I would contend that the net acceleration vector will always be normal to the seat when the motorcycle is in motion.
Yep, that's correct. As lean angle increases, acceleration must also increase to maintain equilibrium. It's more easily visualized by applying the golden rule of steering a cycle, called "Push right- Go right" which means that as you're riding and want to turn to the right, you push with your right hand (You do no leaning at all..) This turns the front wheel to the left, which leans the bike to the right and you steadily apply throttle (accelerating) through the turn. It's the only correct way to maintain steady balance and traction.
So- the more you lean, the more you must accelerate. The more you do this, the more G force you'd see on an indicator, but it's normal to the frame/seat/whatever of the cycle. A single axis gyro and accelerometer could be used to calculate your speed. Your speed and accelerometer could be used for the lean angle. Then again, you could use a potentiometer mounted on the steering column for the degree of "turning" you're doing and your speed to figure it out. Then, filtering would be more simplified because neither parameter is affected to any large degree by terrain (unless you're in the midst of a wreck or losing traction!)
Yep, that's correct. As lean angle increases, acceleration must also increase to maintain equilibrium. It's more easily visualized by applying the golden rule of steering a cycle, called "Push right- Go right" which means that as you're riding and want to turn to the right, you push with your right hand (You do no leaning at all..) This turns the front wheel to the left, which leans the bike to the right and you steadily apply throttle (accelerating) through the turn. It's the only correct way to maintain steady balance and traction.
So- the more you lean, the more you must accelerate. The more you do this, the more G force you'd see on an indicator, but it's normal to the frame/seat/whatever of the cycle. A single axis gyro and accelerometer could be used to calculate your speed. Your speed and accelerometer could be used for the lean angle. Then again, you could use a potentiometer mounted on the steering column for the degree of "turning" you're doing and your speed to figure it out. Then, filtering would be more simplified because neither parameter is affected to any large degree by terrain (unless you're in the midst of a wreck or losing traction!)
It's been a while since I've driven a motorcycle but I believe you actually push the handle bars in opposite direction of the turn to initiate a turn. A small push to the left will cause the bike to start leaning to the right. The handlebar is then moved as needed to balance the bike while leaning and turning right. The handlebar movements can be so slight, I doubt you'd be able to get any useful information from their movements.
Edit: I see you were saying the same thing with "push on the right". I still think the movements would be too slight to use in a meaningful way. (Though I'm not sure.) It would be interesting to test your theory.
...steadily apply throttle (accelerating) through the turn. It's the only correct way to maintain steady balance and traction.
This is not true. It is perfectly possible and natural to transition a motorcycle from straight ahead to seriously banked turn at constant forward speed. No extra throttle is required.
Also consider that if you do that and you happen to be taking the turn at the highest speed possible opening the throttle is only going to break traction on the rear wheel and you are off the road, Similarly braking in a turn is not a good idea.
...No matter what the vehicle is doing there will always be at least one gee present. Since it is constrained to a two dimensional surface you should be able to solve what that vector is. ....
Rich, I think I see what you're getting at. On a flat track, and because the motorcycle is earthbound, there will always be one G normal to the track's surface. Any additional accelerations will be due to centripetal factors encountered during turns....
I believe the confusion arises from the fact you have different frames of inertial reference when you are talking about a static situation vs. a dynamic situation. In the static situation, both the driver of the motorcycle and some dude standing on the ground have the same frame of reference, so Rich's system makes sense and gives useful results. But once the motorcycle is going around in an arc, the driver and a dude on the ground are in two different frames of reference, so Rich's system would break down. If you consider the case of an airplane making a coordinated turn, then consider what a passenger might experience in total darkness without a view of the horizon. They would experience a slightly higher g-load, but there would be no way to know if that higher g-load is from making a turn or accelerating vertically on a straight run. Their frame of inertial reference would be totally divorced from the reference frame of earth. Placing accelerometers in different orientations would only provide you with their respective components of what appears to be a single acceleration vector. As Phil has pointed out, there would be no way to transform or "back out" the acceleration vector associated with the centrifugal effect.
