Hello and thank you for looking.
I am a newbie and need to design a circuit that will measure physical shock to an object and if the shock reaches an adjustable level will signal an audio alarm. Could anyone send me in the right direction? Thank you so much!

 

What kind of shock, mechanical, electrical?

 

physical shock, sensor could be mechanical or electrical just edited the post. Thank you CDrive.

 

These days, I think, Piezo elements are getting a lot of use in this application. Also, I don't know if they make them any longer but the alarm industry once used switches that I think we called "Tremor Switches". They were/are basically a very sensitive Reed Switch, that when whacked, would open or close momentarily. They were typically placed in the loop on windows to replace ugly foils.

I think we have members that have experience using Piezo's. There also may be (I think there are) specialized chips for this. What's the device that needs monitoring?

 

Hi CD!
I need qualitative measuring of the up and down force on a runner's body around the waist. The force should be accurate from 0 to 6 g's in increments of .1 . Don't know if a piezo can do that or it might need to be a thermal axis accelerometer. Again I am a total beginner but maybe the members could shed some light.

 

Ever play pinball? All you need is an old-fashioned "tilt" sensor. They were basically a suspended weight that could be knocked into a nearby conductor if the machine was abused.

The fancy modern version is the 3-axis accelerometer. You could then use a comparator, I believe, to trigger at any preset level of g, along any of the 3 axes. Laptops use these to park the hard drive heads in a collision.

 

Hi CD!
I need qualitative measuring of the up and down force on a runner's body around the waist. The force should be accurate from 0 to 6 g's in increments of .1 . Don't know if a piezo can do that or it might need to be a thermal axis accelerometer. Again I am a total beginner but maybe the members could shed some light.

Whoa, that's way different than your first post!

Do you mean QUANTitative measurement, ie. you need a number?

 

Sorry it's kind of similar even though I am botching it. Basically the project is a any type of small g shock sensor that fits on a person and would sound an alarm if the vertical impact g's from footfall hit a pre-determined, adjustable level, ie 2.6 g's. Sorry to be so confusing but this is where you guys rock...helping us to define the problem is the first step, right? The 3-axis accelerometer using a comparator trigger a might be just the right circuit. Any schematic links would be so appreciated wayneh.

 

The 3-axis accelerometer using a comparator trigger a might be just the right circuit. Any schematic links would be so appreciated wayneh.

I'm assuming the accelerometer outputs an analog voltage. If that's true, the comparator circuit is "easy". I guess you want the circuit to latch and hold something "on", maybe an LED or a buzzer, if the g level (voltage) has been reached?

For your application, I would definitely start learning about accelerometer ICs. Assume the rest is easy with help from the smart folks here - but focus on finding an accelerometer that fits your needs. At this phase you might want to consider how the circuit will be powered. Little button cells might be better than a big battery pack, for instance.

 

I have some experience with a Memsic 2125 dual axis accelerometer I am using with a basic stamp and transmitting the data wirelessly to my laptop- works- like a champ. The 2125 outputs PWM data which is not the same as voltage correct? And I am not sure how g's are computed off of two axis data used mainly for tilt.

I am looking some something a lot less expensive and self-contained on a belt that only sounds a momentary alarm and does not capture data. I will start my homework assignment. Button cells look like the better way to go than a bulky battery pack. Is it too tacky to ask where I might find a good contractor to help me further? Thank you!!!!

 

One potential problem I can see with your project is the orientation of the sensor vs orientation of the body during motion. You would almost need to monitor all three axis directions and do a vector summation to accommodate various body positions

 

I agree, unless possibly the sensor can be expected to always be in one orientation, I suppose vertical. A belt mounted device would be close, and this could simplify the project. I think a 3-axis device could find a market inside helmets and so on.

 

the sensor would always be in one direction for my application.