For those who are skeptical about propellers - Team 2526

[AdamHeard]

I know others have said it, but I'm concerned about the safety here.

If those blades break, which could happen, they will be sent flying at extremely high speeds, with lots of energy.

I'm usually the guy telling people worried about safety to shut up, it's fine. But this is one case where it seriously scares me. You could send shrapnel at extremely high speeds into the audience, which could kill somebody if hit right. It's unlikely, but too high a risk for my liking.

[Rick Wagner]

With two equal weight robots in a pushing contest on this regolith surface, a pound or two of thrust will make a big difference. Done properly (a good safety cage), a propeller in addition to all-wheel drive makes a lot of sense. The robot that pushes another robot into a hostile corner can decide the game.

[DarkFlame145]

I know they will work, I'm just a little frightened by them on a robot =P

[AdamHeard]

Originally Posted by Rick Wagner

With two equal weight robots in a pushing contest on this regolith surface, a pound or two of thrust will make a big difference. Done properly (a good safety cage), a popeller in addition to all-wheel drive makes a lot of sense. The robot that pushes another robot into a hostile corner can decide the game.

I totally agree with that.

I've seen similar sized objects with similar energy and similar speeds break up; it will take a serious, serious object to hold it in. A cage won't suffice, plates of metal or polycarb will be needed, and not thin stuff at that.

[Rick TYler]

Originally Posted by AdamHeard

I've seen similar sized objects with similar energy and similar speeds break up; it will take a serious, serious object to hold it in. A cage won't suffice, plates of metal or polycarb will be needed, and not thin stuff at that.

The propeller blades are pretty low mass. Put a curved Lexan duct around the fan and call it good. The nice thing is that you know the blades are going to fly straight out -- it's not like they're made of C4. Wrapping the perimeter with a 6-inch wide Lexan duct should do the trick. Even if the blade hits the Lexan and bounces back to penetrate the finger-grid, there certainly won't be enough energy left to kill someone standing around the field.

How often do RC plane propellers spontaneously explode? This is a really mature technology -- I would be surprised if they weren't overbuilt for safety. How about replacing those 2-blade props with 3- or 4-blade units? For the same power you can have shorter blades with much lower tip velocities. At least, that's how it works on real airplanes.

[AdamHeard]

Originally Posted by Rick TYler

The propeller blades are pretty low mass. Put a curved Lexan duct around the fan and call it good. The nice thing is that you know the blades are going to fly straight out -- it's not like they're made of C4. Wrapping the perimeter with a 6-inch wide Lexan duct should do the trick. Even if the blade hits the Lexan and bounces back to penetrate the finger-grid, there certainly won't be enough energy left to kill someone standing around the field.

How often do RC plane propellers spontaneously explode? This is a really mature technology -- I would be surprised if they weren't overbuilt for safety. How about replacing those 2-blade props with 3- or 4-blade units? For the same power you can have shorter blades with much lower tip velocities. At least, that's how it works on real airplanes.

RC airplanes aren't hit by other things very often.

[Rick TYler]

Originally Posted by AdamHeard

RC airplanes aren't hit by other things very often.

A properly mounted propeller on an FRC bot shouldn't ever get hit at all. If they are mounted out in the open like the prototype in the first item in this thread -- that's unsafe and negligent. I was assuming (I know, I know) that it would be properly protected from ball-strikes and other structural indignities.

[EricH]

Originally Posted by Rick TYler

A properly mounted propeller on an FRC bot shouldn't ever get hit at all. If they are mounted out in the open like the prototype in the first item in this thread -- that's unsafe and negligent. I was assuming (I know, I know) that it would be properly protected from ball-strikes and other structural indignities.

The R/C airplanes don't get hit either, at all (other than maybe on a hard landing or a midair collision). It's not the props getting hit that we're worried about, it's the robot getting hit and jarring around.

[zrop]

I'm pretty sure that if the props can resist any vibrations that they and the motors generate, they should hold in a robot collision. I mean yeah, it will be a significant change in acceleration, but that's still no reason for the props to spontaneously explode.

Oh and update: We've decided to add power to the back 2 wheels powered by a CIM motor. So basically, that'll add some extra acceleration that we lack, and the props should handle well with the turning. ;D

[Molten]

Originally Posted by zrop

I'm pretty sure that if the props can resist any vibrations that they and the motors generate, they should hold in a robot collision.