Actually it can be done on a motorcycle on flat ground using the 2 axis accelerometers plus a measured forward acceleration using a speed sensor.
Why not go all the way and use a 3 axis accelerometer plus a measured forward acceleration using a speed sensor. Now you also have road angle up or down hills.
Ya, there is a bunch of math involved but it's doable.
If you think the forward angle accelerometer helps, then a two-axis unit should be adequate, since the tranverse acceleration is always zero at equilibrium.
But how to filter those out? I would contend that the net acceleration vector will always be normal to the seat when the motorcycle is in motion.
Also, whether the accelerometer axes are at 45/45 degrees, 0/90 degrees, or 30/60 degrees is irrelevant. Since the measurement vectors are orthogonal, it will be possible to transform the measurement from any sensor orientation to a rotational frame with the Y axis normal to the seat.
-Phil
Sure, but you don't have to look at just the net acceleration. The accelerometer can separate out the accelerations along a particular axis, unlike a human that would only sense the acceleration normal to the motorcycle. It seems to me that being able to measure the net acceleration, individual accelerations on perpendicular axis' and the fact that there is a two dimensional constraint should be enough to determine angle of lean.
The accelerometer can separate out the accelerations along a particular axis, unlike a human that would only sense the acceleration normal to the motorcycle
That's just the problem, though: there is no component of acceleration along the tranverse axis. There's nothing to measure in that direction: it's zero.
If you think the forward angle accelerometer helps, then a two-axis unit should be adequate, since the tranverse acceleration is always zero at equilibrium.
-Phil
True, if the rider on the bike keeps his center of gravity over the bikes center of gravity.
However, when watching road racing you can see the rider crawling all over to maximize the tire traction angles.
So three axis accelerometers plus a measured forward acceleration using a speed sensor are required.
True, if the rider on the bike keeps his center of gravity over the bikes center of gravity.
'Not sure what "center of gravity" means in a non-inertial frame of reference. In any event, the acceleration will always lie along a line connecting the bike-plus-rider's center of mass to the line between the two tires' contact points with the pavement. I guess that's more explicit than saying "normal to the seat." But now that begs the question: what is meant by the "angle of the motorcycle" if the rider is hiked out to one side or the other?
Here is a simple experiment you can do in your mind or in reality if you like:
1) Take a wine glass and glue it to your motorcycles gas tank in as much an upright position as possible.
2) Fill the wine glass half full of wine.
3) Now start riding your bike and get up to a constant speed.
4) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
5) Now take a turn in the road, without changing forward speed.
6) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
Clearly the wine is experiencing some extra acceleration downwards into the glass as you turn but that does not cause it to move. It sees no acceleration left or right or forwards or backwards.
Now replace the glass of wine with a three axis accelerometer. The only axis that will show any change is the vertical.
Measure that. From that calculate the centrifugal force and from that the angle of lean. Job done.
Is that going to get you an accurate result? No idea.
P.S. It's better to use a plastic wine "glass" in case of mishaps.
when watching road racing you can see the rider crawling all over
OK, if we get into that style things get harder, The wine in my experiment starts to slop out to the side.
We now have the C of G of the bike plus rider somehwere perhaps not within the bike itself. Certialy not directly above the tire patches in "bike coordinates".
Let me try to illustrate this. Let's say Rich has a supermodel girlfriend, say her name is Gorga, and she's promised to do whatever he wants so long as she gets to ride his motorcycle for free.
Gorga has 5 arms, so she's able to safely drive the motorcycle with two arms and have three left over to perform this experiment.
Gorga also has 5 eyes but, problem is, she can't see more than 2 feet in front of her, so Rich has to build her an instrumentation board so she can tell what she's doing as she speeds around in a bunch of big circles.
The board has accelerometers attached at each clock position in front of her - at 9 o'clock, 10 o'clock...etc to 3 o'clock. Each accelerometer can be turned on an axis nailed to the board and its measurement can be easily read by any of Gorga's 5 eyeballs.
