Hovercraft

[smartkid]

When does the Q&A open anyway?

[T3_1565]

Originally Posted by Tetraman

And to pose the question as to if we can suction ourselves to the floor is another good question.

The rules say somewhere (now with edit) that you cannot attach yourself to the arena surface.


<G29>
Arena Interaction –ROBOTS may push or react against any elements of the ARENA, provided there is no damage or disruption of the ARENA elements. With the exception of a ROBOT towing a TRAILER, ROBOTS may not grab, grasp, grapple, or attach to any ARENA structure

[Bongle]

To finalize my discussion with myself about fans to augment your own acceleration, I found one designed for a maximum of 300 watts, which is in the range that a FRC robot can legally put out. It puts out 4 newtons while spinning at some mind-boggling RPMs. It would sound really cool spooling up.

So you'd have a drive motor running continuously at near-maximum output for 4 newtons of forward force. Maybe I'm not looking in the right places (after all, R/C fans are designed for applications where they are moving through air quickly), but that seems like it'd be too little force for a hovercraft to be effective.

You could build very light if you wanted maximum acceleration, but you're still hauling a big trailer behind you and you'd lose pushing matches because your absolute pushing power would be lower (you'd lose your wheel traction in proportion to your weight loss).

A maximum-weight robot would get a 15% acceleration boost by using this little fan, assuming it would be a legal thing to do, you could rig a motor and gearbox to the fan, and the constant 300 watt drain wouldn't greatly impair other robot actions.

Here is the fan: http://www.modelflight.com.au/rc_mod..._micro_fan.htm

[Beakerone]

Definitely an interesting idea, but i was wondering how you were going to get past (R18) which defines the height that the trailer must be attached to the robot - 2 and 13/16ths to the center of the Trailer Hitch from the floor., doesn't sound like it can be "floating" at different heights... which I think would occur, also they mention that it has to be rigidly attached to a fixed locations.

[drkiraco]

to keep going:
Thrust = mass flow * velocity
Power = (1/2) mass flow * velocity^2

For a given thrust, we could pick any combination of mass flow and velocity, but to minimize energy used, we want greater mass flow and less velocity.

max power for a single FIRST motor is 480 watts (12V * 40A). call it 500W for round numbers. So for 50 N you could do, for example, 2.5 kg/s air at 20 m/s. Now, to put this in perspective, 20 m/s = 44 mph. Seems a little high to me, but maybe not out of the question. Probably you wouldn't get anything faster than this. Now put 2.5 kg/s of air in perspective. That's 5.5 lb/s. Since a cubic foot of air has a mass of ~0.075 lb at standard temp and pressure, we need 73 ft^3/s. That's 4406 ft^3/min (scfm). That's a rather large number. Maybe not totally out of the question though.

So, unless I messed up the math, I'd guess that 50N of thrust could be achievable without violating the laws of physics or any FIRST rules, but I don't think it would be cheap or easy...

Originally Posted by Bongle

To continue this train of thought and get an idea of what would be possible. Note that neither of these would be legal:

A $2600USD model aircraft jet engine puts out 59N: http://www.trimair.com.au/index2.html. However, it is petrol powered and obviously illegal.

A $740 set of ducted fan components can put out 13-15lbs of thrust (57.9N), but uses a minimum of 3200 watts, which is 6.6 times how much power a FIRST robot can use.

[Bongle]

Originally Posted by drkiraco

to keep going:
Thrust = mass flow * velocity
Power = (1/2) mass flow * velocity^2

For a given thrust, we could pick any combination of mass flow and velocity, but to minimize energy used, we want greater mass flow and less velocity.

max power for a single FIRST motor is 480 watts (12V * 40A). call it 500W for round numbers. So for 50 N you could do, for example, 2.5 kg/s air at 20 m/s. Now, to put this in perspective, 20 m/s = 44 mph. Seems a little high to me, but maybe not out of the question. Probably you wouldn't get anything faster than this. Now put 2.5 kg/s of air in perspective. That's 5.5 lb/s. Since a cubic foot of air has a mass of ~0.075 lb at standard temp and pressure, we need 73 ft^3/s. That's 4406 ft^3/min (scfm). That's a rather large number. Maybe not totally out of the question though.

So, unless I messed up the math, I'd guess that 50N of thrust could be achievable without violating the laws of physics or any FIRST rules, but I don't think it would be cheap or easy...

I did math myself starting with your equations and got the same answer, so you're probably correct.

Now, how big would the swept area on this hypothetical fan be? You've got a column of air 18.5 meters long (for my calcs) long passing through through every second, that contains 2.7kg of air, which is 2.11 cubic meters.

V = pi * r * r * h
2.11 = pi * r * r * 18.5
sqrt(2.11 / (18.5 * pi)) = r
0.19 = r = 19cm radius fan, or 7.5 inches. That's a pretty big fan, but not unusably big. This could be do-able, but the speed out the back and the CFM rating seems pretty insane. I searched for 4000CFM fans, and came up with stuff sold as "whole house fans", and I doubt that those would put out the kind of velocities needed.

Is this a correct calculation?

[mray1031]

Don't want to get into all the math, I don't really remember it, but from fooling around with a human-carrying hovercraft a few years ago I have a couple of things for you to consider:

First: You have to be careful with the center of mass when using a hovercraft- any manipulators carrying over the curtains will tip it. Tipping makes the air curtain hit the ground (a rule violation) and will make your 'craft drift because of the uneven airflow.

