Searching \ for '[OT][WOT][RM] Stopping accelerations and reaction' in subject line. ()
Help us get a faster server
FAQ page: massmind.org/techref/index.htm?key=
Search entire site for: '[WOT][RM] Stopping accelerations and reaction'.

Exact match. Not showing close matches.
'[OT][WOT][RM] Stopping accelerations and reaction '
2005\03\29@051011 by

This is a marginal one - it's technical but I'm seeking opinions for
my interest    .

Otherwise to me directly apptech at paradise dot net dot nz

_______________

Imagine decelerating to rest in a car in two situations.
Dry level road in good condition in each case.
Identical vehicles in each case.

A.     All out maximum possible deceleration stop from 50 kph / 30
mph.

B.    Maximum rate at which it is REASONABLE to expect to have to stop
if a traffic light goes yellow as you approach it. ie this is the
hardest you think people would reasonably expect you to have to brake
to stop - any harder than this and you'd feel justified continuing.
You COULD easily stop harder but considerations of passenger comfort,
not having car behind run into you and general sane behaviour set this
as the upper limit.

QUESTIONS

1.    What percentage of deceleration would you expect in case B
compared to case A.
Some I have asked suggest 30%. Others more. I think about 50% BUT what
do you think ?

1b     AND/OR What deceleration rate do you think should apply in B.
(in m/s/s or ft/s/s)

(Actual results in A will depend on braking system, road conditions
etc. Typically they will *approach* 1g in ideal conditions for many
cars. Some few will exceed 1g due to special factors (sticky tyres,
air dams etc)

2.    What reaction time do YOU think is reasonable to respond to an
orange light. ie time from when light goes orange until you see it,
react and place your foot on the brake and braking just begins. If a
person is much slower on average than this maximum then they are in
danger of being incompetent on the road. If they meet this limit their
response is OK but faster would be better.

___________

I have some official design figures for the above parameters but I'm
greatly interested in what other drivers consider reasonable.

Reason:    My son recently wrote off the 3rd car of mine that has died
in his hands in the last year. First two were not his fault by any
measure (although if he'd noticed the oil light on number 2 ...).
Number 3 MAY be his fault - I have my opinions but I'm interested in a
cross section of opinions on reasonableness.

I'll probably post some pointers to related sites in due course.
There's more to intersection engineering than may at first be obvious.

Russell McMahon

On Tue, 29 Mar 2005 22:09:07 +1200, Russell McMahon wrote:

>...<
> QUESTIONS
>
> 1.    What percentage of deceleration would you expect in case B
> compared to case A.
> Some I have asked suggest 30%. Others more. I think about 50% BUT what
> do you think ?

I think a reasonable B would be about a third of A.  0.5g pushing against the seatbelt is quite a lot for your
passengers!

The question you didn't ask is whether a solo driver would be more inclined to stop in the marginal situation
than if they were carrying passengers, because of passenger comfort.

> 1b     AND/OR What deceleration rate do you think should apply in B.
> (in m/s/s or ft/s/s)

Say 3m/s/s

> 2.    What reaction time do YOU think is reasonable to respond to an
> orange light. ie time from when light goes orange until you see it,
> react and place your foot on the brake and braking just begins. If a
> person is much slower on average than this maximum then they are in
> danger of being incompetent on the road. If they meet this limit their
> response is OK but faster would be better.

I'd say 0.25s.  0.2 is a pretty good average for the population I think, and more than 25% slower than average
could be considered dangerous.  IMHO, of course!

I've always thought that rather than having blood-alcohol limits for driving, there should be reaction-time
limits (these could be checked by the car itself fairly simply) and this would eliminate people who shouldn't
be driving when stone-cold sober!  :-)

Cheers,

Howard Winter
St.Albans, England

> I'd say 0.25s.  0.2 is a pretty good average for the population I
> think, and more than 25% slower than average
> could be considered dangerous.  IMHO, of course!

You'd be surprised, possibly stunned, at real world reaction and
response times.

RM

Russell,

On Wed, 30 Mar 2005 00:28:42 +1200, Russell McMahon wrote:

> > I'd say 0.25s.  0.2 is a pretty good average for the population I
> > think, and more than 25% slower than average
> > could be considered dangerous.  IMHO, of course!
>
> You'd be surprised, possibly stunned, at real world reaction and
> response times.

Don't forget that in the situation you're describing, the driver should be primed, watching the light,
half-expecting it to change.  Agreed that in an emergency the time to recognise it and decide on an action can
be much slower than the plain "reaction time" that you'd get when expecting something, but approaching a
traffic light the driver *should* be expecting the it to change!

Cheers,

Howard Winter
St.Albans, England

> 2.    What reaction time do YOU think is reasonable to respond to an
> orange light. ie time from when light goes orange until you see it,
> react and place your foot on the brake and braking just begins. If a
> person is much slower on average than this maximum then they are in
> danger of being incompetent on the road. If they meet this limit their
> response is OK but faster would be better.