So before Gorga takes off, Rich has her lean the bike over and take some readings. She is told to turn each accelerometer until it gives the highest reading, and then write down the value of each one. Sitting still and twiddling the accels around, she finally notes that all accels are giving their highest reading when pointing straight down toward the center of the earth and each reading is 1 G. She notes the angle of the accel readings vs. the position of the motorcycle, and everything looks fine.
But now Gorga takes off, wind in her hair, and drives around in a big circle. As instructed, she twiddles and fidgets with each accel until each one is pointing in the direction of its highest accel level. To her surprise, all the accels are pointing in the same direction, a vector normal to her seat. No matter where the accels are located on her board, they all have their maximum value in the same direction.
The question is, how would she now be able to extract from any of her accels the vector responsible for the centrifugal effects? Or what angle would there be to measure? She can't see very far, so she must rely only on her accels for information, but what would she be able to tell other than her accels are reading slightly more than 1 g?
This is not true. It is perfectly possible and natural to transition a motorcycle from straight ahead to seriously banked turn at constant forward speed. No extra throttle is required.
Also consider that if you do that and you happen to be taking the turn at the highest speed possible opening the throttle is only going to break traction on the rear wheel and you are off the road, Similarly braking in a turn is not a good idea.
I think we're disagreeing on "proper riding techniques" here, not the physics behind it. You're correct- it's possible and even very common to bank without applying throttle. But when you do so, you begin to turn. And when you turn, you decelerate, given constant throttle. That causes a decrease in centripetal force, which in turn causes you to lean more- which causes you to decelerate more. (It's the exact same principle as when you roll a coin across a table. Once it "leans" and goes into a spiral, the spiral gets smaller and smaller until the coin flops over. There is no force causing the coin to accelerate and maintain the same path.) Most folks do it the old school way- they shift their weight back and forth through a turn as needed, or they push more/less as needed without adjusting throttle. But in motor cycle riding courses, the instructor will always teach you to apply some constant amount of throttle through a curve and adjust your "push" without shifting your weight at all. The increased stability outweighs the increase in velocity for safety.
When you're traveling at a high velocity (for example 60 mph) it takes a surprising amount of force to "push on the right handle bar" and make the handlebar rotate a large amount of degree. Fortunately, at a high speed like that, it takes a very small degree of rotation to cant you into a lean and start turning. So again- you're correct, but in practise it's almost impossible to "over-push" because of the physical force needed to rotate the handle bars enough to cause the reaction you've mentioned. If applying throttle would cause you to lose control, well.... you shouldn't have been going that fast to begin with. (It CAN happen, to be sure!)
You're dead on correct about braking in a turn. Never, ever do it. Straighten out, then brake.
@Duane - I remember a lot of the harley crowd likes to say "Pull left go right" which is the same thing- but they say it that way because a lot of those types of bikes put the rider in a laid back position (thinking about hogs with the Ape Hangar handle bars and the like) and it's much more difficult to "push" when your arm is already extended outward as far as it can go. With bikes like that, it's easier to 'pull left' to go right than to 'push right' but the physics is the same. By pulling, they don't have to lean their body forward in order to get one arm bent a bit so that it can push forward.
The speed of the cycle and anglular rotation of the handlebars can definitely be used to determine the lean angle. At 70 mph, tackling a typical interstate curve to the right will cause your handlebar rotation to be about 4 or so degrees. I'm basing that on remembering visuals of it when I went through rider training courses. So obviously, 4 degrees is not exact. (We had to take 3 courses to get a permit to have a Goldwing or similar bike on base back in my military days.)
OK, if we get into that style things get harder, The wine in my experiment starts to slop out to the side.
We now have the C of G of the bike plus rider somewhere perhaps not within the bike itself. Certialy not directly above the tire patches in "bike coordinates".
That's the point. Real riders ride in the real world.
With saddle bags unevenly loaded.
Racers are maximizing cornering speed.
And then there is the wind pushing sideways.
The question is, how would she now be able to extract from any of her accels the vector responsible for the centrifugal effects?
I don't have enough drugs at had to partake of the Gorga experiment but...
