110ft/s (75mph) robot design

[Aaron Ng]

Awhile back, I saw team 868's Totebot, a robot in a tote that drives at 50mph. I was inspired by their video to design a robot that can go as fast as possible using only FRC legal parts.

I have named the design: Sanic

It has 6 cims, geared at 34-50 at low, and 50-34 at high. achieving 35mph and 75mph respectively.

Amazingly, the math all works out, for the amperage at least. By limiting the amperage for each cim at startup to 40, and then bumping it up to around 50 when shifting to high, it can reach 75mph without the 40 amp snap action cim breaker, or 120 amp main breaker popping. It comes very darn near to dying, but it doesn't.

I'm not a electronics guy, so I'm not sure if there is anything else that will go wrong (there probably is) except for the roboRIO browning out, if that is the correct term.

Of course, this isn't something that I will ever build. It was a fun thought experiment, and a challenge, to see how far I can push the equipment that we have.

And before anyone asks, no. I don't have a way to stop it.

Thoughts? Questions? I would love to hear what the community thinks of a crazy design such as this.

[seg9585]

Originally Posted by Aaron Ng

By limiting the amperage for each cim at startup to 40, and then bumping it up to around 50 when shifting to high, it can reach 75mph without the 40 amp snap action cim breaker, or 120 amp main breaker popping. It comes very darn near to dying, but it doesn't.

Can you explain how running 6 CIMs at 40 amps each will not trip the 120amp breaker switch after a few seconds?

Did you use the JVN Design calculator to figure your expected motor currents with that configuration? Although I don't have the other specs that I need of your design (wheel diameter, weight on wheels), I did some guessing and the resulting "pushing match" current (in this case accelerating from 0) runs up to 152 amps per motor.

[GeeTwo]

How much runway and time do you calculate that you'll need to reach top speed?

[carpedav000]

Could I get a picture of the gearboxes (and cims) only. I'm very intrigued.

[Nathan Streeter]

Originally Posted by seg9585

Although I don't have the other specs that I need of your design (wheel diameter, weight on wheels), I did some guessing and the resulting "pushing match" current (in this case accelerating from 0) runs up to 152 amps per motor.

I don't have the time to check your other points... but the CIM can draw "only" 133A at stall...

[Aaron Ng]

Originally Posted by seg9585

Can you explain how running 6 CIMs at 40 amps each will not trip the 120amp breaker switch after a few seconds?

Did you use the JVN Design calculator to figure your expected motor currents with that configuration? Although I don't have the other specs that I need of your design (wheel diameter, weight on wheels), I did some guessing and the resulting "pushing match" current (in this case accelerating from 0) runs up to 152 amps per motor.

6 CIMs at 40 amps draws 240 amps. AndyMark's data sheet for the 120 amp breaker states that at 200% of the rated current (lets say a bit more because of other electronics) it can last a bit less then 10 seconds. It takes less then 10 seconds for me to reach 75mph. And the 50 amps only happen for like half a second when switching, and the cim snap breakers can draw that for a couple seconds.

wheel diameter is 4in
weight is around 50lbs (the 10lbs for electronics is added in, but not shown)
the pushing match current is if you are pressed up against a wall. in which case, this drivetrain is more then likely power limited which is why the current drawn is so high.

Originally Posted by GeeTwo

How much runway and time do you calculate that you'll need to reach top speed?

Im a perfect world with no opposing force, it would take a 75m runway to reach 75mph. Im guessing it would be around 100 irl.

i'll post all of my calculations when i get home.

[Chak]

Hey Aaron, carpevdav000 has a rather appropriate signature for this thread:

Originally Posted by carpevdav

Always follow your dreams, unless your dream is stupid.

[nuclearnerd]

Originally Posted by Aaron Ng

Im a perfect world with no opposing force, it would take a 75m runway to reach 75mph. Im guessing it would be around 100 irl..

75 meters to reach 75 mph? Are you sure you're not Canadian (We have a similarly "creole" unit system here)

[s_forbes]

I think you may be going a little overkill with the motors there.

