FloorSensor!!!

hi...i'm trying to build a FloorSensor which is able to determine the force of a person exerted on it (but is not like a weighting machine). This FloorSensor will actually determine the action of the person (example: walking, running, jumping, etc) on the mat (FloorSensor) by analysing the the different forces acting on the floor platform...the problem is, i do not know wat kind of sensor to used.

I found one type of force sensor(piezoelectric sensor) but the max force that it can withstand is only about 10kg but i need a sensor that can withstand about 200kg(or can withstand the force of an adult jumping and landing on the platform). The platform size will be around 1x1m, therefore i need quite alot of these sensors.....so it got to be cheap. (i actually found one of these sensor already in a mat form on one website but it is very expensive, so i try to build one low cost sensor platform myself)

Do anyone has any idea wat kind of sensor i can used which is not expensive? can anyone pls give me the the details of the sensor (ex. model, price, max load, etc)

just a reminder....it is not a weighting machine. sensors needed are more like tectile sensors but with the ability to determine forces acting on it (ex. piezoelectric sensors).

Thanks in advance!!!!!

i think u can design your own sensor for your need. u need only to sense the force exerted on it. make 2 platforms(hould be electrically conducting. arrange then in such a way that there is a littel gap inbetween both of them.
when some one jumos onit,both these surfaces touch each other and gonna conduct current which is going to detect. the maximum weight beared by it depends on the material which u use for making it.

**broken link removed**
**broken link removed**
**broken link removed**


they are not exactly what u need,but can be useful to you

srimannarayanakarthik said:

i think u can design your own sensor for your need. u need only to sense the force exerted on it. make 2 platforms(hould be electrically conducting. arrange then in such a way that there is a littel gap inbetween both of them.
when some one jumos onit,both these surfaces touch each other and gonna conduct current which is going to detect. the maximum weight beared by it depends on the material which u use for making it.

Click to expand...

I think u got me wrong there....i need a sensor that can detect force...force sensor, piezoelectric sensor, etc....not a sensor that sense something on it. I'm using a few sensors on the platform so that i can determine the force of each foot n the distribution of force of the foot. and i just need the info of the product of sensors i can use, i'm not goin to build the sensor....my main objectic is not to build the sensor but to analyze the force distribution on the sensor mat[platform]. If there is a readily built sensor mat(platform) that is cheap then let me know (but i doubt it will b cheap cos i have already check on that, therefore i am building my own sensor mat[platform] with the sensors).
Thx anyway srimannarayanakarthik!!

anyone has any idea, need it quite urgently!!!

Two metal surfaces separated by air or a dielectric make a capacitor. Changes in capacitance can be used to indicate that deflection has occured and also indicate the amount of deflection.

You might take two 1 meter by 1 meter sheets of circuit board material with the copper on one side separated into strips of some width. If you put the face to face with the direction of strips perpendicular to one another you'd have a matrix of small squares where one strip crosses another. You could then scan this matrix rapidly and look for changes in capacitance. Note that the flexibility of the circuit board material, the size of the strips and the characteristics of the resilient material all profoundly affect the behavior.

You could cut the foil away from these large boards - or etch it if you have the means.

I was interupted several times and need to clarify.

The conductive surfaces would be separated by a resilient insulating material. A very thin rubber like polymer sheet might serve well in this application.

kwon said:

I think u got me wrong there....i need a sensor that can detect force...force sensor, piezoelectric sensor, etc....not a sensor that sense something on it. I'm using a few sensors on the platform so that i can determine the force of each foot n the distribution of force of the foot. and i just need the info of the product of sensors i can use, i'm not goin to build the sensor....my main objectic is not to build the sensor but to analyze the force distribution on the sensor mat[platform]. If there is a readily built sensor mat(platform) that is cheap then let me know (but i doubt it will b cheap cos i have already check on that, therefore i am building my own sensor mat[platform] with the sensors).
Thx anyway srimannarayanakarthik!!

anyone has any idea, need it quite urgently!!!

Click to expand...

sorry i misundeerstood u

I like the capacitive approach but can't help think it will have some odd frequency effects (the charge won't change instantly but the discharge rate will change). I'm not sure how best to do it.

another approach would be to put a conductive foam (like the stuff that ICs come stuck in) between 2 metal plates. pressure that compresses the foam will cause the resistance to decrease proportionally to the distance between the 2 plates. there are commercial versions of this.

I'm quite interested in force sensors for a similar application but I won't hijack this thread with it, I'll just keep an eye on the ideas and comment on them if they interest me if that's ok

I've never thought of the capacitive idea before... sounds like it might be worth experimenting with.

Brian

Hello everyone,
Just a thought, but wouldn't a 100lb person running or jumping be very similar to a 250lb person walking. Furthermore wouldn't a 250lb, or even 300lb person generate quite a large force when jumping? I'm not trying to discourage, only making sure to consider this in the design. Are you going to be in a situation were you already know this data, or will it be like an alarm sensor where you don't know who will be "using" it?

Sorting out heavy people from light people - and running from jumping will require considerable experimentation to understand the signature. The system might require some calibration for each user to address the differences in weight but the original poster spoke more of dynamic requirement. It may be that the "signature" for jumping is the same regardless of a person's weight.

