough about to do differently and better.

[Tristan Lall]

Originally Posted by Kevin Sevcik

I submit the following two photos into the evidence:
http://www.chiefdelphi.com/media/photos/31159
http://www.chiefdelphi.com/media/photos/23644

A 38" span of bumper isn't likely to survive many of those.

As you say, high quality marine hardwood bumpers with reasonable length spans would be more than fine, and we both know that. But the bumper specifications don't call for high quality plywood, nor maximum span lengths. Instead, they call for the bumpers to be fully supported along their length, which serves a similar function of protecting rookies from underestimating impacts and ending up with snapped bumpers. The rule is, after all, ultimately there to protect less experienced teams from themselves. As are a large number of the rules. Which is why I don't spend time railing against having to electrically isolate everything from the robot frame, the various fusing rules, and other things that I know enough about to do differently and better.

If you compare the failure modes of those electrical rules (burning wires, short circuits, etc.) to the costs of imposing them (it's easy to spot an out-of-place breaker, or test for a grounded chassis), it's a pretty reasonable proposition to mandate those things.

On the other hand, even assuming a violent collision, what's the failure mode of a snapped bumper—plywood hanging limply from cloth? That's hardly in the same league as an electrical fire. (And the bumper repair might even be straightfoward: cut the ends smooth to make two bumpers, and re-brace it somehow.)

So, while you could attempt to define what's a strong frame and what's sufficient structure, and then assess every robot by those standards, I don't think you're going to get much value for the effort (because most bumpers don't break, and the ones that do are probably not a big deal). I also think that you run the risk of inconsistent officiation ("strong enough" is probably a subjective measure) and dispute (because teams will probably believe that their solution is sufficient).

But if a guideline (rather than a rule) captures the intent to caution teams against weak bumpers, why not just opt for that? What's the point of trying to define a frame, and then having to enforce that ruling? And what's the harm in putting the (rather light) burden of building a robust bumper squarely on the teams? I hope that when this gets asked to the Q&A, that they take the opportunity to issue that caution, and just let teams pay harmless lip service to the rule (at their own peril).

After all, although I'm a fan of specificity in rule-making, I don't think that there's a significant benefit for FIRST to interpret this in anything but the most permissive manner.

[dtengineering]

In lieu of a GDC approved definition of "supported", I agree with Tristan that it is not up to the inspectors to make on the site decisions as to what does, or does not, constitute support. If a frame piece provides even the most marginal degree of support... then it is within the rules as they are written now (subject, I am sure, to a Q and A response in the very near future.)

One of the difficult things about being an inspector is to read the rules as they are written... not as you think they should be written. In the spirit of natural justice and fair play that which is not forbidden by the rules must be permitted and that which is required by the rules must be required... even if as an inspector you think the rule is nuts and/or inadequate.

But what would an FRC game be without a few bumper rules to keep things exciting.

Jason

[GaryVoshol]

There's also changes in the bumper attachment rules.

[Paul Copioli]

I noticed the changes in the bumper rules, but the picture didn't change. It looks like it specifically allows other forms of attachment other than what is pictured. Good, now I do not have to "discuss" this with Tristan and Ed Sparks at Championship this year.

[GaryVoshol]

This year's rule: "nut and bolt fasteners are recommended" - Last year that was a MUST.

Additional fastener parts can be attached to the bumper itself this year - the rules specifically mention brackets.

There is in increase to 18 pounds for max bumper weight (was 15).

The pool noodles can be beveled at the ends of the bumper pieces to make a fit around the outside corners of the robot.

[Tristan Lall]

Originally Posted by Paul Copioli

I noticed the changes in the bumper rules, but the picture didn't change. It looks like it specifically allows other forms of attachment other than what is pictured. Good, now I do not have to "discuss" this with Tristan and Ed Sparks at Championship this year.

Believe me, I'm just as relieved as you are that they greatly improved the bumper rule for this year.

[Ed Sparks]

Originally Posted by Paul Copioli

..... Good, now I do not have to "discuss" this with Tristan and Ed Sparks at Championship this year.

I'm sure we'll find other things to "discuss" ......