The Answer My Friend...

OK. Here's one. If a car is doing a 1G turn, then its driver is not in control of it.

Yes - it's a robot car.

Robots do not need to be in control of cars; very few robots hold driving licences.

Why would the robot need to open the window, though?

To get directions.

So he could put his arm on the window sill and show off his Rolex.

In this case the "force of the wind" would be called "drag". You could "measure" it by running multiple tests in various wind conditions, as the drag will have some effect on acceleration...not much but some.

Yes, obviously. My question is, would my (MEMS) accelerometer detect it? For instance, if the car is sitting still, does the accelerometer detect the force of wind blowing on it.

No, a MEMS accelerometer cannot detect wind directly. It should detect the net acceleration effect from all forces on an object, one of these forces can be from wind. Now one certainly can design an anemometer (wind speed instrument) where the sensor is a MEMS accelerometer. This probably would not be a good approach to build such an instrument because a mathematical derivation would have to be performed but it could be done.

Yes if the robot has opened the window. _{Oh, the power of positive thinking.}

I was answering your original question..."For instance, should I expect a different measurement of acceleration doing a 1 G turn in a crosswind vs the same turn in no wind (assuming velocity is the same)?" Standing still is a different question and the answer is obvious in the name of the instrument.

Doesn't it only detect what's relevant to the car, and that it can't determined or separated any outside forces to the car, only what the car is actually doing. unless you have more feed back as to what the car is doing internally and its output, suchas as engine RPM drivetrain, then its only math.

The forum doesn't know whether or not the window is open, though.

Does the wind cause the car to move?

Car moves: yes.

Car does not move: no.

If the window were open, the accelerometer wouldn't be able to tell if it were the car moving it or the wind moving it.

If it moves, it does not matter.

Then shut the window.

I'm not in the car.

this probably could.

https://www.parallax.com/product/28017

My what a cute doggie!

The accelerometer is not an anemometer, but it might respond to wind buffeting of the car, when sitting idle or when already in motion, depending on the magnitude of the buffeting.

Do you also have this on a data logging device in the car's computer, or attached to it? Are you seeing accelerations when you know you are not out driving the car? Maybe there are "gremlins" out and about jumping up and down on the car while you are asleep.

Look for tiny sneaker tracks (not sure which brand is popular right now with gremlins in the age bracket from 7-14 years of age. Older gremlins will leave much larger tracks, bullet holes, and sneaker dents.

You're right, the answer's no. But I had to think for a few moments about gravity. These things are used (or were, in UK before rolling roads came in) to measure brake effectiveness. So would be aligned fore and aft in the vehicle. If it were turned at right angles or thereabouts it would measure centrifugal acceleration going round a curve. If you turned it on its side it would measure acceleration due to gravity, but there's not much point as we already know what that is.

That's my take on it anyway.

Thanks. I actually use gravity as a test input to my (longitudinal) accelerometers, just turning them 90 degrees gives a 1 G input. Very handy for calibration.

It depends upon location. If the accelerometer were in Bogota, then it would calibrate differently compared to if it were in Reykjavik, as the Earth is an oblate spheroid and gravity varies across is surface.

Yes, but if I ever need to know my sensors to that level of accuracy I'll just kill myself.

Ergo. If you have the accelerometers why don you simply set up a test to prove or disprove your theory and findings?

Mainly because my accelerometer and I aren't co-located at the moment. But it seems like that's what I'll have to do.

Yes, good point, i hadn't thought of that.

Why wouldn't it? Imagine the car in neutral with the parking brake off. A strong enough wind would move car and accelerate it over time.

Are not strain gauges a form of accelerometer? If a strain gauge was mounted to a door panel it should detect deflection given enough sensitivity and strong enough wind.

What kind of robot leaves a car in neutral, with the handbrake off and the window open? The car will get TWoC'ed!

What kind of robot leaves a car in neutral, with the handbrake off and the window open?

a robot with its accelerometer in the shop maybe?......

Schrodinger's robot?

The accelerometer will not detect the wind because it is inside the vehicle. I believe wind tests are done in a tunnel with force gauges measuring the forces on the object, a car on a platform and the stresses transferred to the platform are measured.

The accelerometer only measures changing direction forces. In a curve the acceleration will be dependent upon the radius of the curve only, a blowing cross-wind can affect the radius which would affect the acceleration.

Drew K

If there is wind inside the vehicle, then would the robot need to open the window?

Simple, direct answer: the accelerometer will detect the movement of whatever it is attached-to, at the point of attachment. I.e., the same accelerator attached to the underside of the frame of a tall SUV will detect a different amount of movement than if it were attached near the interior domelight inside the same SUV.

Here's a real-life instance: When I drive my SUV through twisty mountain roads I sometimes get a touch of nausea from the motion. When I drive my sports car on those same roads, I'm fine. The difference is the lower seat position and lower center of gravity of my sports car means my inner ear is not affected as much by the body sway as when I'm in my SUV.

<...sometimes get a touch of nausea from the motion...> Open the window, then.

So I can barf down the side of my car instead of into my lap? Yeah, good idea, I suppose.

It's all a question of relativity, which surely ralfcis can clear right up for you.

just don't

A strong gusting crosswind could show an unexpected variation in the g-force...so could a skid...but I think this depends on the speed and design of the vehicle, road surface angle, and other contributing factors....

Is this a three, orthogonal axis accelerometer?

Yes, a MEMS 3 axis accelerometer.

