pH reading changes when another sensor is plugged/unplugged to/from the SmartWater P&S

Post Reply
nachorro
Posts: 13
Joined: Tue Apr 18, 2017 12:06 am

pH reading changes when another sensor is plugged/unplugged to/from the SmartWater P&S

Post by nachorro » Wed Sep 11, 2019 8:17 pm

Hi, good morning.
I want to raise an issue we have with our SmartWater: every time we plug/unplug a sensor to the SmartWater Plug & Sense, the pH readings change. Moreover, if the SmartWater is being plugged/unplugged to/from an external power supply (SmartWater connected to a PC via the USB port), the readings also change.

Could you please let me know how to eradicate such an annoying behavior? Firstly, because readings are incorrect and secondly, because it is nearly impossible to calibrate de sensors, since the calibration values obtained when running your calibration sketches (mv, ohms, etc.) depend so much on the number of sensors connected to the Smart Water.

https://drive.google.com/file/d/1zCtCra ... eXENVze35j

Test environment:
- SmartWater running v15, waspmote-pro-ide 6.16, API libraries 0.42.
- Sensors are placed in a bucket with tap water. Bucket is big enough so as the sensors are far away from each other to avoid field interactions among sensors.
- Water temperature (blue line in the graph) is under control, 23.25 ºC +/- 0.25 alongside the duration of the test.
- The time lapse of each zone was 30 min, to let sensors to get stabilized.

Test chronology:

- ZONE_1
During a whole night, the SmartWater is running with four sensors connected: Temp, pH, ORP and Conductivity.
SmartWater is connected via USB port to a laptop.

- ZONE_2
At 10:06, I disconnect the ORP sensor from the SmartWater. No extraction of the sensor from the water, just the sensor is unplugged from the SmartWater.
Water conditions remain the same, no agitation. Sensors remain in the same position.
Temp, pH and Conductivity sensors remain plugged to the SmartWater.
But pH decreases.

- ZONE_3
At 10:36, I disconnect the Conductivity sensor from the SmartWater. No extraction of the sensor from the water, just the sensor is unplugged from the SmartWater.
Water conditions remain the same, no agitation. Sensors remain in the same position.
Temp and pH sensors remain plugged to the SmartWater.
But pH increases.

- ZONE_4
At 11:06, I disconnect the SmartWater USB cable from the laptop: SmartWater in not being powered from outside, it is running out of its internal battery.
Water conditions remain the same, no agitation. Sensors remain in the same position.
Temp and pH sensors remain plugged to the SmartWater.
But pH keeps increasing.

- ZONE_5
At 11:36, I plugged the Conductivity sensor.
Water conditions remain the same, no agitation. Sensors remain in the same position.
SmartWater is not being powered from outside, it is running out of its internal battery.
Temp and pH sensors remain also plugged to the SmartWater.
But pH keeps increasing.

- ZONE_6
At 12:06, I plugged the ORP sensor.
Water conditions remain the same, no agitation. Sensors remain in the same position.
SmartWater is not being powered from outside, it is running out of its internal battery.
Temp, pH and Conductivity sensors remain also plugged to the SmartWater.
But pH keeps increasing.

- ZONE_7
At 12:36, I plugged the SmartWater to the PC via USB cable.
Water conditions remain the same, no agitation. Sensors remain in the same position.
Temp, pH, Conductivity and ORP sensors remain also plugged to the SmartWater.
But now pH decreases.

As you can see, it is impossible to get a stable reading of the pH despite having a constant water temperature. Depending of the ZONE, it varies from 5.8 to 7.9. Wrong business decision can be taken when pH is varying so much depending on the number of sensors connected or on whether the SmartWater is being powered from outside.

Can you please help me to eradicate such an erratic behavior from the SmartWater?

Thank you very much for your help beforehand,

As for reference, I am attaching also the behavior of the Conductivity and ORP sensors, same time scale. It shows the moments when those sensors are plugged/unplugged to/from the SamrtWater.

https://drive.google.com/file/d/1s5xqPw ... frqqBaLPOW

libelium-dev
Posts: 27967
Joined: Mon Sep 28, 2009 1:06 pm

Re: pH reading changes when another sensor is plugged/unplugged to/from the SmartWater P&S

Post by libelium-dev » Thu Sep 12, 2019 2:31 pm

Hi,

Thanks for the detailed explanation.

The ground isn't isolated in Waspmote, so there are small voltage variations when sensors/usb are connected/disconnected. The variations are millivolts but we agree that some millivolts can affect to the water sensors since the voltage range varies very little.

Our advice would be to calibrate the sensors in the same condition that they will be used. So, connect all sensor to the node, disconnect the usb and send the calibration values through the communication module. Once the values will be stable, you can connect the usb again, update the calibration values and then disconnect the usb.

Regards

nachorro
Posts: 13
Joined: Tue Apr 18, 2017 12:06 am

Re: pH reading changes when another sensor is plugged/unplugged to/from the SmartWater P&S

Post by nachorro » Thu Sep 12, 2019 7:50 pm

Thank you very much for your fast answer.

I appreciate what you say, but milivolts variations when sensors/usb are connected/disconnected is spoiling any reading, be it in calibration mode or in operation mode, since in most cases the sensor readings themselves are in milivolts.

Let me give you some customer feedback to your proposal:
  • I am a bit susprised that we are the first customer sufferig such a problem. Or all your Smart Water customers should be affected by these mvolts variations, and the readings will all be wrong.
  • If your statement is so critical, "milivolts variations exist when sensors/usb are connected/disconnected", I have read nowhere anything in your SmartWater documentation related to this issue.
  • All the sketches that you have developed and that are available to download from your web to calibrate sensors, all use the USB port as a serial port to monitor the readings. This means that you are powering the SmartWater while calibrating the sensors, therefore, having incorrect readings. If your proposal would have been so critical, you would have not developed the sketches to run using the serial port or at least you would have mentioned a note of precaution.
  • Your suggestion to calibrate the sensors under the same condition which the equipment will be used, it is a bit unrealistic. To start with because the equipment will initially be used without external power supply, and only when battery volts decay, it will be recharged via the USB connector. Therefore, readings while in operation mode will vary depending whether the SmartWater is powered by its iternal battery or if it being charged from outside.
Having said that, I can hardly believe that the equipment is so sensitive as to have rotten readings depending on the number of sensors/USB port connected.

Alternatively,
  • Can we think that the problem might be somewhere in the SmartWater's internal hardware?
  • Can this problem be escalated to your 2nd level support to have a second opinion in the diagnostic analysis?
  • What kind of tests should I perform to determine where the problem is?
  • Can you check my SmartWater to confirm it is within quality control limits?
I would very much appreciate if Libelium could reconsider this ticket based on the thorough battery of tests performed prior to the submittion of the ticket.

Thanks very much for your help.

libelium-dev
Posts: 27967
Joined: Mon Sep 28, 2009 1:06 pm

Re: pH reading changes when another sensor is plugged/unplugged to/from the SmartWater P&S

Post by libelium-dev » Mon Sep 16, 2019 2:48 pm

Hi,

We think that your board isn't damaged, all devices are affected for those small variations of tension. The ground of waspmote isn't isolated and when an external power supply is connected it can vary.

We always recommend that the customers calibrate the sensors in the same conditions that will be installed. Our example shows how to calibrate the sensors through the USB because it's an easy way to show the values. When the customers ask for more information we recommend sending the values through the radio module that the node includes.

In addition, the sensors are chemical and degrade over time. The variations in measurements are greater when the sensors degrade.

Regards

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest