Post by thread:[EE]: Looking for method to differentiategravityfr

IOf you use a 3 axis device and do the math it will work. Think vectors. At rest the system will measure a 1 G down. When breaking a breaking acceleration will be added in as another vector so you should be able to remove the 1 G and get a resultant vector.Team 190 from WPI did that a few years ago on their First robot when we put in an inertial navigation system. Larry > > From: "David Van Horn" <spam_OUTdvanhornTakeThisOuTmicrobrix.com> > Date: 2005/08/31 Wed PM 04:21:45 EDT > To: "Microcontroller discussion list - Public." <.....piclistKILLspam@spam@mit.edu> > Subject: RE: [EE]: Looking for method to differentiate gravityfromacceleration > > > 2) If the car is on a banked surface then the device may think the car > > is going downhill since the perpendicular will read less than 1G. In > > this case the light would not come on if you did brake slowly. A > > three axis accelerometer would remedy this. Might only matter at the > > race track and some highway entrance and exit ramps. > > Would it work in San Francisco? > > :) > > > > > --

On Wed, 2005-08-31 at 18:26, David Van Horn wrote: > Well, if you manage to exceed 1G on a flat road, let me know. :) > AAFD class drag race cars do .25 mile standing start in about 4.7 seconds, reaching a peak speed of >300MPH, exceeding 4G for some of the race, on a flat road. -- Gordon - N1MGO n1mgoKILLspamarrl.net

> On Wed, 2005-08-31 at 18:26, David Van Horn wrote: >> Well, if you manage to exceed 1G on a flat road, let me know. :) > AAFD class drag race cars do .25 mile standing start in about 4.7 > seconds, reaching a peak speed of >300MPH, exceeding 4G for some of > the > race, on a flat road. True enoughish. But they are 'glued' to the track surface with a very special expensive dynamic glue. It's stored as a solid on the surface of the big black round things that you see on either side at the back. It's converted to semi liquid glue at the surface layer shortly before 'launch' and during the very few seconds that it is required to glue the car dynamically to the track by application of heat, generally applied by the engine causing friction between the glue dispensers and the track surface. Said glue sticks aka tyres are rather expensive and don't last many races. Suitable dispensers for road use tend to be far cheaper and are usually not used in this mode - classic friction is generally easier on the tyres. Some road vehicles can sometimes manage more than 1g either accelerating or under brakes - but it takes lots of good design, good equipment, skill and luck. RM

> > Well, if you manage to exceed 1G on a flat road, let me know. :) > > > AAFD class drag race cars do .25 mile standing start in about 4.7 > seconds, reaching a peak speed of >300MPH, exceeding 4G for some of the > race, on a flat road. Sort of makes my point.. Aerodynamics, ground effects, and huge engines required, which you won't find on the family grocery getter, or even the local "2-fast-2-furious" machines.

Russell, On Fri, 02 Sep 2005 00:55:10 +1200, Russell McMahon wrote: > Some road vehicles can sometimes manage more than 1g either accelerating or under brakes Far more under the latter than the former... > - but it takes lots of good design, good equipment, skill and luck. Actually any reasonably good modern car should be able to manage 1g braking - I can't remember what the MOT tests here specify as a minimum, but I think it's 0.3g - below that it fails the test as unsafe, so it's expected that most cars will be well above this. Cheers, Howard Winter St.Albans, England

It's too bad people have seemed to miss this post because I think that's the only solution for a stand-alone device. On 8/31/05, .....mchipguruKILLspam.....charter.net <EraseMEmchipguruspam_OUTTakeThisOuTcharter.net> wrote: > IOf you use a 3 axis device and do the math it will work. Think vectors. At rest the system will measure a 1 G down. When breaking a breaking acceleration will be added in as another vector so you should be able to remove the 1 G and get a resultant vector.Team 190 from WPI did that a few years ago on their First robot when we put in an inertial navigation system. > Larry

> Actually any reasonably good modern car should be able to manage 1g > braking Not consistently throughout a whole braking manoeuvre without great care in design. Being able to lock up the wheels is only the first step. Stopping them doing it is the second and harder step. Well designed ABS helps. Without such weight transfer between wheels will leave the lightened wheel locking and the one with more weight on it not braking as hard as it could. RM

>>>>I'd suggest trying rate of change of velocity. >>>What a great idea! And here we were all wasting time talking about >>>measuring accelleration. >>One could reasonably infer, with a prospect of being in error, that >>he meant rate of rate of change, which may well be useful. > Jerk. Whether the effect of the magnitude of the second differential of velocity was sensed as a jerk would depend on whether the rate of rate of velocity change varied extremely rapidly, as in a panic stop, or whether the first differential of the deceleration has a more gradual slope as when eg one starts off braking gently and then gradually increases the braking as the gap shortens more rapidly than may have been intended, which of course only applies if one drives that way, as some jerks do. R :-) M

On Fri, 02 Sep 2005 17:06:07 +1200, Russell McMahon wrote: I had said: > > Actually any reasonably good modern car should be able to manage 1g > > braking > > Not consistently throughout a whole braking manoeuvre without great > care in design. Being able to lock up the wheels is only the first > step. Stopping them doing it is the second and harder step. Well > designed ABS helps. Without such weight transfer between wheels will > leave the lightened wheel locking and the one with more weight on it > not braking as hard as it could. I was sort-of assuming that any reasonably good modern car would have ABS, since my midrange ten-year-old one has it! Cheers, Howard Winter St.Albans, England

On Fri, 2 Sep 2005, Russell McMahon wrote: >> Actually any reasonably good modern car should be able to manage 1g braking > > Not consistently throughout a whole braking manoeuvre without great care in > design. Being able to lock up the wheels is only the first step. Stopping > them doing it is the second and harder step. Well designed ABS helps. Without > such weight transfer between wheels will leave the lightened wheel locking > and the one with more weight on it not braking as hard as it could. http://www.4xclub.co.za/TyreABC.htm Highest listed coefficient is 0.9 . How do you make it 1.0 ? I suppose that with 'glue' tyres (dragster) at low speed and with aerodynamics (pushing down on aerofoils) at high speed you could go a little beyond 1G for a short time. What am I missing. Peter