Go Down

#### panoulisk

##### Sep 18, 2013, 11:44 pm
hi to all, its my first time to write on the site....
my problem:
i have 1 arduino due and i want to take measurements from two sensors

1 linear transducer (input-12v, output- 0-10v) elpt300
http://www.opkon.com.tr/pdf/ELPT_Ver-070213%281%29.pdf

1 piezoresistive transducer(pressure sensor input 12v output 4-20ma)

i tried to match the output from the linear transducer to the arduino by making a voltage divider(1 resistor of 56kohm , and 1 resistor of 24kohm) the output was 0-3v but there was a huge loss in accuracy (saying varying 40-60 values in 10bit resolution when keeping steady the rod of the sensor)

for the other sensor i have a shunt resistor of 250ohm (accuracy 0.05%) making the output 1-5v
i bought 1 resistor of 150 ohm (1% accuracy) hoping to to make the output 0.6-3v havent tried it yet

i have searched the web for a better solution (such as an ic)but with my limited knowledge of electronics i couldnt find any solution

so i was wondering if anyone has any suggestion or correction or could tell me if i am on the right track

ps: i need high accuracy at least on the pressure sensor because the values i will need wont exceed the 30-40% of the max pressure of the sensor

panoulis

#### noblepepper

#1
##### Sep 19, 2013, 12:04 pm
Your setup sounds correct. The "loss in accuracy" may be noise.

My first suspect would be the power supply to the transducer. To check this change the 56k to 120k, and move the connection from the output of the transducer to the 12v power connection. Ideally you should see a steady 2v on the input, if you still see variation then the power supply is the culprit.

#### chriskner

#2
##### Sep 19, 2013, 03:11 pm
Hi panoulis,

--Voltage divider Input --
Feeding an A2D input with a voltage divider is almost always going to result in noisy conversions.

Run the output of the voltage divider into an op-amp with gain = 1 (simplest solution).  This transforms the signal that you are supplying to the A2D from a high-impedance source (undesirable), to a low-impedance source that can adequately drive the A2D input.

There a considerations to make when selecting a good op-amp that works with the supply voltages that are available, and the signal ranges that you are working with.  That is a whole topic on it's own, better answered elsewhere.

-- 4..20mA conversion --
Using a properly-sized resistor to scale 4..20mA to the 0-2.5V input of the A2D-range is ok for prototyping (and will probably result in adequate conversions without noise).  The danger is that the gadget that supplies the 4..20mA signal is probably capable of delivering > 3V.  This is dangerous because if there is a fault in wiring (open-loop), then the gadget may apply over-voltage signals to the DUE, and destroy it.

Find yourself yet another buffer circuit that does the conversion, and limits the input voltage to a safe range (as everything in life, this is easier said than done).

The moral of the stories:  Always use the proper tools for the job, even if it means running to the store.

Regards,

Chris

#### panoulisk

#3
##### Sep 22, 2013, 11:23 am
thank you all for your replies, i will try your suggestions and if i have any progress i will post it..

panoulis

Go Up