[Gases Board] strange behavior and wrong values..?

Justin Time · **Joined:** Fri Jan 06, 2012 4:11 pm **Posts:** 10

Hi *,

I have some questions regarding the (gas)sensor board..:) Maybe there are some more problems which have to be solved..but to be sure, I start with the following first.

I've tried to use this board with an O2 sensor. However, it seems to be that the board is not working correctly. According to the manual, I have to write the following lines to get the voltage value from the sensor:

Code:

SensorGas.setBoardMode(SENS_ON);
SensorGas.configureSensor(SENS_O2, GAIN); //GAIN = 1
value = SensorGas.readValue(SENS_O2);

Unfortunately, this example doesnt work for me :D I've observed that the board is rebooting every time when executing this instruction:

Code:

SensorGas.setBoardMode(SENS_ON);

In the loop(), I've surrounded this instruction with some println()-commands to test whether it is really this command which let the board crash/reboot. And yes, it is. However..the funny thing is: after putting SENS_OFF instead of SENS_ON into the argument, everything is working fine! Ok, I just get 0.00000..... but this is quite ok in my oppinion, if the board is turned off...

But now things are going to be more interesting:
I do not get every time 0.0000..sometimes I get stuff like 0.0096xxx, 0.003xxx, 0.006xxx..which means 9,6mV for example..and this is in the range of what the sensor does provide. And even better: After pressing (and holding) the board at the side where the pins are connected to the 'main-board', I get much higher values like 0.13xxxx (-> over 100mV which is not in the sensors range anymore). In addition, I figured out that these values are completely independing of the GAIN factor.

Therefore, I've currently the following questions:

-Why is it not possible to use SENS_ON?
-Why do I still get values, even the board is apparently turned off?
-Why do I get complete wrong and senseless data after pressing on the board?
-Why does the GAIN not affect the measured values?
-Why is SENS_ON defined to be 0 (zero) and SENS_OFF to be 1 :D

I do own more boards of the base and sensor boards..but I've observed this behavior on all of them :(

I hope that someone has an idea and can help me..:)

Thanks a lot,

Stephan

libelium-dev · **Joined:** Mon Sep 28, 2009 1:06 pm **Posts:** 7815

Dear Justin Time,

We guess that maybe your problem is an I2C interference, due to the O2 sensors uses this bus.

Try adding RTC.ON() just after setBoardMode, in the setup loop.

Quote:

// Configures Gases sensor Board
SensorGas.setBoardMode(SENS_ON);
// Init RTC
RTC.ON();

Also configure the sensor before read it in this way:

Quote:

// Configures O2 sensor
SensorGas.setSensorMode(SENS_ON, SENS_O2);
SensorGas.configureSensor(SENS_O2, GAINO2);
valueO2 = SensorGas.readValue(SENS_O2);

Try to do this things before going further into your problem and going one by one to your questions.

Let us know your comments.

Kind regards.

Justin Time · **Joined:** Fri Jan 06, 2012 4:11 pm **Posts:** 10

Thanks for your fast response :)

I've tried to follow your suggestion, however, the problems are still remaining: every time when executing the method setBoardMode(), the board is rebooting. Again, I've surrounded every instruction with a println():

Code:

  USB.println("Hello World, this is Waspmote!\n");

  USB.println("test1");
  SensorGas.setBoardMode(SENS_ON);
  USB.println("test2");
  RTC.ON();  
  USB.println("test3");
  SensorGas.setSensorMode(SENS_ON, SENS_O2);
  USB.println("test4");
  SensorGas.configureSensor(SENS_O2, GAIN);
  USB.println("test5");

As a result, I still just get "test1" as output (and in the next line a "ÿ"..I don't know from where this is coming from):

[quote=COM8]
Hello World, this is Waspmote!

test1
ÿ Hello World, this is Waspmote!

test1
ÿ
....
[/quote]

However, as soon as I change the parameter SENS_ON to SENS_OFF, every println-instruction is going to be executed. I also tried to put the RTC.ON() to the start of the programm..but this doesn't work, too.

