Sensirion sht7x labview

[meiling]

i have acquired several SHT15 sensors and i am using them to monitor temperature gradients inside a room. i am already able to retrieve data and convert them to actual values. the problem i have is i observe fluctuations of almost +/-6degC in the readings. is this normal even though i can not think of anything that might cause this sharp rise? i sample every 15seconds.

here's a sample of several datapoints i have:

counter, LM35, SHT15

...
12,22.070, 33.160
13,22.119, 33.200
14,22.119, 26.960
15,22.070, 25.720
16,22.070, 25.080
17,22.070, 31.000
18,22.070, 25.720
19,22.021, 24.840
20,22.021, 30.720
21,22.021, 31.760

thank you.

[TheWoost]

Hi meiling,

I'm not entirely familiar with how these sensors work, but it looks like they transmit the temperature data in the form of digital signals.  What data acquisition device are you using to read those signals?  Is it possible that the environment in which you have your test setup could be introducing noise on your digital data lines that translate into variations in temperature level?

Regards,
Andrew W
National Instruments

[Henrik_Volkers]

Andrew, meiling,

information about the sensirion sensors moduls can be found here:

http://www.sensirion.com/en/02_sensors/03_humidity/00_humidity_temperature_sensor/05_humidity_sensor...

Output is a digital datastream, that can be read via DIO lines. Since the communication protocol includes a CRC mechanism digital read errors can be avoided. I hadn't looked into the provided vis, if the CRC is used....

One reason for the temp. drift might be the build in heating resistor to dry a drown humidity sensor, make shure it's turned OFF

 And it's recommended that the sensor is turned of between measurements to avoid selfheating.

Message Edited by Henrik Volkers on 05-04-2007 02:56 PM

[bobcdy]

I usually use  signal express at  my university but I'm trying to learn labview because I can't get SE to recognize the sensirion sensor.  Recently I developed a vi (using M. Lutz's drivers, and using a variety of vi's from the NI forums to kluge together a working vi) for the sensor that does everything I need, but I've been unable to implement a real-time sampling period shut off timer - one that can be set for the sampling period, as for ex. for two hours.   In my attached vi,  the timing block is affected by the 'wait until next..." timer module (used to set the timing for seqiential sampling events) so that only relative time can be set.  Therefore the vi requires a table in order to choose the relative time limit and loop period, both of which affect the length of time that the cycle takes to complete, in order to set a specific sampling period.  I've been unable to isolate the timing block from the timer module.

I wonder if one of the more experienced labview users would be willing to look over the vi and suggest how to isolate the timing block from the time module?

Bob Cody (Dr. Robert D. Cody, Iowa State University)

[max310]

Hello,

I tried to make the code you wrote for Labview 8 (main.vi or main2.vi) work but I didn't succeed. Did you modify it somehow to make it work? Or is there a new version of this program?

Thank you very much.

[bobcdy]

I've given up on my previous vi posted here because I concluded that I'd never be able to get it to work as I wanted.  Attached is another kuge that seems to do what I need.  I've attached it, PCI_sensirion.vi, and also one sub vi, Dew Point.vi that it uses together with a modified sht7x that I had used in the PCI_sensirion.vi.   Perhaps the trouble you had with my older vi was that it probably used the modified sht7x included here.  You're welcome to use any or all of these, and I'd certainly appreciate any improvements, streamlining, or modifications to extend its usefullness.  If you have any questions, let me know.

[bobcdy]

My attachment did not seem to go through??  I'll try again.

[bobcdy]

I forgot to mention that I used a EI-1050 temperature/humidity probe sold by Labjack (http://www.labjack.com/results.php?category=5) rather than the bare sensirion sensor, with a NI PCI-6221. It also works fine with a PCI-6259 board  There's a data sheet for the probe on the Labjack site.  According to the manufacturer rep, the EI-1050 is the sensor encased in a plastic sheath with four leads, but I have no idea whether there could be problems with my vi if you are using wiring the bare sensor only.
Bob 

[max310]

Hello!

  I just tried to get your new version of the SHT-Softwar running. Sadly it won't work ... does not cause any LabVIEW error messages, but I always get a timeout when trying to measure either temperature or humidity using s_measure.vi. The DAQmx_read constantly returns logical TRUE which makes the while-loop run through all 39 cycles and close with a timeout.
  Is there any explanation you can think of for this behaviour? Apart from that everything is running quite fine from a Software point of view.
  Oh yes, and I changed the Lines for the Task-creation in DAQ_Initialize2.vi ... Write and Read shared Dev0/Port0/Line1 - made it Dev0/Port0/Line1 for Write and Dev0/Port0/Line2 for Read. But it did not work this way or that. I would be very grateful for any advice you could give ...
 
Best regards!

[LabJackSupport]

Besides the physical construction, the biggest difference with the EI-1050 is that it adds enable/disable capability to the SHT15 sensor so you can use multiple with the same data and clock lines.

+/- 6 degrees C is definitely too much variation.  I would expect the data to look like your LM35 data, so there is either something wrong with the EI-1050 or something wrong with your software.

One thing I might suggest is that you could buy a U3 from us for about $90 (after your educational discount), and then just run our LabVIEW example to get temp and humidity readings.  In fact, someone at ISU probably has a U12/U3/UE9 that you could borrow if you want to just test the probe.