The point is that a single accelerometer measuring g vertically with respect to the bike will measure 1g when the bike is traveling in a straight line at constant speed.
It will measure something greater than 1g if the bike is in a turn at constant speed.
Assuming the rider is not hanging off the side of the bike racer style that can be used to get the centripetal component of the acceleration and hence the angle of lean. If you also know the forward speed.
Things get harder if the bike has forward acceleration in the curve, or if the rider has shifted his weight over to the side.
In the old days we used to measure the angle of lean by how much metal was scraped off the foot rests by the road surface. BMW owners used to do it by seeing how much of the cylinder heads of their boxer twins were missing, in some cases leaving the valves exposed!
...I don't have enough drugs at had to partake of the Gorga experiment but...
In the old days we used to measure the angle of lean by how much metal was scraped off the foot rests by the road surface. ....
Ah, but that presumes the road surface is not tilted one way or another with respect to the horizon. Allow me to illustrate.... oh, uh, but first, let me introduce you to my good friend Phreandria...
Here is a simple experiment you can do in your mind or in reality if you like:
1) Take a wine glass and glue it to your motorcycles gas tank in as much an upright position as possible.
2) Fill the wine glass half full of wine.
3) Now start riding your bike and get up to a constant speed.
4) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
5) Now take a turn in the road, without changing forward speed.
6) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
Clearly the wine is experiencing some extra acceleration downwards into the glass as you turn but that does not cause it to move. It sees no acceleration left or right or forwards or backwards.
Now replace the glass of wine with a three axis accelerometer. The only axis that will show any change is the vertical.
Measure that. From that calculate the centrifugal force and from that the angle of lean. Job done.
Is that going to get you an accurate result? No idea.
P.S. It's better to use a plastic wine "glass" in case of mishaps.
I can tell youI've come real close to losing a passenger off the back from accelerating too fast. Wouldn't the same hold true with the wine glass?
If you consider the case of an airplane making a coordinated turn, then consider what a passenger might experience in total darkness without a view of the horizon. They would experience a slightly higher g-load, but there would be no way to know if that higher g-load is from making a turn or accelerating vertically on a straight run. Their frame of inertial reference would be totally divorced from the reference frame of earth.
A practical example of this is experienced by anyone who has ever obtained a private pilot license. Your instructor will put a visor on your head that allows you to view only the instrument cluster. Then he puts the plane in a crazy mixed-up attitude and finally turns control over to you. Your senses tell you one thing and your instruments tell you another. The average student's' first reaction is to trust his sense of up and down and disregard the instruments. Big mistake.
BTW, I love the pic of ElectricAye's good friend Phreandria! Never seen a prettier specimen of Amanita muscaria.
I've come close to posting the same observation several times.
You showed good judgement and restraint, thus you didn't post. But me, I've never been hampered by such considerate discretion. THAT'S why I have twice as many posts as you, DD.
A practical example of this is experienced by anyone who has ever obtained a private pilot license. Your instructor will put a visor on your head that allows you to view only the instrument cluster. Then he puts the plane in a crazy mixed-up attitude and finally turns control over to you. Your senses tell you one thing and your instruments tell you another. The average student's' first reaction is to trust his sense of up and down and disregard the instruments. Big mistake.
BTW, I love the pic of ElectricAye's good friend Phreandria! Never seen a prettier specimen of Amanita muscaria.
Foggles - Unusual attitude recovery, teaches you to look at the instruments and recover accordingly, usually throttle, bank then pitch. When they teach you not to trust your senses is when you cannot see outside or inside, then you let the airplane fly and steer with the rudder and throttle only It's pretty freaky, I suppose you have been in that situation a few times. I'm finishing up IFR now, I love flying... and bikes.
On that note, I'll probably be testing some theories before commenting any further. Some people here I'm sure have some degrees related to the subject, however I think that in the real world the way you apply your knowledge to a situation may blind you. From what I've read so far there has been no baseline or control for what anyone is talking about.
What kind of bike? Is it accelerating? Flat road? etc etc.... There are so many details left out I feel like I'm going to have to bail on this one.