Thats two completely different forms of impact. First is vibration. That usually slowly tears something up. The second is turbulence and blunt force. Basically your comparison is as reasonable as stating, 'This doesn't break when I shake it so it shouldn't when I take a baseball bat to it.' Not really a valid statement at all.

A personal request: Once you build your robot so it can drive, purposefully run it into a couple of walls at full speed. See what happens. If it doesn't do any damage, you might be ok in competition, but you still might have trouble.

[Daniel_LaFleur]

Originally Posted by zrop

I'm pretty sure that if the props can resist any vibrations that they and the motors generate, they should hold in a robot collision. I mean yeah, it will be a significant change in acceleration, but that's still no reason for the props to spontaneously explode.

Oh and update: We've decided to add power to the back 2 wheels powered by a CIM motor. So basically, that'll add some extra acceleration that we lack, and the props should handle well with the turning. ;D

Why is it that I shudder every time I hear "I'm pretty sure" when they are talking about engineering?

I'm not too concerned about the vibrations of the motors. I'm far more concerned with the shock of impact from another robot.

**Gets up on soapbox**
Why not be "sure" rather than "pretty sure" by running the bending calculations. I am "pretty sure" that your robot will see a better than 10G impact with another robot or the wall (this will be transmitted throughout the robot and anything cantilevered will feel the full force across its beam.). Calculate that force across the beam of your prop and use the weight of your prop divided by the number of blades. Personally I'd use 20G as a safety margin. If the material does not bend further than where it will permanantly deform then I'd call it fairly safe for the forces it will see ... then you need to protect it from outside forces like other robots and orbit balls.

Props can be very effective. Lets just make sure they're safe as well.
**Gets off soapbox**

[writchie]

Originally Posted by Rick TYler

The propeller blades are pretty low mass. Put a curved Lexan duct around the fan and call it good. The nice thing is that you know the blades are going to fly straight out -- it's not like they're made of C4. Wrapping the perimeter with a 6-inch wide Lexan duct should do the trick. Even if the blade hits the Lexan and bounces back to penetrate the finger-grid, there certainly won't be enough energy left to kill someone standing around the field.

How often do RC plane propellers spontaneously explode? This is a really mature technology -- I would be surprised if they weren't overbuilt for safety. How about replacing those 2-blade props with 3- or 4-blade units? For the same power you can have shorter blades with much lower tip velocities. At least, that's how it works on real airplanes.

It's called a propeller explosion because that what it looks like when contained.

A 12 inch prop rotating at 9000 RPM has a wing-tip velocity over 470 RPM. A 22 inch at that RPM is traveling at over 3/4 the speed of sound.

Propellers do not spontaneous explode when standing still, nor undamaged and moving within their design limits in the applications for which they were designed, tested, and proven. They explode when forces beyond their material strength limits essentially rip them apart. When that occurs two things happen; the failed piece (maybe an inch of tip) no longer has any centripetal force and begins moving in a straight line (usual outward and a bit forward) while aerodynamic forces still act on it, often causing it to tumble; At the same time, the remains are now horribly unbalanced and the remaining rotational energy goes into ripping off the other tip and maybe some other parts - hence the term explosion.

Back to the tip - it will usually bounce off something but since it will be at an angle its direction will change. This happens maybe 200 microseconds after tip separation. If the tip is wood it may splinter into dust. If its nylon it may elastically deform and retain a good deal of its energy after the bounce. Murphy then takes over and the path of the tip may cause it to pass through any protective screening if the opening are wide enough. Or it may bounce inside, hitting the rotating remains. You are told to stay out of the propeller arc because in the open air setting of an R/C field, that is where the prop is likely to go (except if the prop nut loosen off with a tractor prop and the prop travels forward until it hits the face looking at the front of the propeller (i've found props 100 feet away).

On a FIRST field, the whole chain of events could be triggered by the robot hitting the wall, inducing abnormal loads. The point is that if Murphy has his way the collision may be on the side of the field where people like referees, MC's, and cameramen are standing. Tragedy does not require that someone be killed. A nylon tip is plenty sharp, especially when its travelling at 200 fps.

Yes, R/C propellers are a mature technology for the conditions and best practices in which it has been used. That is not what we have here, especially as configurations stray from the conventional. Exactly how are you going to predict the vibrations and resonances of a pimped up belt drive for the prop?

