Problems with sampling rate for PCB pressure sensor 105C13

[SOMA76]

Dear Henrik,

                   The sensor is powered by an external power supply of 30VDC and the output signal is connected to DAQ board through BNC2110. Please see the attached picture

[josborne]

I think I need more explaination about how you gather the data.  Are you doing this:

-- Taking 10 groups of 25000 samples at each sampling frequency.  And immediately progressing to higher sampling frequencies after each group of 10 are taken? 

Also: 

I wonder if you should be increasing the number of samples taken at each sampling frequency.  This is because your sampling period is shortening as you increase the sampling frequency.  You start by sampling for about 0.5 seconds.  And in the end you are sampling for .00625 seconds.

How do you know that the sound pressure you are measuring is the same during each sampling period?  It could be changing over time?  As mentioned, it could be resonating or something?  (sorry, not a sound expert) 

[Henrik_Volkers]

Thank for the picture, seems that you use a correct ICP/ IEPE exitation and a decoupling cap. 🙂   (30V is the V_max , you normally power the sensor with a constant current! )

To make avearage readings of a periodical signal you need to read at least one periode AND usually cut your values in a way that you have n complete periodes in you array. 

Do you use a triggered aqusition? That might explain your increasing results, shift it 1/4 of your periode and you will get a decreasing result with your approach (assuming a fractional periode capture)   🙂

If your signal is a sine and you want the amplitude and phase try the tone detection vi . RMS measurements include noise, humm and THD you maybe don't want.

But you will still have to capture at least one periode (better 10) !! 

[SOMA76]

Dear Friends,

                    I have found the reason for this problem. Th external power supply was not providing correct voltage to sensor. It should give 30VDC but it is giving 16VDC. I had 4 channels on my power supply. I connected the sensor to other channel providing 30 VDC. Using the correct supply solves the problem.

But there is one more issue. For a sound input of 20W ad 70 KHz, I expect a pressure of 124KPa according to the correlation

I = p^2/(2*density*velocity of sound in medium)

where

density of water = 998 Kg/m^3

Velocity of sound in water = 1483 m/s

But I have been able to measure only 13 KPa.

Is it the problem with the custom scale used in the VI. Please see the attachment. 

Please see the attached VI I am using and the spec sheet of the pressure sensor

[Henrik_Volkers]

Again to your sensor : The powersupply should provide a constant current! (2-20mA)

Usually the (bias) voltage at the sensor input is in a range of 10-12V (8-14V) with a connected sensor!!   If your picture is correct you have about 2m cable from the powersupply to your sensor. In that case 4mA is all you need. To test the current replace the senor with a 1k resistor and measure the voltage at the resistor . Each Volt equal one mA.

Next test: With the resistor still attached monitor your signal output. Should be a nice 0 line .

[Henrik_Volkers]

The factor of 10 in your scale: Have you calibrated the ICP powersupply/ amplifier??

Sorry no LabVIEW on my EEEPC .....

[AndreasStark]

Hi,

why don't you contact NI sales department (call me back Link on product page) to evaluate for free one of the Sound & Vibration products (http://www.ni.com/soundandvibration/d/), e.g. NI USB-9234. It comes with integrated signal conditioning and constant current excitation ready to use calibrated in a small USB form factor.

That way you can be sure everything related to your measurement hardware fits your sensor. If it is a TEDS sensor, the 9234 even reads your TEDS calibration information from your sensor.

Best Regards,

[Henrik_Volkers]

---

@AndreasStark wrote:

Hi,

why don't you contact NI sales department (call me back Link on product page) to evaluate for free one of the Sound & Vibration products (http://www.ni.com/soundandvibration/d/), e.g. NI USB-9234. It comes with integrated signal conditioning and constant current excitation ready to use calibrated in a small USB form factor.

That way you can be sure everything related to your measurement hardware fits your sensor. If it is a TEDS sensor, the 9234 even reads your TEDS calibration information from your sensor.

Best Regards,

---

The 9234 would not fit since he want to measure a 70kHz signal 😉

The signal conditioner he uses is specified up to 1MHz (-3db) ....   Wait ...

at channel 4 only!!

SOMAR

the other channels only have a 10kHz spec, so you will not be able to use them for your 70kHz signal !!!   And that might explain the factor 10 in your reading!!

Assuming you still use the amplifier mentioned here

[AndreasStark]

Well, in the first post the requirement was "For a sound input of 40KHz and 20 W".

[Henrik_Volkers]

What is the Samplingrate of the 9234 , 50kHz? Enough for 40kHz??