If you want something to go fast (as in, high speed, not high acceleration) then you just need enough power to counteract the drag you would see at that speed. Getting rid of drag is just as good as adding power. I would suggest focusing more on a small profile robot with a gearing system that lets you go through a wide range of ratios. With the size of the components, I don't think you'd need more than 1 CIM to power it if it was in an aerodynamic shell.

Also, you should get rid of that 6wd tank style steering system... It works good when you are driving a robot around small spaces on an FRC field, but you won't be able to control a fast moving robot with it very well. Cars use Ackermann style steering because it's easy to control when you're driving down the road.

[Nathan Streeter]

Aaron, you've inspired me to think more about this concept... make as fast a robot as possible with 'ordinary' FRC parts. I think this'll be a design project for me for a while. :-)

Having thought about it some, I think a good design would probably look a fair bit more different from the typical FRC robot than your design does.

Originally Posted by s_forbes

I think you may be going a little overkill with the motors there.

If you want something to go fast (as in, high speed, not high acceleration) then you just need enough power to counteract the drag you would see at that speed. Getting rid of drag is just as good as adding power. I would suggest focusing more on a small profile robot with a gearing system that lets you go through a wide range of ratios. With the size of the components, I don't think you'd need more than 1 CIM to power it if it was in an aerodynamic shell.

I agree that you have too many motors and not enough shifting speeds. 4 CIMs are probably about all your battery and breakers can handle for the acceleration period, I'd think. If I were designing this, I'd have 3 speeds (or more) based on a custom ballshifting setup... if you make a custom ballshifting shaft and plunger you can easily get a sequential 3-speed gearbox with two .5" throw pancake cylinders. 4 speeds would be nice, but in that case you might want to start looking at a more elegant (but still highly fast!) shifting setup. 3 speeds should be able to keep you in a 'happy band' of 1800-4200rpm for most of your acceleration time where each motor is always producing 220-300+W and drawing 30-90A.

Originally Posted by s_forbes

Also, you should get rid of that 6wd tank style steering system... It works good when you are driving a robot around small spaces on an FRC field, but you won't be able to control a fast moving robot with it very well. Cars use Ackermann style steering because it's easy to control when you're driving down the road.

I'd definitely second using anything but a 6wd Tank setup... Tank is good for pushing with decent maneuverability in tight spaces. It'll start to be intensely difficult to control above ~30fps or so, I suspect. Ackermann Steering would be ideal, but is substantially less easy. Turntable Steering might be a reasonable compromise, if you're looking for something that has 'adequate' steering. A tank setup would be OK if you're only looking for a drag racer (not an RC car), in which case I'd recommend making the rocker very small.

Originally Posted by Aaron Ng

6 CIMs at 40 amps draws 240 amps. AndyMark's data sheet for the 120 amp breaker states that at 200% of the rated current (lets say a bit more because of other electronics) it can last a bit less then 10 seconds. It takes less then 10 seconds for me to reach 75mph. And the 50 amps only happen for like half a second when switching, and the cim snap breakers can draw that for a couple seconds.

I'm calculating that it'd (sort of) realistically take no less than 15 seconds... I'll share my calculations, but I'd be interested in knowing what you're doing to get 10s or less. I'm guessing when you factor in air resistance, a realistic battery, and battery depletion that it might take 10-15s to get to 50mph and another 10-15s to get to 75mph from there. Internal Combustion engines typically peak in power and torque at high rpms... unfortunately for a design like this, to get from 50mph to 75mph not only do you need to be able to push against twice the drag force and double your kinetic energy, but you're doing it as your motor starts putting out less and less power and torque.

Originally Posted by Aaron Ng

wheel diameter is 4in
weight is around 50lbs (the 10lbs for electronics is added in, but not shown)
the pushing match current is if you are pressed up against a wall. in which case, this drivetrain is more then likely power limited which is why the current drawn is so high.