Hello stevez
Yes, I see what you mean. I presume a jump would begin with a faster transition and a higher spike before settling to the representation of the persons wieght, where as a run would be a slower rising transition and a rounded peak with a slope to zero. Interesting. Thanks!
Edit: I also presume this would take some creative programming to automate (meaning: to get a definitive output of Running, Jumping, Walking or other out of a microcontroller).

Hello,

I am the owner of an alarm company. One type of interior sensor that I like to use alot is called the Pulsor. It is a stress sensor that is epoxied to the bottom of a floor beam. It detects a person walking or running across its feld of detection. It is basically a resistive type of device. 2 can be connected to a loop with up to 4 sensors on the basic 1 zone processor with sensitivity adjustments. There is also a 3 zone processor. These devices can be used in a wide range of applications. I believe the web site is www.sureaction.com. Just to give you an idea of costs, the pulsor sensor is around $35US and the 1 zone processor is around $55US. Like you said these items are not cheap! But they do work great and when installed properly, they do not false alarm. We have them installed on stairwell landings as well as "trap locations" in the traffic pattern of the house.

Hope this helps you out!

Have a nice day,

Alarmman

thx everyone for the help!!!

stevez idea is quite interesting but the problem is it is only one sheet and unable to determine the force distribution of the foot and the location of the foot on the mat. unless i make a lot of tiny ones, and the fact that it need a lot of experiment n time to get it right (dont have much time for that). Preferable a ready made sensor, a lot of them, so that i can place it evenly on the mat. and from there i can get the force distribution of the feet.

oh ya....i have calculated the force of a person jumping....for 0.5m shud b about 3 times the person's weight, if i'm not mistaken (just a rough calculation, didnt consider the fact that the weight distribution is on both feet n i assume the force act on one point or one sensor, so actually sensor that can withstand 150kg shud b find). This project is only a prototype. i will b getting a person less than 50kg to do da jumping, walking, etc.

And this mat sensor is only one part of the project, so i wouldnt wan to built a the sensor (too time consuming). The real challenge is the programming part where i need to input the datas collected and do the analysis.

hi there alarmman, the pulsor sensor sounds interesting but it doesnt seems to give me any info on the force/stress, does it? and isit small...cos like i said earlier, i need to determine the force distribution. therefore i need a quite a few of these. thx for the info.

Hi Kwon,

Yes you could possibly get the information regarding stress if you took the raw output of the sensor and fed it into an analog port on a PIC for instance. You could get many readings over a very short period of time and make your determinations for the info gathered. Remember, this device is just a variable resistor!

AlarmMan

continue...

and i have found a piezoelectric sensor that can withstand 15N of force. that is the maximum i can find....probably goin to use it if dont have other choice. Planning to place all the sensors evenly on a platform and will try design my platform so that it will absord the extra weight and oni apply forces not more than 15N on the sensor....didnt have any info on the model of the sensor yet, but do anyone have any idea how i show place the sensors evenly on the platform so that i can get the most effective result n with least sensors as in how far apart each sensor shud b from one another? any design on the platform is welcome!!!!

but anyway, back to the main purpose of this thread....if anyone got any info on wat type of sensor i should use, pls let me know. Again i need sensors that can give me the force/pressure, preferable in small size cos i need to use a lot of them n place it on a 1x1m platform to be able to determine the FORCE DISTRIBUTION of the feet n the location of the feet. and ofcourse it must must be cheap!!!

Thx everyone for the help!!! keep sending in your ideas n infos.....Thx again ppl

Regarding my home-made capacitive type sensor: if you divide the 1 meter by 1 meter sheets into 10 strips, and have the strips perpendicular to each other you will have a 10 x 10 matrix of square capacitors - or 100 sensing points. You might scan these at a high rate - fast enough to get the resolution you need. Divide the sheet up further for greater resolution though the electronics gets more complicated.

Or you could allways go for the Physics approach.
and that is the weightforce n the 1x1m panel will be shared evenoly over it.
Because of this, you could double up the sensors, or multiples of the sensors, and as such get an accurate result.
If you need help with something like that, I could help.
PM me if interested though, i dont wanna go to all that trouble for nothing.
It sounds like your best bet though, its simple and effective.

If you use a single 1 meter square platform and support it from four corners and measure the force or weight at each corner -

A. If we set a 50 kg bar on to the platform the sum of the weights at each corner would total 50 kg. The weight at each corner would be 50/4 only when the bar is centered on the platform. Any difference in force or weight at each corner is the result of being off center and it would seem that the actual position of the bar could be calculated.

B. If we added another 75 kg bar to the 50 kg bar then we'd see forces that total 125 kg. If we took only the information available when the two bars are on the platform we'd have trouble resolving the location of the bars.

C. If we had the information from 'A' stored somewhere we could subtract it making the task of locating the 75 kg bar managable.

Note that the platform itself would muddy the waters a little but that can be measured and adjustments applied to the raw data. Keep in mind that we've done nothing to address horizontal forces.

If you think of the two bars in a more dynamic situation you could see how you might sort out the impact of one foot from another. You might have to throw out a few moments of data that are ambiguous (both feet contacting at the same exact moment) but that isn't likely to happen too often.

wow ,this topic is so hot ,everyone discuss this problem .
I know some manufacture which produced this sensor .It can determine the action of people such as running ,jumping ,walking ,and other action.