And second: Yes a hovercraft will float the 120 lbs of robot. It will do so for awhile too. I remember using two car fans and the same 12 volt batteries we use now to carry myself as well as the weight of the actual craft and batteries (probably a little over 120 back in the day)

[smartkid]

O.k. Being less far-fetched, from this I've gotten that using fan(s) to boost a drive trains output is possible but to cut out the drive completely is just stupid (or not prudent).

The hovercraft would have been sweet though. Oh well

[drkiraco]

I did the math a little differently and got 7.2 inches for the radius. Probably a round-off difference between you and me. If anyone was seriously considering this, it would probably be better to make the fan bigger, thus it could run slower.

Originally Posted by Bongle

I did math myself starting with your equations and got the same answer, so you're probably correct.

Now, how big would the swept area on this hypothetical fan be? You've got a column of air 18.5 meters long (for my calcs) long passing through through every second, that contains 2.7kg of air, which is 2.11 cubic meters.

V = pi * r * r * h
2.11 = pi * r * r * 18.5
sqrt(2.11 / (18.5 * pi)) = r
0.19 = r = 19cm radius fan, or 7.5 inches. That's a pretty big fan, but not unusably big. This could be do-able, but the speed out the back and the CFM rating seems pretty insane. I searched for 4000CFM fans, and came up with stuff sold as "whole house fans", and I doubt that those would put out the kind of velocities needed.

Is this a correct calculation?

[Bongle]

Originally Posted by drkiraco

I did the math a little differently and got 7.2 inches for the radius. Probably a round-off difference between you and me. If anyone was seriously considering this, it would probably be better to make the fan bigger, thus it could run slower.

I was thinking about this some more, and I realized the original spec of 50N is a bit ridiculous. Sure it would make your robot fast, but once your fan spooled up, you wouldn't be able to stop (without cutting and reversing the fan, which would take a long time). 5-10N as a booster would be more reasonable. I'd bet that 50N is more in the range of what an FRC robot gets on carpet, as it implies acceleration of nearly 1G for a max-weight robot. You'd have a 120lb rocket. (Edit: this is wrong. You'd have acceleration of 1m/s^2, which is nearly 10 times less than 1G. I need to go to bed).

I made an excel sheet to play with these numbers:
If you had a fan with a radius just wide enough to fit on the short length of your robot (35cm), you'd only need an output velocity of 3.1m/s for a 5 newton boost. It would only require 7.8 watts. For a 10N boost with the same fan, you need a 4.5m/s output. However, since a lot of that boost would get caught on the trailer (thus slowing you down), you'd need more.

Edit: But it occurs to me that a giant fan would pose an entanglement risk, and so probably isn't practical, if QA allows it at all.

[Ian Curtis]

Just for arguments sake, remember robots actually weigh about 150 pounds. The 120 base that passes inspection, plus a 13 poundish battery, plus 15 pounds of battery.

Also, can you provide some documentation about motor outputs? I though CIMs output about 300 watts.

[Bongle]

Originally Posted by iCurtis

Just for arguments sake, remember robots actually weigh about 150 pounds. The 120 base that passes inspection, plus a 13 poundish battery, plus 15 pounds of battery.

Also, can you provide some documentation about motor outputs? I though CIMs output about 300 watts.

I'm going by what we're allowed circuit-breaker wise. We have a 12V system with 40amp breakers. Thus, the most power we're allowed using for a single motor is 480 watts. Due to conduction losses, the most useful power we're allowed is a bit less than that, but it is in that region of power.

I don't know the specs for the individual motors, it is possible that 480 watts through one of the kit motors is impossible.

[smartkid]

This might sound stupid but where would you go looking for fans / propellers? And what kind of gearing would get a CIM to spin the most efficient.

EDIT: Talking about using this to give a drive a little boost not to push an entire bot around.

[Bongle]

Originally Posted by smartkid

This might sound stupid but where would you go looking for fans / propellers? And what kind of gearing would get a CIM to spin the most efficient.

EDIT: Talking about using this to give a drive a little boost not to push an entire bot around.

This is an R/C fan that generates 4N, but spins at a pretty crazy RPM to do so: http://www.modelflight.com.au/rc_mod..._micro_fan.htm.

I think it would be worth your time putting a standard house fan on a physics cart and use a spring scale to see how much force it generates. Then you could maybe replace the house fan motor with a CIM and speed it up a bit.

Again, I note that you probably don't want to invest too time or money in this idea unless QA says it is legal. Their "only traction must come from rover wheels" rule could potentially be interpreted (FIRST has reached pretty far in their rule interpretations in the past) that you can't get thrust from fans either.

[drkiraco]

You're right, max output of a CIM is somewhere in the 300 W range. For these back of the envelope calcs, I was assuming 100% efficiency in the CIM motor (and in the gearbox, and in the fan). 12V * 40A = 480W is the maximum input to the motor. It's not 100% efficient. But, you could always use two of them put together if you wanted to...

Originally Posted by iCurtis

Just for arguments sake, remember robots actually weigh about 150 pounds. The 120 base that passes inspection, plus a 13 poundish battery, plus 15 pounds of battery.

Also, can you provide some documentation about motor outputs? I though CIMs output about 300 watts.