I'd generally recommend using larger wheels so you need lower rpms to get to 50-75mph. A 4" wheel needs to spin at 6000rpm or so to get to 75mph... not only is this potentially unsafe for many of the 4" wheels we use in FRC, but it means you need to have your CIMs above a 1:1 in high gear. Use an 8" or larger wheel... you should be able to do your reduction in 1.5-2 stages and it'll be much safer. Additionally, I'd recommend using a pneumatic wheel or something that has some compliance. Driving little wheels with minimal compliance and no suspension on exterior surfaces at 50+mph could easily shock or vibrate apart a frame or damage the electronics or battery.

Your weight seems a little low for including electronics and battery... are you including both?

Originally Posted by Aaron Ng

Im a perfect world with no opposing force, it would take a 75m runway to reach 75mph. Im guessing it would be around 100 irl.

i'll post all of my calculations when i get home.

Because I think it'll take a substantial percentage of the time to cover the final 25mph, I suspect the 75m may be quite low. This will definitely not accelerate linearly.

[Jared]

I'd use one of those fancy internal hub shifting things that are used on bikes to get tons of different gear ratios. I'd also use two batteries with two main breakers.

[Aaron Ng]

Originally Posted by Nathan Streeter

Aaron, you've inspired me to think more about this concept... make as fast a robot as possible with 'ordinary' FRC parts. I think this'll be a design project for me for a while. :-)

Having thought about it some, I think a good design would probably look a fair bit more different from the typical FRC robot than your design does.



I agree that you have too many motors and not enough shifting speeds. 4 CIMs are probably about all your battery and breakers can handle for the acceleration period, I'd think. If I were designing this, I'd have 3 speeds (or more) based on a custom ballshifting setup... if you make a custom ballshifting shaft and plunger you can easily get a sequential 3-speed gearbox with two .5" throw pancake cylinders. 4 speeds would be nice, but in that case you might want to start looking at a more elegant (but still highly fast!) shifting setup. 3 speeds should be able to keep you in a 'happy band' of 1800-4200rpm for most of your acceleration time where each motor is always producing 220-300+W and drawing 30-90A.



I'd definitely second using anything but a 6wd Tank setup... Tank is good for pushing with decent maneuverability in tight spaces. It'll start to be intensely difficult to control above ~30fps or so, I suspect. Ackermann Steering would be ideal, but is substantially less easy. Turntable Steering might be a reasonable compromise, if you're looking for something that has 'adequate' steering. A tank setup would be OK if you're only looking for a drag racer (not an RC car), in which case I'd recommend making the rocker very small.



I'm calculating that it'd (sort of) realistically take no less than 15 seconds... I'll share my calculations, but I'd be interested in knowing what you're doing to get 10s or less. I'm guessing when you factor in air resistance, a realistic battery, and battery depletion that it might take 10-15s to get to 50mph and another 10-15s to get to 75mph from there. Internal Combustion engines typically peak in power and torque at high rpms... unfortunately for a design like this, to get from 50mph to 75mph not only do you need to be able to push against twice the drag force and double your kinetic energy, but you're doing it as your motor starts putting out less and less power and torque.



I'd generally recommend using larger wheels so you need lower rpms to get to 50-75mph. A 4" wheel needs to spin at 6000rpm or so to get to 75mph... not only is this potentially unsafe for many of the 4" wheels we use in FRC, but it means you need to have your CIMs above a 1:1 in high gear. Use an 8" or larger wheel... you should be able to do your reduction in 1.5-2 stages and it'll be much safer. Additionally, I'd recommend using a pneumatic wheel or something that has some compliance. Driving little wheels with minimal compliance and no suspension on exterior surfaces at 50+mph could easily shock or vibrate apart a frame or damage the electronics or battery.

Your weight seems a little low for including electronics and battery... are you including both?



Because I think it'll take a substantial percentage of the time to cover the final 25mph, I suspect the 75m may be quite low. This will definitely not accelerate linearly.

I know this configuration isn't optimal, but I wanted to design something that my team has the resources to make. Something like, for example, a round aerodynamic shroud that covers the entire robot, would be beyond what we can build.

