Amplify Pressure Sensor Output with BJT

[refind]

I have a pressure sensor, it has 4 wires, In+ In- Out+ Out-. I excite In+ and In- with a 12V battery and the outputs are about 6V each, the difference is very small, 0.7mV at 1atm and 12.9mV at 100psi.

I want to read this with a microprocessor which takes in analog 0-5V so I want to amplify the difference (Out+ - Out-) x 50 or so.

I tried a few different op amp circuits with LM318N, I tried an instrument amplifier INA2126 and after 25+ tries none of them work. The reading I get from these amplifier outputs is always 3-6V and it never changes when I put pressure on the sensor or not. So I'm giving up on the amplifiers, and my question is could I use a BJT or some kind of transistor to get the job done?

 

[Crazymechanic]

well since your application is low voltage and low current a BJT wouldn't be a problem.They come in plenty of shapes and characteristics.
I't s late ad I will go to sleep so maybe someone else can advise you a specific transistor for the job, although everything which is rated as small or signal transistor would basically do the job , some 2nxxx from radio electronics etc.

 

[skeptic2]

This is not a good application for a homemade transistor amp. This is a differential voltage application and to match the two input transistors in both gain and offset would be a real challenge. A good quality opamp or instrumentation amp is the way to go.

First, what is the input range you want to measure? You want 5V out at what pressure?
The INA2126 looks very capable of achieving the results you want. The datasheet is very helpful. If you could post some of your circuits perhaps we can help you.

 

[meBigGuy]

Post the specsheet for the pressure sensor. Opamps or instrumentation amp will do what you want.
Search google images for "pressure sensor amplifier" to see different ways to approach the job.

 

[refind]

The sensor is this one from Omega:
http://www.omega.com/Pressure/pdf/PX302.pdf

I can tell you what I tried with the INA2126, I have the dual one (attached diagram).
I tried:

+12V to Pin9
GND (battery - terminal) Pin8
Sensor + output Pin1
Sensor - output Pin2
Lots of resistors (ranging between 1k and 3k) on Pin3 and Pin4
Multimeter on GND and Pin6 or Pin7 (they are the same)

I also tried 2 batteries, with -12V on Pin8 (battery + terminal on GND). I also switched the wires of the sensor output (basically switched Pin1 and Pin2).

Nothing worked. Any ideas?

 

[rbj]

i don't see the attached diagram. is this INA2126 the same i find on the web? http://www.ti.com/product/ina2126

what value are you for R_G? the 1K to 3K?

have you wired up the instrumentation amp with test voltages independent of your pressure transducer?

by "nothing worked", was the output voltage zero? pinned to the power supply rails? oscillating? something else?

 

[skeptic2]

The INA2126 is meant to be used with a dual supply. If battery 1 was connected between pins 9 and 5 and battery 2 was connected between pins 5 and 8, that should work. Maybe your gain was too low.

Look on page 7 of my attachment for the value of the resistor you should use between pins 3 and 4. If you want 100 psi (12mV) at the input to produce 5V at the output, you will need a gain of about 800. This will require a resistance of about 80 ohms.

 

[Curl]

I want 5V at 850psi. I used a 1.4k resistor, a 1.7k resistor, and a potentiometer with everything from 1k to 3k. The output pin, measured with the multimeter, was always 5.4V, NOT changing at all even though I was putting 0-100psi on the sensor. Is it fried? I could try to see if I have another one.

rjb, yes that is the correct datasheet.

 

[skeptic2]

Curl, are you the same person as refind?

I have a pressure sensor, it has 4 wires, In+ In- Out+ Out-. I excite In+ and In- with a 12V battery and the outputs are about 6V each, the difference is very small, 0.7mV at 1atm and 12.9mV at 100psi.

When you measure these voltages, what node are you measuring them with respect to? Again, when you measured 5.4V on the output, which point did you measure that voltage with respect to? When you measured those voltages, was pin 5 connected to ground?

What is the suffix of your pressure sensor? The suffix to PX302 should be something like "-100GV" or "-1KGV".

 

[refind]

It is the 1KGV. I didn't ground Pin5, I think that's my mistake. I'll try it tomorrow to see if that's the only problem.
I feel stupid for not knowing how to hook this up, I did this in undergrad, ahhh...

P.S. On page 8 of the datasheet it says:
"The INA126/2126 can be operated from a single power supply with careful attention to input common-mode range, output voltage swing of both op amps and te voltage applied to the Ref terminal."

What does that mean?

 

[meBigGuy]

Certainly not connecting pin 5 will be a problem. Hopefully the only one :)

The statement on page 8 is referring to the fact that pin 5 and pin 12 can be thought of as reference pins. They do not carry amplifier internal bias currents, just the currents through the reference divider resistors. Look at figure 4 and 5 to see ways of playing with it.

You could connect a voltage divider across the battery to provide a reference, although that would introduce error terms as the current through the internal 40K resistors changed due to output voltage. If you supply a low impedance reference (through an op-amp for example, or a zener as shown), that would be best.

Maybe the best reference would be the average of the sensor output.