Searching \ for '[EE]: Looking for method to differentiate gravity' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/index.htm?key=looking+method+differentiate
Search entire site for: 'Looking for method to differentiate gravity'.

Exact match. Not showing close matches.
PICList Thread
'[EE]: Looking for method to differentiate gravity '
2005\08\31@110038 by Gus Salavatore Calabrese

face picon face
I have been designing a vehicle tail light which would start flashing  
at a slow rate
as the brakes were applied and at a faster rate as the brakes were  
applied harder.
Ideally this unit would be standalone except for power ( which might  
be solar provided ).

I see no way to differentiate between braking and going downhill. Any  
ideas ?




Augustus Gustavius Salvatore Calabrese 720.222.1309    AGSC
http://www.omegadogs.com   Denver, CO

Disclaimer: The above statements are not intended as an endorsement of
any kind and any inference of having any verifiable knowledge about
anything referenced above is purely coincidental. ( we hope )

If you are not part of the solution, you are precipitate.





2005\08\31@111946 by David Van Horn

picon face

> I see no way to differentiate between braking and going downhill. Any
> ideas ?

Good luck. That's one of those basic physics problems. You can't
differentiate gravity from acceleration (negative, in this case).

You can measure pressure in the brake lines though, which should do the
job.




2005\08\31@112240 by William Couture

face picon face
On 8/31/05, Gus Salavatore Calabrese <spam_OUTgscTakeThisOuTspamomegadogs.com> wrote:
>
> I see no way to differentiate between braking and going downhill. Any
> ideas ?

None whatsoever.  This is the basis of Einstein's General Theory of
Relativity.

Bill

--
Psst...  Hey, you... Buddy...  Want a kitten?  straycatblues.petfinder.org

2005\08\31@115127 by Mark Rages

face picon face
On 8/31/05, Gus Salavatore Calabrese <.....gscKILLspamspam@spam@omegadogs.com> wrote:
> I have been designing a vehicle tail light which would start flashing
> at a slow rate
> as the brakes were applied and at a faster rate as the brakes were
> applied harder.
> Ideally this unit would be standalone except for power ( which might
> be solar provided ).
>
> I see no way to differentiate between braking and going downhill. Any
> ideas ?

I brake when I go downhill.

Regards,
Mark
markrages@gmail
--
You think that it is a secret, but it never has been one.
 - fortune cookie

2005\08\31@130125 by John Pfaff

picon face
Check the state of the brake lights?

William Couture wrote:

{Quote hidden}

2005\08\31@130233 by John Pfaff

picon face
Check for a difference in altitude or elevation between the front and
the rear?

Mark Rages wrote:

{Quote hidden}

2005\08\31@130357 by David P Harris

picon face
David Van Horn wrote:

>>I see no way to differentiate between braking and going downhill. Any
>>ideas ?
>>    
>>
>
>Good luck. That's one of those basic physics problems. You can't
>differentiate gravity from acceleration (negative, in this case).
>
>You can measure pressure in the brake lines though, which should do the
>job.
>
>
>
>
>  
>
Surely you are braking when the brakelight is turned on, which you can
detect, since you are attached to the brakelight - or am I totally out
to lunch?

David


2005\08\31@131124 by M. Adam Davis

face picon face
You need to detect the orientation of the car using something other
than acceleration.  I suspect your best chance is with a solid state
gyroscope.  If you place it so the axis of rotation is parallel to an
axle then you should be able to tell, after integrating the output,
the angle of the car (solid state gyros give rate of rotation, not
rotation).  Then you can use a simple one or two axis accelerometer to
detect acceleration/deceleration.

Rather than putting in constants (G=1) I would make the unit self
calibrating, and would use a two axis accelerometer (one up and one
forward, or both forward with one up 45 degress and one down 45
degrees).  When the car is stopped you can use the two accelerometers
to determine the initial angle of the vehicle, so you can park on a
hill and still expect the unit to perform correctly when you start up
the next day.  This is easy to determine regardless of the
gravitational pull in any given area.

You will experience two interesting sources of error:
1) Sensor drift
2) The earth is round, and after a few hundred miles the gyro will
report you're going downhill

These can both be managed by using a very good integration algorithm
or use an external analog integrator on the output of the sensors and
re-calibrating the device on each stop.  You should be able to
determine when you stop if you've been tracking all the variables
closely enough and the drift isn't too bad.  However, you may also be
able to tell the difference in vehicle vibration between a moving
vehicle, a stopped but still running vehicle, and a vehicle with its
engine off (important for energy savings anyway).  When the
acceleration algorithm says you are close to zero speed, check the
vehicle vibration and if you really are stopped check the angle using
the accelerometers and zero everything out.