The question is, how would she now be able to extract from any of her accels the vector responsible for the centrifugal effects? Or what angle would there be to measure? She can't see very far, so she must rely only on her accels for information, but what would she be able to tell other than her accels are reading slightly more than 1 g?
I would ask Gorga to rotate one of the accelerometers until it read exactly one gee and take note of it's angle and which side it was on relative to where it read the maximum value.
I would ask Gorga to rotate one of the accelerometers until it read exactly one gee and take note of it's angle and which side it was on relative to where it read the maximum value.
Maybe I'm not understanding what you mean by taking "note of it's angle and which side it was on relative to where it read the maximum value," but if I understand you correctly, then I believe that it will not matter which side of the board any of the accelerometers are on - they will all see the same maximum acceleration vector (let me call it the Big A) pointing in the same direction, which is a vector normal to the plane of the motorcycle seat. Tilting the accels this way or that will simply cause them to see only a portion of that Big A acceleration vector. For example, you could tilt the 9 o'clock accel until it sees 1 G, but that tilt could be performed either clockwise or counter clockwise by the same angle around its nail. If you tilted the accels so they are 90 degrees to the normal of the seat, then they would read zero. In other words, there would be no difference what the 9 o'clock accel sees vs. what the 3 o'clock accel sees. They all see the same Big A vector, and tilting them only causes them to read a component of the Big A.
Again, I'm not saying it's impossible to derive tilt angle from a single accelerometer under some circumstances - I'm just saying it's not possible with a single accel to tell the difference between the > 1 G value caused by a turn vs. a > 1 G value caused by a vertical acceleration. Also, a single accel could not tell you in which direction you are turning. You would need a gyro or some other reference system for that.
Place a two axis accelerometer on a merry-go-round such that one axis is pointed at the center of the merry-go-round (A) and the other is pointing straight up (Z). I think we can agree that the while the merry-go-round is stationary that the Z axis will read one gee and the A axis will read zero gees.
Now, spin the merry-go-round. Obviously the A axis reading is going to increase but what about the Z axis? I assert that the Z axis will remain at one gee.
Now lets rotate the accelerometer so that the Z axis is at it's highest value. Let's say it was rotated five degrees. That will be the angle that you would have to stand in order to not fall off the merry-go round. The difference between the max gee and one gee readings will tell you the lean angle.
Now, spin the merry-go-round. Obviously the A axis reading is going to increase but what about the Z axis? I assert that the Z axis will remain at one gee.
unless... the merry-go-round is on the back of a flat bed truck and the truck is moving up and over a small hill. The Z axis is no longer one gee and you can't tell if it was the hill or the tilt that changed the Z reading..
The problem is fundamentally the same as trying to fly a plane in clouds - it needs an artificial horizon and gyros are very cheap nowdays.
Comments
Yep, that's correct. As lean angle increases, acceleration must also increase to maintain equilibrium. It's more easily visualized by applying the golden rule of steering a cycle, called "Push right- Go right" which means that as you're riding and want to turn to the right, you push with your right hand (You do no leaning at all..) This turns the front wheel to the left, which leans the bike to the right and you steadily apply throttle (accelerating) through the turn. It's the only correct way to maintain steady balance and traction.
So- the more you lean, the more you must accelerate. The more you do this, the more G force you'd see on an indicator, but it's normal to the frame/seat/whatever of the cycle. A single axis gyro and accelerometer could be used to calculate your speed. Your speed and accelerometer could be used for the lean angle. Then again, you could use a potentiometer mounted on the steering column for the degree of "turning" you're doing and your speed to figure it out. Then, filtering would be more simplified because neither parameter is affected to any large degree by terrain (unless you're in the midst of a wreck or losing traction!)
It's been a while since I've driven a motorcycle but I believe you actually push the handle bars in opposite direction of the turn to initiate a turn. A small push to the left will cause the bike to start leaning to the right. The handlebar is then moved as needed to balance the bike while leaning and turning right. The handlebar movements can be so slight, I doubt you'd be able to get any useful information from their movements.
