Design Request - Deceleration Activated Brake Lights

[sign216]

Zapper,
There's a difference between aggressively downshifting like a race car driver approaching a hard turn, and merely rolling off the throttle and then shifting down a gear or two to match rpm with the bike's speed.

I do all my own maintenance/repair, and fully understand how cheap and easy brake work is compared to clutch/transmission jobs.

But...I will give that there's two schools of thought on this. Some do all the braking with the brakes and a depressed clutch, while others use the engine in combination with the brakes. It's all a matter of degree, as long as you're not extreme in either practice.

Joe

 

[sign216]

I measured the bike's deceleration with engine braking, no throttle, in 3rd gear from 30 mph (48 kmh) over a 50 ft (15.2 m) distance, and it was 0.12 g.
A little higher than expected, but not a crazy number.
The bike is a '09 Moto Guzzi V7, with a number of performance enhancements.

 

[crutschow]

Some do all the braking with the brakes and a depressed clutch, while others use the engine in combination with the brakes.

I leave the clutch engaged until just before I stop, but never downshift.
Downshifting seems like a lot more trouble and effort than it's worth just to save a little brake lining.

I measured the bike's deceleration with engine braking, no throttle, in 3rd gear from 30 mph (48 kmh) over a 50 ft (15.2 m) distance, and it was 0.12 g.
A little higher than expected, but not a crazy number.

Interesting
But wouldn't you want the brake-light to come on when you back off the throttle in high gear?

I nearly got read-ended on the freeway one day a number of years ago from an inattentive driver.
I was travelling about 70MPH on my bike when I could see traffic up ahead starting to slow so I backed off on my throttle, although I was still a couple hundred feet from the car ahead of me.
Suddenly I hear the screeching of brakes right in my ear.
I looked in my mirror and there was a car about 10 feet behind me, fish-tailing from jamming on his brakes. Apparently he didn't notice my slowing until he was right on my tail.
I don't know how close he was to me when he hit his brakes, but I think I was about 10ms from getting creamed.
He apparently was pretty shook up about that as I noticed in my mirror that he pulled over to the right shoulder and stopped as I was driving away.

 

[Nigel Goodwin]

May save wear on your brakes but it puts added stress on the engine, clutch, gears, and chains.
Brakes are cheaper than those items.
I never down-shift to brake on my bike.

Extra stress is imaginary, it causes no problems, and is the correct was to ride (or drive).

Try watching motor bike racing, or car racing - watch what they do

In the distant past I knew someone who had two broken crankshafts sequentially happen on her VW bug.
After the second failure the mechanic determined that she was aggressively downshifting when slowing down and popping the clutch during the down shifts without blipping the throttle.
The continued shock to the crankshaft from that caused the breakages.

I'm EXTREMELY dubious that such an action would break a crankshaft, however she wasn't doing it properly anyway, by not blipping the throttle.

 

[sign216]

But wouldn't you want the brake-light to come on when you back off the throttle in high gear?

I think the decel. brake light will come on at highway speeds too.
The measured deceleration in 3rd (0.12 g) is probably higher, or similar, to the figure you'd get with engine braking in high gear at highway speeds.
Either way, it'd be good for the circuit to have a rheostat or other method to tune it, much like the commercial version.

 

[KeepItSimpleStupid]

Wat were your raw numbers? From to mph and seconds.

 

[sign216]

I made a dozen runs, if not more. Starting at 30 mph, closing the throttle, and ending at 25 mph, over the 50 ft stretch.
In essence, I would slow 5 mph through engine braking over 50 ft.
Margin of error approx. 20%.

 

[crutschow]

Extra stress is imaginary, it causes no problems, and is the correct was to ride (or drive).

Says you.
What possible advantage could that have in normal driving?

ry watching motor bike racing, or car racing - watch what they do

What's that got to do with driving on the highway at normal speeds?
They are out to win, prevent their brakes from overheating, and don't particularly care about extra stress on the drive train.
None of those apply to my driving conditions.

 

[crutschow]

I think the decel. brake light will come on at highway speeds too.
The measured deceleration in 3rd (0.12 g) is probably higher, or similar, to the figure you'd get with engine braking in high gear at highway speeds.
Either way, it'd be good for the circuit to have a rheostat or other method to tune it, much like the commercial version.

The reason I wanted the values in high gear at highway speeds is because I wanted to know the minimum G force we need to detect and that's the conditions for the minimum.
You just have to measure the time say to go from 60 to 50 or 50 to 40 in high gear, for example.
No need to measure distance.
Do you have a watch with a stop-watch capability that you could use?

 

[alec_t]

One factor to take into account, if you use an accelerometer, is the attitude of the bike. Going up or down hill will affect the measurement.

 

[Nigel Goodwin]

Says you.
What possible advantage could that have in normal driving?

It makes everything smoother and more efficient.