Do you have some more ideas? :)

Kind regards

Stephan

libelium-dev · **Joined:** Mon Sep 28, 2009 1:06 pm **Posts:** 7815

Dear Justin Time,

It seems to be an I2C bus interference problem, due to the electronic involver in O2 sensor.

When you use SENS_OFF, you are not using this electronic, therefore interference disappears. However, you have not powered the O2 sensor correctly.

This is a hardware issue, and we would like to ask you a few question more before start a RMA process to repair your hardware.

- Do you have other Waspmote to test if the problem is still present? Just to be sure that the problem is only in one of your Waspmotes.

- What happens if you unplug the Gas board and run the code? Is it still blocking?

We will wait to your answers.

Kind regards.

Justin Time · **Joined:** Fri Jan 06, 2012 4:11 pm **Posts:** 10

Ok, I've just tested a bit more..here are the results :)

Conditions: Three waspmotes (v1.1) and two gas sensor boards.

Result: I tested every combination of these two components but, unfortunately, it was not possible to get at least one of them working correctly. However, one combination worked temporary..but just once and only for a short time. The interesting point here is that I did not get every time the same failure. Either it was not even possible to upload the code (due a sync error), or the strange behavior occurs, as described above. However, if the code is going to be uploaded while the sensor boards aren't attached to the waspmotes, everything worked fine and all of the println() instructions were executed.

Hope that helps.

Kind regards,

Stephan

libelium-dev · **Joined:** Mon Sep 28, 2009 1:06 pm **Posts:** 7815

Dear Justin,

One last question before applying for an RMA process,

What happens if you unplug the Gas board and run the code? (with Waspmote code previously successfully uploaded)

Is it still blocking?

Kind Regards

Justin Time · **Joined:** Fri Jan 06, 2012 4:11 pm **Posts:** 10

Dear libelium-dev,

as I said before, it seems to be that the code is running 'correctly' after detaching the gas sensor board. However, I still get pseudo values nevertheless when using the readValue() method (a bit strange for me). In addition, it is sometimes not even possible to upload the code while the gas sensor board is attached to the waspmote.

According to my failure description, I would guess that only the hardware of the gas sensor boards is defective. So, do I have to send just these sensor boards to you or all of the waspmotes, too?

Kind regards

Stephan

libelium-dev · **Joined:** Mon Sep 28, 2009 1:06 pm **Posts:** 7815

Dear Justin Time,

Thanks for answering.

After this conversation, finally it seems that there is some hardware issues.

We are asking you to contact technical service here and ask for a RMA number and repair your Hardware.

They will tell you what you have to do.

Kind Regards.

Wirthm_U · **Joined:** Tue Jan 31, 2012 3:57 pm **Posts:** 36

I got a similair problem.

If I upload the "wasp_start_program_full_ZB_v1" to the Waspmote I get correct answers with the serial monitor either having the Waspmote directly on the PC or wirelessly via gateway and coordinator ZBee:
-mac:0013A20040714C00 -x:27,y:45,z:1034 -temp:27 -bat: 92%
-mac:0013A20040714C00 -x:-21,y:47,z:1027 -temp:27 -bat: 92%
-mac:0013A20040714C00 -x:30,y:47,z:1034 -temp:27 -bat: 91%
-mac:0013A20040714C00 -x:30,y:47,z:1034 -temp:27 -bat: 91%
-mac:0013A20040714C00 -x:30,y:47,z:1034 -temp:27 -bat: 92%
-mac:0013A20040714C00 -x:30,y:47,z:1035 -temp:27 -bat: 92%
etc....

If I just attach the Gas Sensor Board using the same conditions (same code) I get a few cryptic symbols and then communication stopps:
~ & }3 @qL B }3 @qL

Taking off the Gas Sensor Board resolves the problem.

libelium-dev · **Joined:** Mon Sep 28, 2009 1:06 pm **Posts:** 7815

Hi Wirthm_U,

yes, it is a normal behavior, since in the Waspmote start program doesn't turn on the sensor board the I2C bus won't work properly.

What are you trying to test with this code and the sensor board?

Regards.