Edit: I see you were saying the same thing with "push on the right". I still think the movements would be too slight to use in a meaningful way. (Though I'm not sure.) It would be interesting to test your theory.
Also consider that if you do that and you happen to be taking the turn at the highest speed possible opening the throttle is only going to break traction on the rear wheel and you are off the road, Similarly braking in a turn is not a good idea.
I believe the confusion arises from the fact you have different frames of inertial reference when you are talking about a static situation vs. a dynamic situation. In the static situation, both the driver of the motorcycle and some dude standing on the ground have the same frame of reference, so Rich's system makes sense and gives useful results. But once the motorcycle is going around in an arc, the driver and a dude on the ground are in two different frames of reference, so Rich's system would break down. If you consider the case of an airplane making a coordinated turn, then consider what a passenger might experience in total darkness without a view of the horizon. They would experience a slightly higher g-load, but there would be no way to know if that higher g-load is from making a turn or accelerating vertically on a straight run. Their frame of inertial reference would be totally divorced from the reference frame of earth. Placing accelerometers in different orientations would only provide you with their respective components of what appears to be a single acceleration vector. As Phil has pointed out, there would be no way to transform or "back out" the acceleration vector associated with the centrifugal effect.
Actually it can be done on a motorcycle on flat ground using the 2 axis accelerometers plus a measured forward acceleration using a speed sensor.
Why not go all the way and use a 3 axis accelerometer plus a measured forward acceleration using a speed sensor. Now you also have road angle up or down hills.
Ya, there is a bunch of math involved but it's doable.
Duane J
If you think the forward angle accelerometer helps, then a two-axis unit should be adequate, since the tranverse acceleration is always zero at equilibrium.
-Phil
Sure, but you don't have to look at just the net acceleration. The accelerometer can separate out the accelerations along a particular axis, unlike a human that would only sense the acceleration normal to the motorcycle. It seems to me that being able to measure the net acceleration, individual accelerations on perpendicular axis' and the fact that there is a two dimensional constraint should be enough to determine angle of lean.
-Phil
However, when watching road racing you can see the rider crawling all over to maximize the tire traction angles.
So three axis accelerometers plus a measured forward acceleration using a speed sensor are required.
Duane J
-Phil
1) Take a wine glass and glue it to your motorcycles gas tank in as much an upright position as possible.
2) Fill the wine glass half full of wine.
3) Now start riding your bike and get up to a constant speed.
4) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
5) Now take a turn in the road, without changing forward speed.
6) What do you see the wine doing? - It's not slopping forwards or backwards or sideways it's just sitting level in the glass as normal.
Clearly the wine is experiencing some extra acceleration downwards into the glass as you turn but that does not cause it to move. It sees no acceleration left or right or forwards or backwards.
Now replace the glass of wine with a three axis accelerometer. The only axis that will show any change is the vertical.
Measure that. From that calculate the centrifugal force and from that the angle of lean. Job done.
Is that going to get you an accurate result? No idea.
P.S. It's better to use a plastic wine "glass" in case of mishaps.
OK, if we get into that style things get harder, The wine in my experiment starts to slop out to the side.
We now have the C of G of the bike plus rider somehwere perhaps not within the bike itself. Certialy not directly above the tire patches in "bike coordinates".
-Phil
Gorga has 5 arms, so she's able to safely drive the motorcycle with two arms and have three left over to perform this experiment.
Gorga also has 5 eyes but, problem is, she can't see more than 2 feet in front of her, so Rich has to build her an instrumentation board so she can tell what she's doing as she speeds around in a bunch of big circles.
The board has accelerometers attached at each clock position in front of her - at 9 o'clock, 10 o'clock...etc to 3 o'clock. Each accelerometer can be turned on an axis nailed to the board and its measurement can be easily read by any of Gorga's 5 eyeballs.
So before Gorga takes off, Rich has her lean the bike over and take some readings. She is told to turn each accelerometer until it gives the highest reading, and then write down the value of each one. Sitting still and twiddling the accels around, she finally notes that all accels are giving their highest reading when pointing straight down toward the center of the earth and each reading is 1 G. She notes the angle of the accel readings vs. the position of the motorcycle, and everything looks fine.