When you move outside the envelope, the technology is no longer mature. And prop failures in R/C do occur surprisingly often where the envelope is being pushed.

I have no doubt that some thruster designs will be deemed insufficient to be operated within a reasonable margin of safety for the conditions present on the field (5 other 150 lb robots sliding under marginal control all over the field.) I'm hoping that some teams will have confronted and surmounted the challenges presented and be allowed to demonstrate their engineering on the field.

I do like this added dimension to the competition. But to paraphrase a movie quote "The lack of humility before nature that's being displayed here, uh ... staggers me."

[writchie]

I think the following comments on propeller safety from a manufacturer are worth the lengthy quote- From APC.

APC Propeller Safety Concerns

All propellers are inherently dangerous. Model airplane propellers are especially dangerous. Model airplane propellers used in high performance racing are extremely dangerous. Model airplane engines designed and modified to achieve maximum operating capabilities create unpredictable and potentially severe loads, leading to various forms of potential propeller failure. Ignoring reasonable safeguards may likely be catastrophic. This concern is the motivation for the following discussion.

Warnings included with propellers are intended to protect consumers. They also protect manufactures against claims resulting from misuse of the product. Most products with potential for causing injury contain ample warnings about misuse. Some advertisements for products now contain warnings, even before the product is sold! There is a strong proliferation of warnings in most products having potential for creating injury or damage. This inundation of warnings may cause consumers to become inured to product warnings.

The warnings about propeller use must be taken seriously, especially for racing applications. It is very risky to assume that a racing propeller blade will not fail, especially when used with state-of-the-art racing engines. Yet, nevertheless, occasionally model aircraft operators are observed standing in the plane of propeller rotation of high performance racing engines running at full power. This is very frightening. The following information reinforces the assertion that dangers of misuse are very real.

Ideally, a product can be designed with credible knowledge of the environment (loads acting on the product) and capabilities of the product to withstand that environment (not fail). There is nothing ideal about designing a model airplane propeller because some major components of propeller loads are very uncertain. The principle load components acting on a propeller are:

Centrifugal (from circular motion causing radial load)
Thrust/drag (from lift and drag acting on blade sections)
Torsional acceleration ( from engine combustion and/or pre-ignition)
Vibration (from resonant frequencies or forced excitation)
Another potential source of loading is aero elastic tip flutter. This may be caused by self exciting aerodynamic loads at a resonant frequency.

These loads are discussed next in order.

Centrifugal loads are very predictable, given rotational speed and mass density distribution of a blade. Their contribution to total stress is relatively small.

Thrust/drag loads are somewhat uncertain due to complexities of aerodynamic environments. The relative axial speed at the prop (at any radial station) is aircraft speed plus the amount the air in front of the blade is accelerated by the mechanics creating thrust. The latter may be approximated using first order classical theory. Much empirical lift/drag data (from wind tunnel tests) exists to quantify lift/drag loads, once relative velocity and angle of attack distributions are established.

Torsional acceleration loads are generally not known. Analytical estimating technique used by Landing Products to quantify torsional acceleration loads suggests that they can become dominant when pre-ignition or detonation occurs. These analytical observations are supported by test experience with very high performance engines running at elevated temperatures. The latter causes a high torsional load (about the engine shaft) which creates high bending stresses, adding to those from centrifugal force and lift/drag effects. These torsional acceleration loads depend on unique conditions for specific engines. Engines "hopped up" for racing appear to be especially prone to create high torsional loads when lean mixtures lead to high cylinder temperatures and pre-ignition/detonation.

Vibration causes additional loads from cyclic motions. These motions occur when resonant frequencies are excited or when cyclic load variations exist on the blade. The magnitude of these variations depends on how close the driving frequency is to the resonant frequency and the level of damping in the propeller material. Engine combustion frequency is an obvious excitation. Obstructions in front of or behind the blade can cause cyclic variations in thrust load. Once a blade starts to flutter, those motions alter the flow, causing variations in loading. High performance engines have caused propeller tips to break, presumably due to fatigue failure from vibration.

Aero-elastic flutter is speculated to be a dominant mechanism causing rapid fatigue failure near a tip when insufficient or destabilizing tip stiffness exists. The interaction between variable loading and deflection induces a high frequency vibration with unpredictable magnitude.

