February 3, 2010
about 4 years ago
I just wrote out a (lengthy) response to Gizmoguy explaining things in more detail, so you might want to check that out, but it sounds like you get the circuitry. As for sensing, just use ONE of the sensors. The other is just there to balance the Wheatstone bridge. And actually, because of the nature of a Wheatstone bridge and the fact that the two sensors behave the same way, applying a load to BOTH sensors would completely cancel out the signal. If you have the circuit wired you can see that pushing on one sensor will result in a positive Vout while pushing on the other will result in a negative Vout. Pushing equally on both at the same time will give you nothing because you’re keeping the bridge balanced.
Now I can think of some situations where you might use this behavior. For example, you could have a board or something supported on each end by the two sensors. Then a balanced force applied uniformly across the board or directly in the middle would lead to no voltage change, while an uneven force distribution across the board would lead to either a positive or negative voltage change depending on the side.
As for the actual mechanical application of the force to the sensor, you might have to get a little creative. For my application I ended up gluing a nut right onto the little round nib and then screwing a bolt into it, but there are definitely other (better) ways depending on your application.
I used 1 differential amp with the two red leads input to +Vin and -Vin. It doesn’t really matter which goes to +Vin and which goes to -Vin, it’ll just reverse the sign of the output. So basically in my setup I have the two sensors in parallel (black-black-ground, white-white-Vs) with the two red leads going into the +Vin and -Vin of an instrumentation amplifier with adjustable gain. As for the instrumentation amplifier, I am supplying it with 5V (same line that’s supplying power to the sensors) to +Vsupply and grounding the -Vsupply.
I wish I could just draw a schematic… it would be a lot easier, but I’ll try to explain in words.
Let’s say you’re using a 9V battery for supply (you could really use any voltage as long as it falls within the range of that the instrumentation amplifier can handle), so you have +Vs and -Vs (Ground).
Wired to +Vs should be the two white leads and a jumper wire to the +V supply of the instrumentation amp. Wired to -Vs (ground) should be the two black leads and a jumper to the -V pin of the instrumentation amp. Now, the red leads from the sensors should go into the +Vin and -Vin pins on the inst amp. The only remaining pins on the inst amp (8 pin) should be the two Rg pins, the Vout, and the REF pin. So you connect a resistor (choose a fairly low resistance for the greatest gain) across the two Rg pins (pins 1 and 8) and don’t worry about the REF pin. Now, if you want to filter the output a little, you could set up a simple passive RC circuit. Attach a resistor and capacitor in series to ground (choose whatever values you need to get the right cutoff) and then measure across the capacitor. If you want to forgo the filtering, just measure the Vout from the inst amp with respect to ground. Voila! When you push on either of the sensors, the voltage will change.
Note: Because one of the sensors is only there for temperature compensation and completion of the Wheatstone, you don’t need to worry about it once it’s wired in. It doesn’t matter which sensor you make the passive element. Pushing on one sensor will make the Vout positive, pushing on the other will make the Vout negative. So pick whichever sensor makes you happy (probably the one that causes a positive Vout) and make it the active element.
See my responses to RandyF and Member201737 above. Good Luck!
See my response to RandyF above. Basically you arrange the two in parallel (white-white-Vin, black-black-Grnd, read voltage across the red). You’ll probably want a differential amp as well. Also, the final output is pretty noisy, so you might want to filter it with a quick low-pass (If I remember correctly I used a cutoff somewhere between 20-40Hz).
Ben121 above was correct. Use two in parallel for best results. I got mine to work with really great sensitivity (it picks up the weight of a pen). I applied voltage (9v) to the white wires, grounded black, and then input the two red wires into a differential amp (AD620AN) with a gain of 1000. On the output I low-passed with a simple passive RC filter. I have yet to conduct any scientific tests regarding linearity, etc. but judging by the signal I get I’d be surprised if I couldn’t get within a few grams of accuracy.
No public wish lists :(
Forgot your password?
No account? Register one!