But now Gorga takes off, wind in her hair, and drives around in a big circle. As instructed, she twiddles and fidgets with each accel until each one is pointing in the direction of its highest accel level. To her surprise, all the accels are pointing in the same direction, a vector normal to her seat. No matter where the accels are located on her board, they all have their maximum value in the same direction.
The question is, how would she now be able to extract from any of her accels the vector responsible for the centrifugal effects? Or what angle would there be to measure? She can't see very far, so she must rely only on her accels for information, but what would she be able to tell other than her accels are reading slightly more than 1 g?
I think we're disagreeing on "proper riding techniques" here, not the physics behind it. You're correct- it's possible and even very common to bank without applying throttle. But when you do so, you begin to turn. And when you turn, you decelerate, given constant throttle. That causes a decrease in centripetal force, which in turn causes you to lean more- which causes you to decelerate more. (It's the exact same principle as when you roll a coin across a table. Once it "leans" and goes into a spiral, the spiral gets smaller and smaller until the coin flops over. There is no force causing the coin to accelerate and maintain the same path.) Most folks do it the old school way- they shift their weight back and forth through a turn as needed, or they push more/less as needed without adjusting throttle. But in motor cycle riding courses, the instructor will always teach you to apply some constant amount of throttle through a curve and adjust your "push" without shifting your weight at all. The increased stability outweighs the increase in velocity for safety.
When you're traveling at a high velocity (for example 60 mph) it takes a surprising amount of force to "push on the right handle bar" and make the handlebar rotate a large amount of degree. Fortunately, at a high speed like that, it takes a very small degree of rotation to cant you into a lean and start turning. So again- you're correct, but in practise it's almost impossible to "over-push" because of the physical force needed to rotate the handle bars enough to cause the reaction you've mentioned. If applying throttle would cause you to lose control, well.... you shouldn't have been going that fast to begin with.
You're dead on correct about braking in a turn. Never, ever do it. Straighten out, then brake.
@Duane - I remember a lot of the harley crowd likes to say "Pull left go right" which is the same thing- but they say it that way because a lot of those types of bikes put the rider in a laid back position (thinking about hogs with the Ape Hangar handle bars and the like) and it's much more difficult to "push" when your arm is already extended outward as far as it can go. With bikes like that, it's easier to 'pull left' to go right than to 'push right' but the physics is the same. By pulling, they don't have to lean their body forward in order to get one arm bent a bit so that it can push forward.
The speed of the cycle and anglular rotation of the handlebars can definitely be used to determine the lean angle. At 70 mph, tackling a typical interstate curve to the right will cause your handlebar rotation to be about 4 or so degrees. I'm basing that on remembering visuals of it when I went through rider training courses. So obviously, 4 degrees is not exact. (We had to take 3 courses to get a permit to have a Goldwing or similar bike on base back in my military days.)
With saddle bags unevenly loaded.
Racers are maximizing cornering speed.
And then there is the wind pushing sideways.
All this influences the bike angle.
Duane J
I don't have enough drugs at had to partake of the Gorga experiment but...
The point is that a single accelerometer measuring g vertically with respect to the bike will measure 1g when the bike is traveling in a straight line at constant speed.
It will measure something greater than 1g if the bike is in a turn at constant speed.
Assuming the rider is not hanging off the side of the bike racer style that can be used to get the centripetal component of the acceleration and hence the angle of lean. If you also know the forward speed.
Things get harder if the bike has forward acceleration in the curve, or if the rider has shifted his weight over to the side.
In the old days we used to measure the angle of lean by how much metal was scraped off the foot rests by the road surface. BMW owners used to do it by seeing how much of the cylinder heads of their boxer twins were missing, in some cases leaving the valves exposed!
Ah, but that presumes the road surface is not tilted one way or another with respect to the horizon. Allow me to illustrate.... oh, uh, but first, let me introduce you to my good friend Phreandria...