No. An accelerometer measures change of velocity, either magnitude or direction. If the wind does not affect the movement of the car, it will not be detected by an accelerometer.

As I understand it, the question is can it differentiate the wind affect to the car's velocity.... which it can't.

Well, yes it can.

A change in velocity is due to acceleration. It can measure the acceleration which will indicate a change in velocity.

Yes, but as I understand it, can it differentiate between the cause of the change of velocity between how much is due to the wind and how much due to the car.

Or am I reading way too much into this.

The phrasing by the OP is ambiguous on this and so very many other ways it is hard to tell. An automobile with enough lateral traction that it can stay on the road during a 1G turn will likely be so streamlined that gale force winds might not have any effect regardless of who or what is driving the sports car.

Somewhere, I've got some lateral acceleration data from my time in a race car.

I'll look for that later and post the numbers.

Nothing like dissecting the ambiguity and find out, who knows what'll become of it.

That's the spirit!

Even negative results are results

Let me try to be more precise. Let's say that I'm designing an intelligent suspension system. I'd like that suspension system to be able to adjust to forces acting on the car. One class of such forces is centrifugal force going around a turn. Another is lateral force from accelerating and decelerating. Both of these can be detected by the accelerometer independent of moving the vehicle's body. (Einstein's famous elevator thought experiment for example.) A third is traveling perpendicular to a cross wind, or the experience we've all had of sitting still and having a large truck zoom by. Will a MEMS accelerator detect the force of the wind, or can it only react to the movement of the car's body in this instance?

It can't detect wind force directly, only if the wind causes change of speed. E.g. if driving along and a head wind occurs, as long as the driver opens the throttle to maintain constant speed, the accelerometer does not detect it. If however engine power is kept fixed so the car slows, it detects that.

I thought your original post was clear enough. Some posters have got hung up on the 1-g turn, which would be pushing it for adhesion etc, but I assume you just want to know whether the accelerometer detects turning, the 1-g being an arbitrary figure (not well-chosen )

Also if stationary on a hill, the accelerometer reading can be used to determine the slope.

Yes.

These intelligent suspension systems exist today and have done for many years. High speed trains use these intelligent suspensions for banked corners and it has been common on cars for a long while.

You are not asking about wind speeds but should be asking about compression and decompression of air from one side, when looking at passing vehicles, as wind speed and compression is two different dynamics. i.e. High speed trains entering tunnels may be the place to discover your answers, along with turbulence in air.

Physicists describe turbulence as "turbulent flow," or the movement of a gas or liquid in which the fluid undergoes irregular mixing, causing changes in the fluid's speed, pressure and direction.

Can it differentiate between the cause of the change of velocity? No. Lateral acceleration may not change the velocity.

But it can't differentiate between the cause of acceleration either.

OP was asking about a single variable, wind. All else being equal and wind the only variable the accelerometer can detect the difference.

So, who knows?

Exactly, No reference point really.

Just to pick a point, speed and velocity are different

Acceleration is by definition a change in velocity

How did speed enter into this discussion?

ds/dt with no particular place to go, or are you funnin' me?

Is the window really open if only a robot is there to see it?

An accelerometer attached to the seat might detect the wind, usually sub-1kHz and sometimes almost zero Hertz flat . . . . fluctuations.

You meant flatulent asians?

Umpteen possible forces could impinge on the car and accelerate it. But no internal meter can tell one of these forces from another. (Wind, vehicle impact, braking, steering, acceleration, collisions with animals, plants, or minerals, etc.) The whole question is completely ridiculous.

I disagree, this is not a ridiculous question. It is a very telling question what one has learned in class and can they apply it in a proper fashion.

Oh, well, from that point of view....

I don't recall if it was motorcycle or vehicle, but I was reading in a magazine about how tyres were getting better and approaching 1 G forces in turns. But I don't think that is the question here, it was a statement.

The question is will an accelerometer detect wind. The "for instance" on the 1 G turn in a crosswind is just an example. If you make the paramater of a 1 G turn, then the turn will be 1 G no matter the wind or speed or anything else, you just keep turning till you hit 1 G.

The answer to the original question is overall no, the accelerometer cannot by itself detect wind. Now if it were tied into measuring devices that detected steering wheel position, force on the steering wheel and some sort of positioning sensors then when a crosswind pushed the vehicle causing it to leave a specified path causing the positioning sensor to correct the steering then the accelerometer would detect the correction.

As the OP stated, it will also detect the shaking and swaying due to the wind blowing the body around.

Drew K

My point was the ambiguity of the sparse information we actually have been provided. Yes, the 1G turn was a "red herring" of distraction. This can also be said about the robot pilot. We don't even know if the OP wonders if an effect from wind on this accelerometer could be a possible source of error or if a moving weather station can correct vehicle movements from wind speed data.

Skid Pad Test Corvette vs 911

(g) 1.08 1.04

Interesting discussion, but no amount of guessing and opinions will alter the fact that the only thing that an accelerometer measures is the rate of change of the summation of forces acting on the mass that it is attached to. Without independent data about the forces and inputs acting on that mass then it is impossible to resolve the source of those forces are nor where they acting. Mathematically it's simply a system of N equations with N+1 variables; unsolvable without assumptions about the extra variable.

To answer the question directly, knowing the wind velocity (Magnitude) without knowing its direction (Angle), does not allow the accelerometer to determine its affect on the vehicle.

You'll have to use a manometer with trending capability along side of the track your testing on. You must be able to time correlate it exactly with the accelerometer readings during different wind conditions to see if there is an effect or not.