Help devising a project using a strain sensor

Thread Starter

TormundB

Joined Jan 6, 2018
9
Hello,

I am a 3rd year EEE student and am working on a project that will go on to form my 3rd year dissertation. I am having trouble formulating a project idea from the guidelines. I was told I should use a strain sensor to form some kind of self sensing smart material that could be used to detect forces acting upon it and maybe some damage too.

The outline was "Monitoring a composite material using strain sensor"

I have reviewed classic strain sensors using the wheatstone bridge in conjunction with an amp to get readings but this doesn't feel like a project. I have seen fibre optics used but this seems more like phd level work. I have also seen a self sensing technique where the material itself self senses. I believe it was a carbon composite the resistance through the material would change with strain/stress. much like a strain sensor. My concern is ending up going too far into the civil/material engineering.

If anyone could offer some pointers of areas to look at I might have missed that would be much appreciated.


Thanks
 

wayneh

Joined Sep 9, 2010
16,400
Bike riders are always interested in things like pedal pressure, crank torque, frame stresses and so on. Might be an area to look at.
 

sc0tch

Joined Nov 6, 2018
64
Hello,

I am a 3rd year EEE student and am working on a project that will go on to form my 3rd year dissertation. I am having trouble formulating a project idea from the guidelines. I was told I should use a strain sensor to form some kind of self sensing smart material that could be used to detect forces acting upon it and maybe some damage too.

The outline was "Monitoring a composite material using strain sensor"

I have reviewed classic strain sensors using the wheatstone bridge in conjunction with an amp to get readings but this doesn't feel like a project. I have seen fibre optics used but this seems more like phd level work. I have also seen a self sensing technique where the material itself self senses. I believe it was a carbon composite the resistance through the material would change with strain/stress. much like a strain sensor. My concern is ending up going too far into the civil/material engineering.

If anyone could offer some pointers of areas to look at I might have missed that would be much appreciated.


Thanks
There are many types of strain sensors. And until you decide what you want to accomplish, such as what you are detecting strain on it is impossible to determine a viable route. Fiber optics are actually simpler than one may think in these uses as it basically acts as a stretched out optocoupler where you send light across a somewhat fragile fiber and if it is strained to a certain extent the phototransistor no longer remains closed and "strain" is detected. You could even use a photo-sensor to determine a level of strain across a fiber optic cable(would need to be a cheap cable to have defects as it is bent or compressed in passing it's light)

You can also use impedance based sensing using ropes of conductors, thin gauge to increase impedance with the strands close together in such a way that when strain(whether compressive force, or shearing force) is applied it will cause a short. You can determine where the short is and therefore where the damage is by measuring the impedance across the rope.
 

Thread Starter

TormundB

Joined Jan 6, 2018
9
There are many types of strain sensors. And until you decide what you want to accomplish, such as what you are detecting strain on it is impossible to determine a viable route. Fiber optics are actually simpler than one may think in these uses as it basically acts as a stretched out optocoupler where you send light across a somewhat fragile fiber and if it is strained to a certain extent the phototransistor no longer remains closed and "strain" is detected. You could even use a photo-sensor to determine a level of strain across a fiber optic cable(would need to be a cheap cable to have defects as it is bent or compressed in passing it's light)

You can also use impedance based sensing using ropes of conductors, thin gauge to increase impedance with the strands close together in such a way that when strain(whether compressive force, or shearing force) is applied it will cause a short. You can determine where the short is and therefore where the damage is by measuring the impedance across the rope.
Hello, I took your advice and I have been able to speak with my supervisor and have a more complete idea of what the aims should be, although there is still scope to change if the circumstances require it.

The reference to a composite material is more to do with the end product which will be an embedded sensor network incorporated into a material giving it additional properties rather than using the properties of a material to detect strain.

The aim is to detect position and magnitude of force on a given piece of material using strain gauges. So I would need to design a circuit that could be integrated into a material to detect the location of any forces. Then I can look for a way to create a prototype, maybe something like 'Direct Write' I have seen or a flexible PCB board. Detecting damage isn't required though additional credit would be awarded.

Does this sound more reasonable for a 3rd year EE? Would be great to hear some feedback.