I can tell youI've come real close to losing a passenger off the back from accelerating too fast. Wouldn't the same hold true with the wine glass?
I'd like to just go with your signature and start testing
I've come close to posting the same observation several times.
A practical example of this is experienced by anyone who has ever obtained a private pilot license. Your instructor will put a visor on your head that allows you to view only the instrument cluster. Then he puts the plane in a crazy mixed-up attitude and finally turns control over to you. Your senses tell you one thing and your instruments tell you another. The average student's' first reaction is to trust his sense of up and down and disregard the instruments. Big mistake.
BTW, I love the pic of ElectricAye's good friend Phreandria! Never seen a prettier specimen of Amanita muscaria.
You showed good judgement and restraint, thus you didn't post. But me, I've never been hampered by such considerate discretion. THAT'S why I have twice as many posts as you, DD.
Quantity before quality, I always say.
Foggles - Unusual attitude recovery, teaches you to look at the instruments and recover accordingly, usually throttle, bank then pitch. When they teach you not to trust your senses is when you cannot see outside or inside, then you let the airplane fly and steer with the rudder and throttle only
On that note, I'll probably be testing some theories before commenting any further. Some people here I'm sure have some degrees related to the subject, however I think that in the real world the way you apply your knowledge to a situation may blind you. From what I've read so far there has been no baseline or control for what anyone is talking about.
What kind of bike? Is it accelerating? Flat road? etc etc.... There are so many details left out I feel like I'm going to have to bail on this one.
You are correct, the app uses the accelerometer, gyroscope, and GPS sensors in the iPhone to detect the lean angle calculation. Here are a few articles on how it works:
http://eatsleepride.com/c/18976/lean_angle_calculation_on_the_eatsleepride_app
http://eatsleepride.com/c/15299/get_your_knee_down_collecting_data_for_lean_angle
No external hardware required, all on your phone
I would ask Gorga to rotate one of the accelerometers until it read exactly one gee and take note of it's angle and which side it was on relative to where it read the maximum value.
Maybe I'm not understanding what you mean by taking "note of it's angle and which side it was on relative to where it read the maximum value," but if I understand you correctly, then I believe that it will not matter which side of the board any of the accelerometers are on - they will all see the same maximum acceleration vector (let me call it the Big A) pointing in the same direction, which is a vector normal to the plane of the motorcycle seat. Tilting the accels this way or that will simply cause them to see only a portion of that Big A acceleration vector. For example, you could tilt the 9 o'clock accel until it sees 1 G, but that tilt could be performed either clockwise or counter clockwise by the same angle around its nail. If you tilted the accels so they are 90 degrees to the normal of the seat, then they would read zero. In other words, there would be no difference what the 9 o'clock accel sees vs. what the 3 o'clock accel sees. They all see the same Big A vector, and tilting them only causes them to read a component of the Big A.
Again, I'm not saying it's impossible to derive tilt angle from a single accelerometer under some circumstances - I'm just saying it's not possible with a single accel to tell the difference between the > 1 G value caused by a turn vs. a > 1 G value caused by a vertical acceleration. Also, a single accel could not tell you in which direction you are turning. You would need a gyro or some other reference system for that.
Place a two axis accelerometer on a merry-go-round such that one axis is pointed at the center of the merry-go-round (A) and the other is pointing straight up (Z). I think we can agree that the while the merry-go-round is stationary that the Z axis will read one gee and the A axis will read zero gees.
Now, spin the merry-go-round. Obviously the A axis reading is going to increase but what about the Z axis? I assert that the Z axis will remain at one gee.
Now lets rotate the accelerometer so that the Z axis is at it's highest value. Let's say it was rotated five degrees. That will be the angle that you would have to stand in order to not fall off the merry-go round. The difference between the max gee and one gee readings will tell you the lean angle.
unless... the merry-go-round is on the back of a flat bed truck and the truck is moving up and over a small hill. The Z axis is no longer one gee and you can't tell if it was the hill or the tilt that changed the Z reading..
The problem is fundamentally the same as trying to fly a plane in clouds - it needs an artificial horizon and gyros are very cheap nowdays.