Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Simple accelerometer and tilt sensor

Status
Not open for further replies.

be80be

Well-Known Member
This would make a good accelerometer. How it works is it has three mirrors in it two on the led side of the flaps. And one on the bottom, with the LDR facing down out side of the flaps. When the car or what ever you put it on
accelerates they move back by the amount of force to make your car move. Which angles the light from the led towards the LDR. Which lowers the resistance by how much force is applied. You could also use it for sensing tilt.
I made one with ball and spring to see if the car moved but with flaps and mirrors you have better precision if you wanted to use it for tilt. It starts at 112K and gos down to 20k like turning a adjustable potentiometer very good control.
**broken link removed**
I guess people think this is big or hard to make it is only a 1/2 by 1/2 by 1 1/2 inch the mirror is made of
paper you get it at hobby shops. The flaps are hinged on a piece of copper wire with copper foil and the mirror glued to it an I coated it with solder for more weight. Here a pic of the first one I made not to good looking but it worked as good or better then what you can buy. And it cost less then a dollar to make.
Here shows how the adc values change on LCD
[embed]http://www.youtube.com/v/xsrZR8J2Y_c&hl=en&fs=1[/embed]
 

Attachments

  • acceleration.PNG
    acceleration.PNG
    6.4 KB · Views: 756
be8, it might be nice if we could see the actual sensor in that video, also it's going to be so mechanically sensitive it couldn't be used in any situation where there'd be high shock forces or vibration. Accelerometers are not that expensive.
 
You can tell shock same way sparkfun 27.00 Accelerometers dose it calibration. Acceleration is shock like the force that pushes you back in the seat when you hit the gas Or the shock of some one hitting your car in the rear end. Still Acceleration.
I was raising it with my hand. I wanted to see the voltage change more then watch it raise. I could fix it so vibration is not a factor. It should hold up good. I was thinking about using it where there is not much vibration
 
I like what you are doing with DIY sensor but here size works against you.

The force due to acceleration increases with an object's mass. force = mass x acceleration

The larger a sensor is the more sturdy it has to be.

Sturdy things tend to be massive.
 
This is not that big **broken link removed**
And from what I tested so far mine can do the same things it can it's just a little faster and it's decoder is built in mine a pic.
Use that one if you want to it can handle 1000 g Shock Survival not powered up and 500g with the power on. But from what
I tested mine would win hands down at tilt. That's not why I made it I like it.
 
Last edited:
Sturdy things tend to be massive
Good point my testing rig i glue it on with hot glue. I was getting the same voltage reading got my cam out and it started changing. The glue I had holding it to the arm didn't hold up lol
 
Sturdier things are more massive, but they also fall a lot harder. You should now work on a home-built vibrating gyro.
 
Last edited:
This would be fun **broken link removed** I have most of the junk to make it and I think my sensor would work out find to help keep you up right.
 
Accelerometers don't work for things like that. As soon as you accelerated forward, the platform would knock you on your back (or vice versa).
 
Last edited:
You no I always jump the gun . I was thinking about tilt control. It has two sensors and you could find the middle. It can tell if you tilt forward or backwards. With some fast Adc I'm sure you could keep it upright.
 
You still would need a gyro. ADC speed isn't the problem. It's that the acceleration when moving would skew the gravity reading making it think the ground is not quite where it actually is. Two sensors don't help either because acceleration skews both sensors in the same direciton.
 
Last edited:
I don't have any good links to gyro when you google you get food I done ate Lol. I see what your saying But I think there a way to make it work like I'm thinking. If it tilts back one LDR will increase in voltage and one will decrease in voltage. That will tell which way your moving. Now if i compensate for acceleration I could keep it upright. May work but if not still some thing to do. I like the ideal of a gyro but it would have to be big to keep me upright.
 
Last edited:
Hay I found this
One ceramic rate gyro, of the same kind that's in your camcorder (to detect your hand jiggling and stabilize the video) or RC helicopter (to stabilize the tail), and a 2-axis accelerometer to correct for drift
I have one
didn't no what it was now I do. LOL going to dig it back out and see what else is in it. I no I have seen one of these 2-axis accelerometer in my junk box 1300sf of basement LOL.
 
Last edited:
www.sparkfun.com has a bunch. You only need a single-axis gyro for a segway (but the PCB would have to be mounted vertically so the axis being measured is correct). If you used a dual-axis gyro then you could mount the PCB horizontally (but one axis would go unused).

How did you have one of those and not know? lol
 
Last edited:
I didn't no a camcorder had one in it till I just read about it. All I wanted out of it was this **broken link removed**
eyes for my bot but I have a long way to go before I can make that work. The pic is like what I have
 
Last edited:
Actually to increase accuracy if I'm not mistaken a Segway uses multiple independent but integrated gyro's, to reduce drift. And they're also sampled REALLY fast.
 
Last edited:
Do they use multiple gyros to try and average out the bias and drift? I knew they used multiple gyros for redundancy so if one fails you don't hit your head.
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top