Thanks
 

sc0tch

Joined Nov 6, 2018
64
Hello, I took your advice and I have been able to speak with my supervisor and have a more complete idea of what the aims should be, although there is still scope to change if the circumstances require it.

The reference to a composite material is more to do with the end product which will be an embedded sensor network incorporated into a material giving it additional properties rather than using the properties of a material to detect strain.

The aim is to detect position and magnitude of force on a given piece of material using strain gauges. So I would need to design a circuit that could be integrated into a material to detect the location of any forces. Then I can look for a way to create a prototype, maybe something like 'Direct Write' I have seen or a flexible PCB board. Detecting damage isn't required though additional credit would be awarded.

Does this sound more reasonable for a 3rd year EE? Would be great to hear some feedback.

Thanks
It's feasible and even electrically simple to do, however will have mechanical and cost hurdles. The most apparent solution would be to use load cells at joints or between 2 parallel beams. Either way I look at the problem I see more of a mechanical challenge than I do an electrical one as where to mount and orient the sensors is far more difficult than how to design the electrical schematic. The elcelectri side of it will be basically the same as a kitchen scale with multiple load bars.

Of course there are many more options depending on how in depth you want to get into it. Can make a strand of 2 parallel conductors that when deflection occurs in the material being strained by shorting the 2 out and messuring the impedance to tell where the short occured .
 

Thread Starter

TormundB

Joined Jan 6, 2018
9
It's feasible and even electrically simple to do, however will have mechanical and cost hurdles. The most apparent solution would be to use load cells at joints or between 2 parallel beams. Either way I look at the problem I see more of a mechanical challenge than I do an electrical one as where to mount and orient the sensors is far more difficult than how to design the electrical schematic. The elcelectri side of it will be basically the same as a kitchen scale with multiple load bars.

Of course there are many more options depending on how in depth you want to get into it. Can make a strand of 2 parallel conductors that when deflection occurs in the material being strained by shorting the 2 out and messuring the impedance to tell where the short occured .

Thanks, that has always been my problem with the task as it has always felt like more of a mechanical or even civil challenge once you get past the initial problem of using a strain gauge. My sticking point seems to be placement of sensors and deciphering the strain to locate the position which I believe falls under a mechanical challenge.
 

sc0tch

Joined Nov 6, 2018
64
Thanks, that has always been my problem with the task as it has always felt like more of a mechanical or even civil challenge once you get past the initial problem of using a strain gauge. My sticking point seems to be placement of sensors and deciphering the strain to locate the position which I believe falls under a mechanical challenge.
Which in my opinion an engineer of any field should be able to solve basic challenges in the core engineengin fields. Not so proficient at mechanics to be a "mechanical engineer" but at least knowledgable enough to tackle the mechanical issues of applying a circuit to real world problems.
 

Thread Starter

TormundB

Joined Jan 6, 2018
9
Which in my opinion an engineer of any field should be able to solve basic challenges in the core engineengin fields. Not so proficient at mechanics to be a "mechanical engineer" but at least knowledgable enough to tackle the mechanical issues of applying a circuit to real world problems.
Fair enough, it's just that I am marked mainly on the electrical content of the project so I was worried about going deep into other areas, although it may be unavoidable in this scenario. However I can look at methods of embedding the sensors and circuit to create a prototype as well as wirelessly passing the data to a 3rd party to bump up the electrical/electronic content.
 

sc0tch

Joined Nov 6, 2018
64
Fair enough, it's just that I am marked mainly on the electrical content of the project so I was worried about going deep into other areas, although it may be unavoidable in this scenario. However I can look at methods of embedding the sensors and circuit to create a prototype as well as wirelessly passing the data to a 3rd party to bump up the electrical/electronic content.
Yes. I know personally if I was to tackle the project I would probavly use either a rPu 0W or an Arduino with a wireless module for the microcontroller.collect data for various sensors. Either use a small display panel or make a web api to report the data. There is hundreds of routes you can take for the 0roject ranging in difficulty, expense, and complexity.

No matter which option you take will certainly involve some mechanical engineering, which I'm sure your professor is very interested in seeing how these challenges are overcome.

So then you need to determine how intricate if software you want. From simply dumping data to a log file, or outputing to a small display Al the way to designing a web based app that the controller will sync to when it's in range if wifi or possibly even a simple Android app to receive the data over Bluetooth.
 
Top