Yes, looking back, more shifting speeds would definitely be more optimal than my solution of more motors. Less stress on all the other electronics while being lighter. However, the reason for all the motors is that it has to start and stop within a 200m runway, as that is the longest we have at school. An artificial restraint, of course, as we won't build it, but was definitely a fun, additional challenge for me. Would more stages compensate for lower power? I don't know how to design shifters, but I’m thinking of having two or even three shifters next to each other.

The idea behind the tank drive was that the track at school has a turn in the beginning, so for the first, perhaps 70m I will be turning while accelerating. I was thinking of sticking my swerve design on the front, and taking out the CIMs, (is that what turntable is?) but the extra weight stopped me. Ackerman steering sounds quite difficult to design as well as build. I was thinking a simple caster, but traction may become a factor.

The reason for the smaller wheel was all because of weight. not only is the wheel larger, but now I need a way to mount the wheel above the drivetrain as I would think that keeping the robot as low as possible would be the best. But perhaps I am wrong on that point.

The weight is that low because everything is made from 1/16. That is also how I got an 18t sprocket to work. We have custom weights on most COTS parts, and solidworks estimated everything to be around 40lbs (with battery). I’m guessing that the electronics and chain would add another 10. Thus, 50

Wow! That just tore my design completely to pieces. My calculations were all done in an ideal world. No voltage drop, no air resistance, etc. I did calculate air resistance, but only at 75mph, and with a lot of guessing, since that is the extent of my knowledge in physics. So I guess that this design is no longer an "it might work in theory" but rather an "it will probably crash and burn in theory". Though I won’t be working on this for quite a while. It has already taken up enough of my time that I was dedicating to college apps.

My calculations are as follows:

Amperage and pushing strength at taken from JVNcalc

Pushing strength at low gear, with motors limited to 40amps is 27lbs or 120N
F=MA
120=22.68A
A=5.29m/s^2

Pushing strength at low gear, with motors limited to 40 amps is 13lbs or 57.8N
F=MA
57.8=22.68A
A=2.55m/s^2

Vf=Vo+At

15.64=0+5.29t
t=2.96s

33.5=15.64+2.55t
T=7s

7s+2.55s=9.55s

120 amp breaker pops in 10s when under 200% load,

Of course, this is all in a ideal world. So irl, it probably won’t be as nice as this. Acceleration drops over time, but amperage does as well. Is that enough to stop parts from dying? Im not sure.

I'd use one of those fancy internal hub shifting things that are used on bikes to get tons of different gear ratios. I'd also use two batteries with two main breakers.

Shifting chain would save a lot of weight, but I wouldn't necessarily trust it while going that fast. As for more batteries and breakers. that would make it FRC illegal. but I don't recall any rule that limits robot speed

[Jared]

Originally Posted by Aaron Ng

Shifting chain would save a lot of weight, but I wouldn't necessarily trust it while going that fast. As for more batteries and breakers. that would make it FRC illegal. but I don't recall any rule that limits robot speed

https://rideons.files.wordpress.com/...i-hub-gear.jpg

This is the sort of shifting transmission I was thinking about. Somebody made a cool battery powered tricycle using an 8 speed one that could get going pretty quickly.

[Aaron Ng]

Originally Posted by Jared

https://rideons.files.wordpress.com/...i-hub-gear.jpg

This is the sort of shifting transmission I was thinking about. Somebody made a cool battery powered tricycle using an 8 speed one that could get going pretty quickly.

ahhh. i've seen those before, but i don't know how they work. do they shift with a wire like a traditional bicycle?

[nuclearnerd]

Originally Posted by Aaron Ng

ahhh. i've seen those before, but i don't know how they work. do they shift with a wire like a traditional bicycle?

They work with friction wheels that turn in a toroidal cylinder (some car CVTs work in a similar method). They're also really wasteful, < 85% efficient.

If you were going to use a bicycle hub transmission, get something like a shimano alfine 11. . Just make sure to cut the current during the shift, because the won't shift under full torque.