Angle measurement switch.

[140hp900]

I'm trying to make a switch that will activate at a specific angle. It's for a motorcycle so it can't be affected by centrifugal force. Any ideas? Thanks!

 

[Boncuk]

check for acellerometer circuits.

Analog devices and Maxim both have the full programm of one to three axis acellerometers.

Boncuk

 

[Diver300]

If you are trying to measure the lean angle on a motorbike, it is a really difficult problem.

A tilt sensor normally works by measuring the acceleration due to gravity in one direction. A motorbike leans into the corners so there is no apparent sideways acceleration.

If you want to measure how fast the motorcycle is cornering, and you don't need to know whether it is turning to the left or right, you could measure the apparent vertical acceleration and work out the lateral acceleration from that.

If the apparent vertical acceleration is A, the angle of the bike is:-

Angle = arctan ( sqrt ( A^2 - 1 ))

You will need damping to remove the effect of bumps, and it is not at all accurate for small cornering angles, but it is good at larger angles.

 

[140hp900]

Thanks for the replies! I was thinking maybe of a simpler, mechanical way of accomplishing this. I had an idea to mount a gyro at the end of a pendulum. In theory (at least in my head) this should stay perpendicular to level. The same way an airplanes artificial horizon works. I don't need to measure the angle accurately, just close enough to 10 degrees off center to activate a switch. Any thoughts?

 

[dknguyen]

Thanks for the replies! I was thinking maybe of a simpler, mechanical way of accomplishing this. I had an idea to mount a gyro at the end of a pendulum. In theory (at least in my head) this should stay perpendicular to level. The same way an airplanes artificial horizon works. I don't need to measure the angle accurately, just close enough to 10 degrees off center to activate a switch. Any thoughts?

Have you considered the size, complexity and fragility of mounting a motor at the end of a swinging arm?

 

[Diver300]

I was thinking maybe of a simpler, mechanical way of accomplishing this. I had an idea to mount a gyro at the end of a pendulum. In theory (at least in my head) this should stay perpendicular to level. The same way an airplanes artificial horizon works.

This is an electronics forum. Mechanical solutions are always more difficult to do.

Seriously, the artificial horizon in an aeroplane is a gyroscope in gimbals and is very difficult to make.

If you just want to know when the bike is tilting more than 10 degrees, a vertical accelerometer will need to sense about 1.015 G, so it would need to be very accurate, so my suggestion would probably not work.

I'm sorry that I don't know what would.

 

[dknguyen]

This is about the closest thing I can think of
**broken link removed**

You'd have to look at the "damping response" graphs to see the tradeoff between response speed and settling time. Depending on how accurate this 10 degrees has to be, 500ms is about as fast as you can get without have oscillations. THe critically damped response is almost as fast as the undamped response with almost the same lack of oscillation as the double damped response but may skirt your 10 degree threshold via oscillations depending on how fast everything is turning.

Anything else I can think of would require a gyro + accelerometer airplane-like system.

I wouldn't use any of this for a critical system on a motor bike.

 

[140hp900]

Well, it wouldn't be for any critical systems, just wanted to rig up a cornering light system. The electronic path looks to be more work than it's worth. I may just manually activate it. Thank all of you for the suggestions!

 

[vsmith]

This is my first post and I may be way off-base, but..
I put cornering/lean lights on my motorcycle using two mercury tilt switches, that I took out of old termostats. I hot glued them to a plate under my battery cover. I configured them at opposite angles, so they would activate at the appropriate lean angle to turn on my ground effects and cornering lights. Mercury is now banned from use in homes and vehicles, but a ball switch would work equally well.
Great Forum, glad to have found you!

 

[140hp900]

I was thinking of that as well but wasn't sure it would work. How well does it work?

 

[Roff]

This is an electronics forum. Mechanical solutions are always more difficult to do.

Seriously, the artificial horizon in an aeroplane is a gyroscope in gimbals and is very difficult to make.

If you just want to know when the bike is tilting more than 10 degrees, a vertical accelerometer will need to sense about 1.015 G, so it would need to be very accurate, so my suggestion would probably not work.

I'm sorry that I don't know what would.

I'm no physicist, and college physics was almost 50 years ago, but...
I can't see how, for a constant lean angle, the vertical acceleration (toward the center of the earth) can be other than the acceleration of gravity. Are you defining vertical to be normal to the axes of rotation of the wheels?
Or am I just confused?

 

[vsmith]

140hp900:
For my simple needs it works great. Although, I just re-read the title of your original post, this of course does not measure any angle it just defines, or alerts you to a specific lean angle. I was so pleased with it that I also used another one to detect when my bike was moved from its kick-stand position. It then sends a wireless signal to my pager/alarm. This alerts me to tampering while I'm in Radio Shack, etc..