You don't need to know how strong the gravitational pull is in a given
location if you have two accelerometers 90 degrees apart.  However,
this value may play a part in you calculation of acceleration, so you
may need to calibrate that automatically as well when you stop.

Sounds like a fun project.  Good luck!

-Adam

On 8/31/05, Gus Salavatore Calabrese <EraseMEgscspam_OUTspamTakeThisOuTomegadogs.com> wrote:
{Quote hidden}

> -

2005\08\31@133139 by PicDude

flavicon
face
Might be simpler to measure the brake pedal position directly, or use a
pressure sensor.

Cheers,
-Neil.



On Wednesday 31 August 2005 10:00 am, Gus Salavatore Calabrese scribbled:
{Quote hidden}

2005\08\31@133502 by Bob Blick

face picon face

> I see no way to differentiate between braking and going downhill. Any
> ideas ?

Since you specified you wanted a standalone unit, I'd say you need two
sensors - an accelerometer and also a tilt sensor in order to subtract out
gravity.

Make sure you get the right kind of tilt sensor, many of them are really
accelerometers, you don't want that kind.

Unfortunately this device is not going to be very accurate unless you can
predict the amount of tilt in the suspension caused by braking, and that
will vary considerably.

Cheers,

Bob


2005\08\31@135517 by Herbert Graf

flavicon
face
On Wed, 2005-08-31 at 11:22 -0400, William Couture wrote:
> On 8/31/05, Gus Salavatore Calabrese <gscspamspam_OUTomegadogs.com> wrote:
> >
> > I see no way to differentiate between braking and going downhill. Any
> > ideas ?
>
> None whatsoever.  This is the basis of Einstein's General Theory of
> Relativity.

Yes, there is a difference, it's just really hard to measure.

Force due to acceleration is constant along the direction of the
acceleration.

Force due to gravity changes dependant on the distance between the two
objects.

In most cases the difference is too small to measure, but that doesn't
change the fact that there IS a difference, and it CAN (and has) been
measured.

TTYL


-----------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

2005\08\31@142638 by Michael Rigby-Jones

picon face


>-----Original Message-----
>From: @spam@piclist-bouncesKILLspamspammit.edu [KILLspampiclist-bouncesKILLspamspammit.edu]
>Sent: 31 August 2005 18:35
>To: Microcontroller discussion list - Public.
>Subject: Re: [EE]: Looking for method to differentiate gravity
>from acceleration
>
>
>
>> I see no way to differentiate between braking and going
>downhill. Any
>> ideas ?
>
>Since you specified you wanted a standalone unit, I'd say you
>need two sensors - an accelerometer and also a tilt sensor in
>order to subtract out gravity.
>
>Make sure you get the right kind of tilt sensor, many of them
>are really accelerometers, you don't want that kind.

Are there tilt sensors available that would work in this application?  Any system I can imagine would be affected by both gravity and acceleration to some extent.

Regards

Mike

=======================================================================
This e-mail is intended for the person it is addressed to only. The
information contained in it may be confidential and/or protected by
law. If you are not the intended recipient of this message, you must
not make any use of this information, or copy or show it to any
person. Please contact us immediately to tell us that you have
received this e-mail, and return the original to us. Any use,
forwarding, printing or copying of this message is strictly prohibited.
No part of this message can be considered a request for goods or
services.
=======================================================================

2005\08\31@143602 by M. Adam Davis

face picon face
What exactly is the difference, then?  Or, more appropiately, how do
you meaure one without measuring the other?  Gravity is a force.
Acceleration is a result of force.  Acceleration may be caused by
gravity or some other force, or some combination.  Implicit to the
definition of both is the concept of mass.  You cannot have gravity or
acceleration without mass.  Therefore we measure both gravity (the
force), and acceleration (the result of force over time) using mass.
I can see that since acceleration involves the variable time it may be
possible to differentiate between the two, at least on paper.

How does one measure either acceleration or gravity without involving
the other?  Or can you point me to research where the difference has
been measured?

I'm very interested in this, it has some implications for work I am involved in.