I have always felt that orange light times should match the "2 second rule"
for following another vehicle. Now precisely how this matches reaction and
stopping times, I am not sure.

Russell McMahon wrote:
> but I'm
> greatly interested in what other drivers consider reasonable.

I think this is an area where people's intuition is likely to be quite
different from fact.  Anything short of real measurements would just add to
confusion and be a real disservice.

*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com

>> > I'd say 0.25s.  0.2 is a pretty good average for the population I
>> > think, and more than 25% slower than average
>> > could be considered dangerous.  IMHO, of course!
>>
>> You'd be surprised, possibly stunned, at real world reaction and
>> response times.

> Don't forget that in the situation you're describing, the driver
> should be primed, watching the light,
> half-expecting it to change.

I'd alreay tried this

http://www.phy.ntnu.edu.tw/java/Reaction/reactionTime.html

Interesting.
0.32s absolute best. Typically 0.36 to 0.4 Easy to get higher.
Anything less than hair trigger waiting - say more like you would
really be when driving but taking care, takes me up into 0.5 - 0.6.

RM

2005\03\29@092028 by
> Russell McMahon wrote:
>> but I'm
>> greatly interested in what other drivers consider reasonable.

> I think this is an area where people's intuition is likely to be
> quite
> different from fact.  Anything short of real measurements would just
> confusion and be a real disservice.

Not at all.
The medium is the message in this case.
Disparate OPINION is what I am seeking, and the legal test, here at
least, seems to be reasonableness. I know how hard I would consider
stopping in a a typical orange light scenario. Whether my measure of
"reasonable" and the law's coincide is yet to be seen. The standard
used by the official book that everyone here uses to design
intersections is less rigorous than what I would have applied.

RM

>
>Disparate OPINION is what I am seeking, and the legal test, here at least,
>seems to be reasonableness. I know how hard I would consider stopping in a
>a typical orange light scenario. Whether my measure of "reasonable" and
>the law's coincide is yet to be seen. The standard used by the official
>book that everyone here uses to design intersections is less rigorous than
>what I would have applied.

I assume this is what I'm used to as a yellow, the warning that the light
I would not expect to have to brake hard, and I would expect that the
timing would be adjusted according to the speed limit. Braking hard to
avoid running a red, could cause a rear end collision, or damage/injury in
my vehicle.  Granted, I'll do what I have to do, but traffic signals
shouldn't force me into this.

When the light goes yellow, there should be plenty of time to react to it,
and to slow down in a reasonable manner.  Remember, the yellow is there to

Russell McMahon wrote:
> Not at all.
> The medium is the message in this case.
> Disparate OPINION is what I am seeking, and the legal test, here at
> least, seems to be reasonableness.

But my point is that what people imagine to be reasonable coming up with
numbers while sitting in front of a computer may be quite different from
what you get from measuring actions these same people consider reasonable
when directly performing them.  In other words, people are very bad at
describing their perception of reasonable, even then they have a good
understanding of what feels like when driving a car.

I could use my car to show you what I think is a panic stop, a reasonable
but stop to a suddenly changing light, and a slow stop.  However, I expect
any numbers I give you just imaginging this sitting in front of a computer
to be way off from the real values.  You want to know what people consider a
reasonable stop, not how much they err in reporting the figure.

*****************************************************************
Embed Inc, embedded system specialists in Littleton Massachusetts
(978) 742-9014, http://www.embedinc.com
Russell,

On Wed, 30 Mar 2005 01:51:01 +1200, Russell McMahon wrote:

> I'd alreay tried this
>
>         www.phy.ntnu.edu.tw/java/Reaction/reactionTime.html
>
> Interesting.
> 0.32s absolute best. Typically 0.36 to 0.4 Easy to get higher.

There's something wrong with his software - it shows a braking distance of 25.51m regardless of the speed!

> Anything less than hair trigger waiting - say more like you would
> really be when driving but taking care, takes me up into 0.5 - 0.6.

I managed 0.06 by "guessing" when the light was going to change... but by doing it properly I got 0.29.
Interestingly, if I looked "past" the light so I was using vision outside the centre of my eyes (I forget the
name for this) I got faster scores (0.26) than if I stared straight at the light!

Cheers,

Howard Winter
St.Albans, England

And don't forget the time it takes to physically move your foot from the
accelerator to the brake - let alone the processing/decision making delay -
which may, for some, take the longest period of all.

RP

{Quote hidden}

I'd alreay tried this

http://www.phy.ntnu.edu.tw/java/Reaction/reactionTime.html

Interesting.
0.32s absolute best. Typically 0.36 to 0.4 Easy to get higher.
Anything less than hair trigger waiting - say more like you would
really be when driving but taking care, takes me up into 0.5 - 0.6.

RM

My opinion:

1. The maximum deceleration is less than 1g in any case and should
*never* be used (assume someone is driving behind you).

