Signal Conditioning

cancel
Showing results for 
Search instead for 
Did you mean: 

signal conditioning for conductive polymer

Solved!
Go to solution

I want to design a signal conditioning circuit for the conductive polymer which can give the output voltage in the range of 0-5 volts when the strain is applied from 0% to 25%. This conductive polymer gives the resistance change when the strain is applied.

I have also attached the sensor response.
https://onedrive.live.com/embed?cid=371B2D9E3A748E90&resid=371B2D9E3A748E90%211647&authkey=AEyMSkPPR...
https://onedrive.live.com/embed?cid=371B2D9E3A748E90&resid=371B2D9E3A748E90%211648&authkey=AIMdg_GK0...

0 Kudos
Message 1 of 7
(3,988 Views)
Solution
Accepted by topic author parthbhimani00

Common ways:

  1. -Apply a constant current (1µA?) , U=R*I, use an OP to scale the outputvoltage
  2. -Use a bridge circuit  and a instrumentation OP....
  3. -use another resistor (about same range 1-4.7 Meg) , build a voltage divider , buffer the outputvoltage , scale the outputvoltage

Looking at the response... I would choose 3.  a dual CMOS OP with Rail to Rail output, 5V supply ... 

 

Some resources: AoE.. (Art of electroniks, Horowitz/Hill)  and TI appnotes on OPs . 

Greetings from Germany
Henrik

LV since v3.1

“ground” is a convenient fantasy

'˙˙˙˙uıɐƃɐ lɐıp puɐ °06 ǝuoɥd ɹnoʎ uɹnʇ ǝsɐǝld 'ʎɹɐuıƃɐɯı sı pǝlɐıp ǝʌɐɥ noʎ ɹǝqɯnu ǝɥʇ'


Message 2 of 7
(3,949 Views)

Hello henrik,

Thanks for the solutions but the problem i am having is that the resistance of the sensor is not the same for certain force. but the resistance change is same. so what could be oprions for the auto zero design, Which can gives the voltage 0 to 5 for the force of 0 to 0.45 N.

0 Kudos
Message 3 of 7
(3,934 Views)
Solution
Accepted by topic author parthbhimani00

From the appearance of the strain/resistance graph you posted it seems that the resistance change is not instantaneous, is not directly proportional, and may have some hysteresis.

 

The first thing you need to do is to get information about the theoretical behavior. Is there a known function or algorithm which describes the relationship? Once you understand that relationship, then you can begin to develop a process to convert the measured values to strain.  The memory effects and non-linearities may require an iterative numerical approximation method to get the results you want.

 

Lynn

Message 4 of 7
(3,922 Views)

Thank you for your reply.

Actually there is no theoretical explanation as far as the strain vs resistance behaviour is concerned, but the  resistance change is constant every time you stretch the material. The material also shows the creep resistance.
Can you throw some light on how to find the transfer function of strain vs resistance?

 

Thank you. 

0 Kudos
Message 5 of 7
(3,858 Views)
Solution
Accepted by topic author parthbhimani00

A transfer function is just the output divided by the input. You may need to take precautions when the input is near zero to avoid infinities.

 

Because your device has a time dependence (creep) as well as strain dependence, it may be more difficult to separate the effects.  For example a family of curves with the same strain endpoints but different timing may help identify the characteristic time associated with the creep. Static tests might also be useful. Keep the stain constant until the resistance stops changing. Then ramp it to a different value and keep it there until stable.  Similarly it might be worth testing with several different strain endpoints separated by the same amount, (0-25%, 10-35%, 25-50%, ...). This could help identify possible non-linearities with total strain, changes in the creep with strain, or other effects.

 

I also note that in the data image you posted the resistance change in one direction is about 10% larger than the change in the other direction for the same magnitude of strain change. Overshoot appears in one direction but not the other.  There are a lot of unknowns in this system!

 

I think it may be premature to be thinking of transfer functions. Collect data under various condtions and look at the plots.  See if you can identify any patterns. The human eye-brain system is fairly good  at spotting patterns. If you think you see something, then try curve fitting to a similar function to see whether you can develop a relationship with predictive value.

 

Lynn

Message 6 of 7
(3,847 Views)

thanks for the solution.
What could be the possible soultion to convert the resistance change into the voltages. Since the initial resistance of the material is not constant. but it gives the specific resistance change for specific strain/force input.?

Thank You.

Best Regards,
Parth

0 Kudos
Message 7 of 7
(3,833 Views)