Need a Quick, Low Cost Solution Switch Function

Hello.

I've been lurking & reading here for a while. I seems this is a great forum where even newbie’s can get some help, so I decided to ask for input.

I am frustrated trying to make a simple sensor which will activate a switch when a change in inertia, along a single axis occurs.

The circuit has to live on a moving (water proof) platform & activate a switch, which just turns on a LED when acceleration occurs.

It has to work at speeds between 1-200 MPH. I have tinkered with MEMs an IMU (which was way over my head) & some analog devises, but I can't make it work.

It's a school project & I'm running out of time. I need parts & circuit building advise if anyone has any to offer, please.

The operative phrase here is "Simple". I need a small, cheap, stand alone solution which can function without a laptop.

Thanks in advance.

Sounds like what you need is a very simple accelerometer switch.

Think of a small conductive tube, maybe 5-6mm diameter. Mount it upright, perhaps on a copper sheet or bare PCB. In the center of the tube, mount a small metal spring with a weight at its top and the bottom securely fixed to the same plane as the tube. Do not let the spring and tube come into contact, the spring/weight should be electrically isolated. Attach one wire to the base of the spring and one to the tube.

Whenever the device is moved, the mass of the spring and weight will damp its motion and it will touch the inside of the tube. The electrical connection will give you an indication that the device has been moved.

Cost = pennies, Build time = 30 minutes.

Brian.

We're thinking in the same direction. That was an idea I tried when I was trying pure analog solutions, but the NVH noise caused false triggers.

I tried an RC heli gyro & the gyro out of a car alarm system but I couldn't use the output / didn't know how to use the output.

To make it more interesting, this has to sence a changes like from 45 to 50 mph, so the transitions & forces are low at times. I'd like to be able to adjust the sensitivity too, incase I change the weight of the vehicle.

I have some thin film sensors, like from a washing machine, but they require too harsh a transition to make a signal.

betwixt said:

Sounds like what you need is a very simple accelerometer switch.

Think of a small conductive tube, maybe 5-6mm diameter. Mount it upright, perhaps on a copper sheet or bare PCB. In the center of the tube, mount a small metal spring with a weight at its top and the bottom securely fixed to the same plane as the tube. Do not let the spring and tube come into contact, the spring/weight should be electrically isolated. Attach one wire to the base of the spring and one to the tube.

Whenever the device is moved, the mass of the spring and weight will damp its motion and it will touch the inside of the tube. The electrical connection will give you an indication that the device has been moved.

Cost = pennies, Build time = 30 minutes.

Brian.

Click to expand...

Understood.

Plan B:

Use a potentiometer with it's shaft horizontal. Drop an arm downward from the shaft with the pot in it's half way position. Fit a weight to the end of the shaft so it works like a pendulum. Then measure the resistance change to calculate the amount of movement.

Could easily be adapted to provide vector information by using an upside down joystick arrangement in the same fashion.

Brian.

That's a good idea! It isolates the NVH issues of the sprung element also. OK, back to the test bench.

Do you think a transistor could be used for the switch, once the pot reached the trigger voltage?

A comparator would be better than just using a transistor but it would be a little more complicated. With a comparator you compare the voltage from the potentiometer with a fixed voltage (could be from a pre-set potentiometer) and see if it is higher or lower.

You can do interesting experiments if you use and op-amp too, for example splitting the voltage from the potentiometer down two routes, one to the '+' input of the amplifier and the other through a low-pass filter to the '-' input. The output of the op-amp would then show rate of change of acceleration as well.

Brian.