Efficient propeller design practice utilizes analytical/computational models to predict propeller performance and stresses. However, the uncertainty in impressed and inertial loading from complex phenomena requires testing to assure safe performance. Unfortunately, it is not possible to assure testing that convincingly replicates worst case conditions. The large combinations of engines, fuels, temperature, humidity, propeller selection, aircraft performance and pilot practices creates an endless variety of conditions. If the origins of severe loads were well understood, quantified, and measurable, structured testing might be feasible that focuses on worst case stack up of adverse conditions. However, since the origins of severe loads are really not well understood, it is essential to provide sufficient margins in material properties and design to assure safe performance. Propellers that are used in fairly routine and widespread applications (sport and pattern) lend themselves reasonably well to test procedures that provide reasonable confidence. In time, a sufficient data base develops that can be used to empirically quantify performance and "anchor" or "tune" assumptions used in analytical models.

However, propellers that are used for increasingly extreme performance applications do not benefit from the large empirical data base sport and pattern propellers enjoy. Assumptions and design practices developed for current generations of engines may not be valid for emerging engines whose technologies continue to push engine performance to greater extremes. Consequently, propellers that are used in applications where performance is already relatively high (and expanding) must be used with great caution.

An adverse cascading effect occurs when propellers are permitted to absorb moisture in high humidity environments. Composite strength, stiffness and fatigue endurance all reduce with increased moisture content. Reduction in stiffness typically causes resonant frequencies to move toward the driving frequency (increasing torsional loads) and, the reduction in strength reduces fatigue endurance. Composite propellers should be kept dry.

In summary, please abide by the safety practices recommended by propeller manufactures. This is especially important for high performance propellers. Assume that propellers can fail at any time, especially during full power adjustments on the ground. Never stand in or expose others to the plane of the propeller arc.

[Racer26]

In FIRST, safety has always been #1, and its imperitive that teams take it seriously. As several others have mentioned, and linked to in this thread, accidents DO happen, and CAN cause serious injury. For the 2008 season, 1075 built a robotic forklift cart, driven by an IFI system from 2004. When we built it, we knew for sure that the safety advisors would be all over us about it, since driving a vehicle through the crowded pits has the potential to be quite dangerous, never mind that the vehicle itself has its own concerns. It was built with a very long list of safety features designed to prevent accidental operation, as well as to reduce possibility of serious injury as a result of its use. It was also built to reduce RSI's in our own team members, associated with lifting the robot, and working on it at the wrong height.

Its features are as follows:

on the old white joysticks:
trigger: enables the lift to be operated with aux1 and aux2
trigger + thumb button: enables the X and Y axes of the joystick to drive the cart

other lockouts:

seat is spring loaded, with a switch to prevent operation of the cart without a driver IN the drivers seat

front bumper on the end of the deck has a switch in it as well to stop the cart from continuing to drive into something (people, walls, whatever)

the lift has holes crossdrilled in the legs for safety pins, to ensure the deck doesnt fall.

the motor driving the deck up and down has a shroud over the chain, to protect any stray fingers, etc

it has a rotating light off our 2003 robot, to warn the area of its presence and ability to move unexpectedly.

for 2009 I believe we are planning to add an automotive horn, for an audible warning.

AND, its ALWAYS operated with a crew of flagpeople to clear the path and raise awareness to its presence.

As for benefits it provides to us, it was designed to be able to reach over the edge of the field, however, none of the events we've taken it to so far have allowed us to do so. This was to reduce back strain associated with lifting it on and off the field.

It allows us to raise up the robot to a proper height to work on in the pits without back strain and similar issues.

It also has a number of features to aid in testing and debugging the robot at competition.

This is the kind of attention to detail required in FIRST when it comes to safety. I hate to use my own team as an example, because it sounds like i'm bragging or gloating, but when there's a high risk for danger (ie. props breaking up at thousands of rpm, with the potential for sending shrapnel flying at high velocity), safety should be paramount.

[KGood]

Originally Posted by Akash Rastogi

I'm taking Eric's word on this one.

I highly doubt that the inspectors will give a crap ( or too much) about the safety of the precious orbit balls.

I say that if a finger or even a hand can pass through the cage, then it is not safe.

+$0.02

Not just breaking and having shrapnel in the field, but if something flies into the stands it's going to be trouble.