2. The reaction time to a light, including taking the proper action,
should be much less than 2 seconds imho., and assume that people who are
slow reacting are not driving at the maximum allowed speed. I read
somewhere that traffic and signal lights must be visible for 100 meters
so someone driving at 50kph (~14 m/sec) and seeing the signal at 100
meters will go at most another ~28m before hitting the brakes full on
(in an emergency), and stop in another at most ~85 meters before
entering the junction (assuming no-one rams him from behind and that the
traffic light is recessed from the junction by a few meters).

Braking track length at 1g is (at 0.5g the track doubles):

50kph        14m/sec                98m
40kph        11m/sec                61m
30kph         8m/sec                34m
20kph         5m/sec                15m
10kph         3m/sec                4.5m

The formula used was x = v^2 / (2 * a)

Probably any acceleration above 0.5g is unpleasant for most people
(total g=1.11). City buses and other vehicles have acceleration limiters
set around such values afaik (they are overridden when the driver floors
the brake or the accelerator in an emergency).

Peter

On Tue, 29 Mar 2005, Olin Lathrop wrote:

> Russell McMahon wrote:
>> but I'm
>> greatly interested in what other drivers consider reasonable.
>
> I think this is an area where people's intuition is likely to be quite
> different from fact.  Anything short of real measurements would just add to
> confusion and be a real disservice.

I don't think that it would be very easy to set up an experiment to
duplicate real behavior. The easiest way would be to have pollsters
watch the time it takes drivers 'in' the 100 meters before the junction
to hit the brakes (looking at brakelights from behind) after a traffic
light changes color.

Peter
In australia i believe its 3 second rule.
but anyway that distance is based around "emergincy stop" + reaction time +
margin.
i dont think its based around "nice gentle stop"
In australia lights are yellow for a minimum of 4 seconds.
if you can "nice gentle stop" for them you should
so in other words the practical rule seems to be if you can get across by
anything other than downshifting *Punch it* otherwise people will madly
pullout from behind you and attempt to get across.

question. You are travelling at 110kph (70mph)
how long and how far is a "nice gentle stop"
and the same again for "emergincy stop"

b) work it out based on some form of "scientific" reasoning

> {Original Message removed}
i dont know about you but it dosent take me 98 meters to stop from 50kph
v^2 = u^s + 2as
s = v^2/2a
s = (14^2)/(9.8*2)
s = 196 / 19.6
s = 10 meters

I regularly practise my 50kph emergency stop (when was the last time you
practised something that could save your life)
in my 1985 ford telstar i have it down to a bit over 2 parked car lengths
(well a car and a van, eh what can i say they always park next to each
other) now that is with the advantage that i know its coming and i begin
braking at the start of the first car.

that to me agrees fairly well with the "theory"

> {Original Message removed}
Peter,

On Tue, 29 Mar 2005 21:29:23 +0200 (IST), Peter wrote:

> Braking track length at 1g is (at 0.5g the track doubles):
>
> 50kph        14m/sec                98m
> 40kph        11m/sec                61m
> 30kph         8m/sec                34m
> 20kph         5m/sec                15m
> 10kph         3m/sec                4.5m

Errr - I don't think so!  That top one means that even if you didn't slow down it would take seven seconds to
travel that 98m, and stopping from 50kph certainly doesn't take 7 seconds...

> The formula used was x = v^2 / (2 * a)

I suspect you should have square-rooted the answer to this.

Cheers,

Howard Winter
St.Albans, England

On Wed, 30 Mar 2005, Jake Anderson wrote:

> i dont know about you but it dosent take me 98 meters to stop from 50kph
> v^2 = u^s + 2as
> s = v^2/2a
> s = (14^2)/(9.8*2)
> s = 196 / 19.6
> s = 10 meters
>
> I regularly practise my 50kph emergency stop (when was the last time you
> practised something that could save your life)
> in my 1985 ford telstar i have it down to a bit over 2 parked car lengths
> (well a car and a van, eh what can i say they always park next to each
> other) now that is with the advantage that i know its coming and i begin
> braking at the start of the first car.
>
> that to me agrees fairly well with the "theory"

Sorry for the goof, a decimal dot got misplaced in my table below. The
correct numbers are as shown, divided by 10. I.e. 50kph ~= 9.8 meters as
you say. That also means that it's enough if the traffic light is seen
from 50 meters, but my source says 100 meters (or so I remember).

>> {Original Message removed}

On Wed, 30 Mar 2005, Howard Winter wrote:
>> Braking track length at 1g is (at 0.5g the track doubles):
>>
>> 50kph        14m/sec                98m
>> 40kph        11m/sec                61m
>> 30kph         8m/sec                34m
>> 20kph         5m/sec                15m
>> 10kph         3m/sec                4.5m
>
> Errr - I don't think so!  That top one means that even if you didn't slow down it would take seven seconds to
> travel that 98m, and stopping from 50kph certainly doesn't take 7 seconds...
>
>> The formula used was x = v^2 / (2 * a)
>
> I suspect you should have square-rooted the answer to this.

How about a decimal point ? 9.8 meters, 6.1 meters, ... my mistake
anyway.

Peter

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