Back to Smartcitizen.me

Smart Citizen kit 2.1!

Yesterday I received an email announcing the (pre-order) for SCK 2.1 .
This looks great, so congratulations for that !

( I assume we skipped SCK2.0?)

Is there any additional documentation available on this already ?
Also, is the housing included or do I need to get that somewhere else?

Thanks for any insights,

Alexander.

Good news! But not so much activity in the last days here, except on the API:


Hi James,

But not so much activity in the last days here, except on the API:

Yes indeed, that was something I was wondering as well. I would have expected more activity and something of a more community-minded announcement if there is indeed the goal of recreating a worldwide network of those kind of measurement-nodes.

It also now feels a bit as if the hardware is (just) a project on Seeed. For example, even the information link on that webpage just refers to some document on a domain called ā€˜docsend.comā€™, instead of pointing to smartcitizen.me

does someone understend what sensors are mounted on SCK2.1 ???
VOC or NO2 ???

A brief sum:
Last summer I receved an e-mail that tell SCK2.0 will be available from September.
Until today no news on this almost dead forumā€¦ on twitterā€¦ or over e-mailā€¦
All the sensors on the world map seems offline and it is logic to think that the project was killedā€¦
Today I recevied an e-mail telling me that SCK2.1 is on preorder now. (so the SCK2.0 was cancelled?)
I follow the linkā€¦ https://www.seeedstudio.com/smartcitizen ā€¦ and seems that NO2 and CO sensors of the never sold SCK2.0 was removedā€¦ now there is a stupid VOC and CO2 sensor (Why ???).
I follow the link to the preorderā€¦ https://www.seeedstudio.com/Samrt-Citizen-Kit-p-2864.html ā€¦ the pictures and description is for a SCK2.0 with NO2 and CO sensorā€¦
The comunity is offline as usual and no answers as usualā€¦

Is this a joke ?

Will I buy a brick ?

Whatā€™s wrong here ?

Indeed, I agree it is now very unclear what will be in the new SCK2.1.
Also, it is very unclear if I will get just a standalone measurement device from Seeed, branded as ā€™ SCKā€™ , or if it will still be part of this global ā€™ Smart Citizenā€™ network ( since indeed most of the nodes are offline at the moment and activity here is minimal).

There has been some activity on GitHub, so at least some people should be aware, but also on GitHub I cannot find documentation regarding the kinds of sensors present.

The Plantower PMS 5003 as listed in:

ā€¦seems to refer to: http://www.plantower.com/en/content/?108.html

On that link, the manufacturer advertises PM2.5, so PM1 and PM10 would be derived values? It would be really great to know why this was selected. Like for the SCK 2.0, a different sensor was cherished a year ago:

https://luftdaten.info/en/construction-manual/ which seems to be a much more active project, currently uses the Nova SDS011. That has a built-in ventilator and uses a tube to suck the air into the device. Iā€™ve no idea if that sensor is better or worse, but any reasoning as to why hardware is selected would be really great.

I think not. ā€œInstalling the Smart Citizen Kitā€ in the Document section on the Seeed website refers to https://docs.smartcitizen.me/Smart%20Citizen%20Kit/#build-your-own-enclosure

Indeed, that refers to 2.0, but I guess for 2.1 no enclosure is included either.

Hi alexander!

Thank you for the message! :slight_smile:
We have been working quite hardly to get this SCK2.1 version up and running. The version SCK2.0 was indeed developed, but it was mainly used for an EU2020 European Project called iScape. Now, during this research project we have learnt quite a bit and the new kit has a few changes that we will explain in the forum and the documentation. These changes are mainly aiming to provide with more reliable (short and long term) measurements that have a more robust calibration process, and with proven performance in research literature.

As for documentation, we are working on this (beta Version here), and it will be available very soon. Very soon there will be an announcement in the forum as well!

As for the enclosure: there will be one available at seeed studio as well as a build your own version, although it has not yet been included in the seeed studio site.

Thanks again for reaching out!

Dear @aidrini

Apologies for not clarifying enough on this. The SCK2.0 was mainly a research-project version of the SCK from which we have learnt quite a lot. The SCK2.0 version was not sold openly for various reasons and due to that, the new SCK2.1 aims to improve it based on what we have learnt. For the new SCK2.1, the list of sensors is as follows:

Metric Units Sensor
Atmospheric Pressure kPa NXP MPL3115A2
Air Temperature / Relative Humidity ĀŗC / %rh Sensirion SHT31
VOCS, eCO2 ppm AMS CCS811
Noise Level and FFT Spectrum dBa TDK ICS43432
Particle Matter (PM1/2.5/10) ug/m3 Plantower PMS5003
Ambient Light lx Rohm BH12721FVC

The NO2 and CO sensors (SGX MICS4514) have been removed from the Urban board for many reasons. Mainly, the poor reliability of the measurements, the difficulty to calibrate them to obtain actual concentration values, and the battery consumption. As we have mentioned in the past, there have been some projects using the SGX MICS4514 as a low cost pollution sensor, but the list of cons from our point of view was larger than the potential gains. Instead, we have replaced it with a VOC/eCO2 sensor able to measure organic compouds and give proper concentration values. These values are also corrected for temperature and humidity effects.

