Skip to content

Frequently Asked Questions

uThing's integration

I'm using Linux and I get "Permission Denied" when I try to access the serial port

You have to give your user permission to read and write to the port. Please check here for more details.

How can I send the dongle data over MQTT?

Please check the uBridge and the MQTT plugin for more details.

What's the easiest way for logging data?

Please check the uBridge and the CSV logger plugin for more details (most robust solution for long term logging).

Alternatively, for short time logging, the output data can be sent to a plain text file with the following command: $ cat /dev/ttyACM0 > activity.log

Additionally, if the “&” symbol is added at the end of the command, the process will be forked, and the system will keep storing the data until the user logs out or the dongle is disconnected. This trick could be very useful for quick datalogging. By first configuring the CSV format, then the generated file can easily be opened in Excel or any spreadsheet application.

How can I build a datalogging and data visualization solution for a PoC?

Please check the uBridge and the InfluxDB plugin for more details on how to configure this database + dashboard / analytics solution.

uThing::VOC

Can I use the device outdoors?

You will need to consider 2 things:

  • Water-proofing. Obviously, you will have to consider how to install the board + host in a way that rain won't reach the electronics while still keeping a reasonable amount of air-flow into the sensor.

  • The concentration of VOCs is usually much higher indoors than outdoors, due to products like cleaning, paints and even the varnishes from some furniture. The BME680 is also relatively sensitive to CO2 from breathing, so you will notice the difference in a closed small room with a few people after a few minutes. Outdoors, the VOCs concentration is much lower in general, unless you have some factory or storage facility close-by emitting these gases. For outdoor air quality, what is usually interesting to measure is the particulate matter (2.5-10 um in size) generated by cars exhaust, dust storms, fire, and other sources. To measure particulates, the most common sensors are based on photo-optical absorption.

Why I'm seeing an offset between the reported and ambient temperature?

Regarding the temperature offsets (and humidity, since RH is highly dependent on the temperature):

Because of heat conduction through the PCB, the temperature seen by the sensor will change when you plug more devices to an USB hub, or if you have it directly connected to a Raspberry for instance, the heat generated by the CPU highly depends on its load, so a "perfect" calibration with ambient temperature will be even impossible in some cases.

The device comes with a fixed calibration performed while connected to a 4-ports USB hub, but you can use the {"temperatureOffset": -.-} command to calibrate against ambient-temperature for your specific setup.

If precise temperature values are important for your use case, the device can be connected through an USB extension instead, reducing the heat transfered from the host device. In addition, you can disable the LEDs with the command {"led":false}. The power used by the Blue LED is not much (~10mW) but is close enough to the sensor, so it may help to reduce the self-heating.

Remember that you can save the configuration in Flash to make it permanent with {"saveConfig":true}

Is there a plastic casing for the uThing::VOC?

Although there is no injection-molded case for this board, we have designed a 3D printable model which could be useful if you have access to a 3D printer or a printing service. The case consists of a top and bottom parts which snap together with small snap-fit features.

Check here for the design files.

Between multiple units, why the raw resistance seems to be quite far from each other?

The heated element (hotplate) is designed in a way that its resistance will change in a very wide range due to varying VOC concentrations (the higher the concentration of reducing VOCs, the lower the resistance and vice-versa). Since this raw signal is influenced by parameters other than VOC concentration (e.g. humidity level), the raw values are transformed to an index for air quality (IAQ) by the BSEC algorithms. BSEC keep track of the changes in this value over relatively longs period of time (days) to obtain baselines to improve the IAQ estimation.

uThing::MNL

Can I use the device outdoors?

You will need to consider a few things:

  • Water-proofing. Naturally, you will have to consider how to install the board + host in a way that rain won't reach the electronics while still keeping an opening for the noise sensor.

  • The PIR sensor will not perform good when is directly exposed to sunlight (it will have too many false triggers).

  • For the luminosity sensor, a transparent plastic window could help to expose it to sunlight.

uThing::iPH

Why the pH / voltage readings are slowly drifting?

This is most likely caused by the sensing electrode, not by the meter itself:

For instance, it may take 10 to 20 minutes for the sensor temperatures to equilibrate, if the glass temperature is different than the compensation sensor temperature, the output value will show an offset.

The sensor voltage depends on the pH of the liquid at the surface of the pH sensitive glass. If the bulb is dirty or fouled, it may take some time for the liquid to diffuse through the coating and reach the membrane. The response of the sensor to step changes will be slow, and readings will drift to the final values.

Memory of past junction potentials can also lead to drift. When a pH sensor is transferred from one sample to another, liquid that has diffused into the reference junction is carried with the sensor. When the sensor is first placed in the new solution, both the old and new solutions determine the junction potential. As the new solution diffuses into the junction and the old solution diffuses out, the liquid junction potential gradually changes. The changing junction potential causes the pH reading to drift. The effect is most severe when the ions involved have substantially different mobilities. If the solutions are more or less equitransferent, the memory effect will be relatively small. The type of junction also influences the amount of drift. In flowing junctions, the filling solution washes out the previous sample, so the time for the memory to disappear is a few minutes to an hour. In gel type sensors, diffusion alone is responsible for movement of ions through the junction. Therefore, the time required for the memory to dissipate can be quite long.

I'm getting noisy readings even with a new electrode, what can I do?

In some industrial environments when there is high-power equipment operating close to the sensor, EMI can induce noise in the probe cable due to the very high impedances of the electrodes and the measurement input (GigaOhms range). Place the coax cable far as possible from AC sources. Alternatively wire the cable inside a metalic conduit connected to earth on the probe side.