Do I Need Multiple Power Supplies for This Op-Amp Circuit?

[atlbraves49]

so this is the op-amp circuit I'm going to have to implement...

http://www.tekscan.com/images/flexi-circuit-new.jpg


Does that mean I need a power supply to power the supply voltages of -9V and 9V, and then another power supply to power the drive voltage of -5V?

 

[skeptic2]

Why not use a 5.1 V Zener between ground and V(t), and a resistor between V(t) and -9 V?

 

[atlbraves49]

Why not use a 5.1 V Zener between ground and V(t), and a resistor between V(t) and -9 V?

could you explain more about the zener? (and the resistor between V(t) and -9V should work, thanks)

 

[skeptic2]

Zener diodes are reversed biased diodes that conduct at specific voltages. They come in a variety of voltages usually the same as the 10% resistor values. Zeners below about 4 V are of a different type and their voltages are not as stable as those above that value. The actual voltage varies a little not only from unit to unit but also due to the amount of current they are conducting and their temperature.

Zeners come in various wattages so you will need to pay attention to how much current they are conducting and how much power they are dissipating.

For your circuit you would connect the Zener's cathode, the end marked with a band, to ground and the other end to V(t). For starters you might try using a 1 K resistor from the Zener anode to -9 V. Note: If you connect the Zener in backwards it will work like a diode and the voltage drop will be about 0.65 V instead of 5.1 V.

 

[atlbraves49]

so something like this (just to make sure I am understanding the setup)?
http://img443.imageshack.us/img443/3008/opampuu6.jpg

And is there a reason you can't just put a resistor between the -9V and -5V to get the voltage drop of 4V across the resistor? Like, why is the zener diode even needed?

 

[skeptic2]

You're right, you don't really need the Zener (your diagram was correct, however). You could do it with a resistive voltage divider, especially because the Flexiforce is such high resistance. Keep the resistances at less than 1/10th of the lowest value you expect to see with the Flexiforce.

 

[berkeman]

Consider decoupling that -5V reference voltage with capacitors to lower the impedance of that reference.

 

[Proton Soup]

i guess it depends on how accurate you need things. zeners are generally crap as a reference. you can get some noise coupled through the power supply with an op-amp, and you want to put some capacitors on the power inputs near the device... but overall, probably more error is introduced directly via Vt.

edit: if you measured both Vt and Vout to compute each measurement, i don't guess it would matter much.

 

[dlgoff]

You will need to calibrate the amplifier/sensor with some sort of standard forces applied to the sensor. What are you requirements? If you want really good results, you could use NIST traceable weights as your calibration standards (only for small values of force; cost prohibitive for large weights).