-Adam

On 8/31/05, Herbert Graf <RemoveMEmailinglist2TakeThisOuTspamfarcite.net> wrote:
{Quote hidden}

2005\08\31@145127 by Herbert Graf

flavicon
face
On Wed, 2005-08-31 at 14:36 -0400, M. Adam Davis wrote:
> What exactly is the difference, then?  

As I explained, the force due to acceleration is constant along the
direction of the acceleration, while gravity is not.

Consider a rocket with a force meter on the nose, and one on the tail,
and the rocket is on the earth pointed straight up. On the pad the two
meters would differ in reading, since the one on the tail is closer to
the earth then the one on the nose.

Now consider the same rocket in an environment devoid of any other mass
(zero gravity). Light that rocket up. The force on the two meters would
be the same.

> Or, more appropiately, how do
> you meaure one without measuring the other?  

AFAIK you can't, and I didn't say you could, force is force.

{Quote hidden}

I don't think you can, all I said is that the force caused by each is
different.

Off hand, if you know the mass of your object and the mass of the Earth,
any force difference you read between the tail sensor and the nose
sensor could be inferred to be this part gravity and this part
acceleration (since you'd have something between the linear line of
force due to acceleration and the curved line of force due to gravity),
but I think that would only work when you know the difference in
direction of your travel vs. the direction of gravity.

I'm certainly no expert though, it's possible someone out there much
more cleverer then me has come up with a way to do it, or at least a
theoretical way of doing it.

TTYL

-----------------------------
Herbert's PIC Stuff:
http://repatch.dyndns.org:8383/pic_stuff/

2005\08\31@151114 by M. Adam Davis

face picon face
Ah, of course - I should have recognized that from your first post.
Still impractical outside of a lab, alas.

-Adam

On 8/31/05, Herbert Graf <TakeThisOuTmailinglist2EraseMEspamspam_OUTfarcite.net> wrote:
{Quote hidden}

> -

2005\08\31@160928 by Oscar T.

picon face
2005/8/31, Gus Salavatore Calabrese <RemoveMEgscspamTakeThisOuTomegadogs.com>:
>
> I see no way to differentiate between braking and going downhill. Any
> ideas ?
>
 I think you are meaning to differentiate between braking to
decelerate (slow down) and braking to non accelerate going downhill
(maintain speed). If so you only need an one axis acelerometer,
Freescale have same ones you may even ask for a sample.

Oscar

2005\08\31@193942 by William Chops Westfield

face picon face
Isn't "x-axis" acceleration due to typical hills liable to be MUCH
smaller than acceleration caused by  even gentle breaking?  Maybe
you can get by with a simple threshold.  Have you measured the sorts
of values that actually occur instead of getting hung up on theory?

BillW

2005\08\31@201738 by William Chops Westfield

face picon face
On Aug 31, 2005, at 4:39 PM, William Chops Westfield wrote:

> Isn't "x-axis" acceleration due to typical hills liable to be MUCH
> smaller than acceleration caused by  even gentle breaking?

(and in fact the "databrake" Jinx mentions only does anything for
"hard" baking.  Victory Dance!  ($198.  Wow.))

As for driver habits, the idea is not so much to alert the person
behind you exactly what is happening, but to do something unusual
so that they have a clue that SOMETHING is happening.  It wouldn't
work if they were following too close, but might work very well
indeed for those cases where the following distance is relatively
large (and the driver is therefore paying LESS attention to the
car in front of them.)  It's already recommended (?) that you
flash your brakes when you're required to slow down to a rate
(or AT a rate) atypical for the environment you're in...

BillW

2005\08\31@223022 by Peter

picon face


On Wed, 31 Aug 2005, Gus Salavatore Calabrese wrote:

> I have been designing a vehicle tail light which would start flashing at a
> slow rate
> as the brakes were applied and at a faster rate as the brakes were applied
> harder.
> Ideally this unit would be standalone except for power ( which might be solar
> provided ).
>
> I see no way to differentiate between braking and going downhill. Any ideas ?

The magnitude of the acceleration vector is much larger when
decelerating/accelerating (uphill) than g. If you measure the magnitude
of the vector (regardless of the direction) you can threshold it at say
5% and get pretty reliable detection imho. So if you use a 2 axis sensor
and rms the output of the two axes you should have a good signal to work
with.

Peter

More... (looser matching)
- Last day of these posts
- In 2005 , 2006 only
- Today
- New search...