Comparing accelerometers with electromechanical sensors in the context of airbags

Hi there,

I’m a second year pre-engineering student and i’m working on airbags for my end of the year project. In a particular part of my project, i am looking to compare the use of accelerometers VS electromechanical sensors in airbags for shock/deceleration detection, in order to showcase why car manufacturers have moved on from electromechanical sensors. I was wondering if doing two different experiments : one with an arduino and an accelerometer, and another with a rolling ball switch would allow me to conclude?

Thanks in advance for any advice that will be given :slight_smile: !

Do you have an automotive-grade accelerometer to experiment with?
What do you hope to demonstrate?

30 years ago I was involved in creating a crash, impact, sensor. The technology ordered before my time was a 3 dimensional electronical accelerometer. Sure it detected solid crashes head on. However the customers wanted to detect if their fork lift trucks did cut corners and bend the legs of the 300 feet tall racking legs.
What was very disturbing was that the "impact sensor" false triggered from other mechanical noise, vibrations in the truck.

My advise is to drop the highly sensitive electronical accelerometers and use more low frequency sensitive devices.

Define the axis for the sensor! Front/rear, side,..... Know the rule of impacts. The Q-factor, the goodness factor means more than a lot. Think about colliding with a 1 ton bale of cotton or a 1 ton block of concrete. I assure You the readings from every sensor will difer a lot!

1 Like

Really you ought to be researching this properly . As a student you are supposed to be learning how to research - when you finally get a job you can’t be asking questions on a hobby forum for free consultancy .
Look at technology available, look for existing research papers, look at what is down now etc .

Sorry - my opinion , but short cutting your work this way doesn’t really help you .

Yes, i’m planning on using a MEMS accelerometer hooked to an arduino card, two diodes, and a small toy electric car ; and according to the crash code that im gonna write, if the defined threshold is met, the illumination of the green diode simulates the deployment of the airbag.

if i had found satisfying answers online i wouldn’t be here, but thanks for the advice :slight_smile:

Thank you for your response !

Found this in thirty seconds - might not be relevant , but stuff is out there and the Internet is not the only source

Example

1 Like

I'm not sure that answered my question, but just remember that if that green LED lights for any reason other than a severe collision, you could have caused an injury, or the death of the occupant

1 Like

At about the same time we had the same problem with a physical 1D sensor on a vibrating table where it happened to trigger even at much lower excitation than predicted by the manufacturer supplied mathematical model. Fortunately I only did the sensor check, not the real life application :slight_smile:

thank you so much for your response but unfortunately i’m not really looking to study physiognomies sensors in the context of airbags.

Thank you for your response ; well im looking to essentially compare the sensitivity of these two airbag sensors and thus demonstrate that accelerometers are preferred to electromechanical sensors in the context of airbags.

My point is that data is out there for you to research !!!

okay i think i got that from your first reply. i’m not gonna waste my time explaining how much research i’ve done on this topic. If you have no other point to prove, thank you for your participation to this topic.

Your university has access to a vast library of scientific papers, including the entire IMechE and IEEE outputs. Have you searched those?

1 Like

Thank you for your response. Will i be able to access these if i’m not attending an american university ?

Here …

1 Like

Imaging bolting the "impact sensor" to a 4 ton steel truck, and a light hit by a hammer triggered it.....

1 Like

Thank you so much. I’ll check it out.

...or simply slamming the door

1 Like