What's that got to do with driving on the highway at normal speeds?
They are out to win, prevent their brakes from overheating, and don't particularly care about extra stress on the drive train.
What's that got to do with driving on the highway at normal speeds?

Everything, it's about efficiency and safety as well as improving performance, correct use of gears greatly improves handling.

But I would have thought preventing your brakes overheating is a good idea whatever the circumstances?.

 

[crutschow]

It makes everything smoother and more efficient.

Downshifting is certainly not smoother than just using brakes.
Define efficient because I don't see how it can be.
Certainly it uses more petrol.

Everything, it's about efficiency and safety as well as improving performance, correct use of gears greatly improves handling.

If going thorough corners at high speeds, perhaps, but that's not the driving I do.

But I would have thought preventing your brakes overheating is a good idea whatever the circumstances?.

The circumstances I drive in never come close to overheating my brakes.

Am I being trolled?

 

[sign216]

The reason I wanted the values in high gear at highway speeds is because I wanted to know the minimum G force we need to detect and that's the conditions for the minimum.
You just have to measure the time say to go from 60 to 50 or 50 to 40 in high gear, for example.
No need to measure distance.
Do you have a watch with a stop-watch capability that you could use?

How about this;

1. Use the 0.12 g as an expected high value, and plan for a slightly lesser value.

2. Make the unit adjustable, like the commercial version, to account for real-life issues.


And...this afternoon I went out on the bike and didn't use any engine braking, because I had new boots, and they were so thick and stiff it was hard to shift.
The boots better break in, because I genuinely like downshifting as I slow down.


Joe

 

[crutschow]

Make the unit adjustable, like the commercial version, to account for real-life issues.

My circuit has that adjustment.
But it's still good to know what the minimum signal is to see if my circuit can reliably respond to that.
Much below 0.1g, noise and offsets can become problematic.

because I genuinely like downshifting as I slow down.

I'm not arguing against it if that's your preference.
I just don't see another good reason to do it in normal street driving.

 

[sign216]

Below 0.1 g? Yes, the system is likely to operate below that. Engine braking for a car varies between 0.04 and 0.08 g (if I recall correctly), and if the motorcycle has 0.12 g as its high value, then below 0.1 g can be expected.

Can the circuit be adapted to account?

 

[MrAl]

Hello,

Really you want to use a method that works for any variation that causes a change in acceleration. This can come from many different sources.

Measuring can be done with an accelerometer or you can glean that information from a speedometer that has reasonable resolution because acceleration is the first derivative of velocity.

So to use a speedometer we have:
a=dv/dt

and using a first order approximation to the derivative we get:
a=(v2-v1)/t

where 't' is a small amount of time over which two speed readings are taken, v1 at time t1 and v2 at time t2, and here t=t2-t1.

So the way to do this is to take two readings spaced maybe 0.1 seconds apart, then do the simple math, and that gives you 'a' the acceleration.

For example, say we are traveling down the road and we get two readings v1=20 mph at time t1=1 and v2=21 mph at time t2=1.1.
This gives us:
a=(v2-v1)/0.1=10*(v2-v1)=10*(21-20)=10*(1)=10 miles per hour per second
and this value 10 is positive so we are accelerating.

A second example, we read v1=21 and v2=19 then we have:
a=(v2-v1)/0.1=10*(v2-v1)=10*(19-21)=10*(-2)=-20 miles per hour per second
and this value -20 is negative so we are decelerating.

So in one case we had +10 m/h/s and in the other case we had -20 m/h/s and so we calculated the acceleration in each case.

The requirement on the speedometer is that it must provide stable readings with high enough resolution for the application, and of course the readings must be accessible to a microcontroller unless the speedometer puts out an analog signal, and then we would handle the derivative with s small analog circuit.

 

[crutschow]

The sensitivity of the ADXL335 from Sparkfun is 300mV/g so 0.1g is 30mV.
It's difficult to reliably detect signals below about 10mV or .03G so that's my concern about the minimum g that needs detection.
If .03g or so seems adequate as minimum detectable then the circuit I posted should work.

 

[hyedenny]

Since you don't really need that actual g measurement of an accelerometer, and you're interested in doing it on the cheap -- you could easily build an ersatz accelerometer with a microswitch, a weight, and maybe a spring to sense just that single unidirectional axis of acceleration.

OR, from a piezo sensor. Very cheap and very small.

 

[crutschow]

you could easily build an ersatz accelerometer with a microswitch, a weight, and maybe a spring to sense just that single unidirectional axis of acceleration.

The problem is critically damping such a system so the weight doesn't oscillate during acceleration or deceleration and give false flashing.

 

[MrAl]

The problem is critically damping such a system so the weight doesn't oscillate during acceleration or deceleration and give false flashing.

Hi,

Yes i dont think it's that easy either. Plus the movement has to be changed into an electrical signal too.

To the OP:
What kind of speedometer do you have, as per post #56 ?