That being said, we agree that a more complete sensing solution for urban air quality should have at least CO, NO2, PM and HC -hidrocarbons- (and surely we could think of many more, like O3, NH3ā€¦). Our aim is to cover the PM and HC with the combination of the Plantower PMS5003 and the AMS CCS811. For CO and NO2, we have worked on a more sophisticated board that can be plugged into the grove auxiliary port, and itā€™s based on electrochemical sensors. These sensors have been tested during the iScape project and give much more accurate measurements with much lower battery consumption. During the iScape project, they are all part of what we call the Smart Citizen Station, which is currently being deployed around Europe for evaluation. If successful, in the near future, we plan to include this board, or a single electrochemical sensor board to cope with the CO, NO2 measurements.

As for the SCK2.0 pictures that you see on seeedstudio, the thumbnails will change for the final SCK2.1, so donā€™t consider them fully representative.

Finally, as for the sensors offline, at the moment of writing this post, there are 131 kits online. I believe there is a small bug on the platform that is messing up the filters and that doesnā€™t show the online kits properly (I see them if I only unclick offline from the filters list).

And as a final comment: keep in mind that we donā€™t sell for profit, and that we would love to share the tools we build with everyone (data analysis, firmware, hardware components and so on).

Hopefully this helps! Let us know your comments!

1 Like

Dear @arjan,

Thanks for the message! :smiley:

On that link, the manufacturer advertises PM2.5, so PM1 and PM10 would be derived values? It would be really great to know why this was selected. Like for the SCK 2.0, a different sensor was cherished a year ago

The reason for this change is because the PMSX003 sensor family uses a light scattering detector, which is a much more robust method than the optical method used by the MAX30105, and has been evaluated by scientific literature with quite good off-the-shelf results. Comparing it to the MAX30105, this sensor offers better robustness and the manufacturer provides with processed results, rather than raw values that require a questionable calibration.

The PMS, as every light scattering sensor, actually does not measure PM, but it rather estimates Particle Number (PN) for different sizes, and it then calculates the PM1, PM2.5 and PM10. The sizes of the raw PN values range from 0.3um to 10um. This means, that the calculations for PM2.5, PM1 and PM10 are equally derived from the PN values. Note that I said that the sensor also estimates the PN values, because the sensor needs to somehow derive the values by the difraction of a laser, and this method, relies on particle physics that at certain sizes are not so obvious. Of course, in low cost sensors one has to make assumptions, and in the case of the PMS, the manufacturer makes assumptions for the average particle density, the shape, and even the actual diameter of the particle that the laser is hitting, and with this, obtains PM values in ug/m3. In our case, we have been/are testing the PMS5003 in the iScape project and we promise we will share the results of the evaluation as soon as we have them.

Keeping all this in mind, the sensor that our friends from luftdaten are using is really not that different, and to be honest, it would be interesting to test one against the other to see how they relate :slight_smile:

Let us know your comments! Happy to be back in the forum :smiley:

1 Like

Hi oscgonfer,

Thanks for clarifying a lot in your extensive reply!
This definitely helps understanding what is going on and what is in the SCK2.1 (your posts above might make a good basis for a blogpost on the main page I think).

Questions I do have:

  • Is it the idea that the ā€™ Smart Citizen Kitā€™ can be upgraded to the ā€™ Stationā€™ , or does one need to start with different components? ( i.e. is the Kit part of the Station, or not?)
  • And is the idea that the ā€™ stationā€™ can also be ordered in its totality?
  • Will the advanced sensor board be plugged in to the '2.1 kit ā€™ later on, or will it be part of another (later)version ? And are there any timelines for that?

Thanks,

Alexander.

Thanks for the clarification.
Iā€™m very sad to hear that I have to wait (again) for a complete shared pollution logger; I received the invitation to buy the 2.0 version on september but then there was no sellers.
Now the most important sensor is removed and replaced.

I hope will be disponible again as soon as possible.

I will wait for the 2.2 :slight_smile:

hi @alexander,

I will reply one by one:

Thatā€™s the idea. The auxiliary port of the citizen kit can be used to plug in a lot of supported sensors. Actually, the idea is that any sensor that can be connected with the grove connector could be logged into the kit. As of now, the list of supported sensors is quite long (sorry for the file format, weā€™ll put it nicer), but since the project is fully open, one can modify the firmware and include other sensors into the structure.

Regarding more advanced gas measurements, these extensions would be from our side:

  • the gases pro sensor board (either single or three-sensor)
  • the full station, which you can see below. You see that there is an intermediate board (in the center) that has a lot of ports: GPIO, SPI, I2C and the JST connector for the PMSX003 series

  • And is the idea that the ā€™ stationā€™ can also be ordered in its totality?

Yes. It will take some time for us to complete this solution, since it comprises more elements and itā€™s currently being evaluated. The price itā€™s not fixed yet either.

  • Will the advanced sensor board be plugged in to the '2.1 kit ā€™ later on, or will it be part of another (later)version ? And are there any timelines for that?

As I said above, the gas pro board can be plugged in to the 2.1 (and actually to the 1.5 as well). The timeline is not fully fixed, but around 4-6 months.

I hope this helps!

1 Like

Hi everyone,

We updated our documentation to help clarify any issues between SCK 2.0 and SCK 2.1

https://docs.smartcitizen.me/Smart%20Citizen%20Kit/

We will keep working on it during the next few weeks to improve the SCK 2.1 technical description and support.

Thank you all for your support!