Topic: I'm designing a new air quality monitor!

[yan]

Hey all,

My name is Yan, I am a mechanical engineer, and I am new to the Wxforum.

I have been working on an open-source air quality monitor for the Baltimore Open Air Project https://github.com/baltimoreopenair, a community based air quality monitoring network. I am now working on a new device to make air quality monitoring less expensive and more available for everyone.

I would like your help to decide which features are most important, for example, measured species, connectivity, and access to the data.

Right now our stations measure PM, T/Rh, Ozone, Nitrogen Dioxide, Hydrogen Sulfide and Sulfur Dioxide. I have put together a website https://forcasters.landinglion.com/mistralbytroposphere/ that shows the specs and design of our new station and optionally allows you to reserve an early customizable outdoor station as well as our new indoor model.

Please put your suggestions in the comments, It will be a big help to us in our ongoing design efforts.

Happy Air Forecasting!

Yan

[Bushman]

Looks awesome, but  over $600 USD with an indoor unit?  Ouch.

[yan]

Looks awesome, but  over $600 USD with an indoor unit?  Ouch.

Hey Bushman, I hear you. This is mainly due to the fact that good gas sensors are fairly expensive and that for small production devices, the cost of FCC or CE certification is very high. I am also making most of the production in the US to be able to calibrate each device.

If you don’t live downwind from a power plant, a factory, a water treatment plant, a volcano or a swamp you probably don’t need to have H2S and SO2 in the outdoor device, except if you want to know on what day your neighbor is emptying his septic tank. PM, NO2 and O3 are enough in most cases. Assuming that, you end up with a general device at $245 which tells you about T/rh, PM (1, 2.5 and 10), O3 and NO2.

What are your thoughts?

[johnd]

I'd be most interested in a sensor with an SDI-12 interface and minimal power requirement (eg 1mA @3v or something similar) so that it could in principle hook directly into other data handling platforms. I suspect that this might well appeal to users who already have a weather station (as well as others too of course) so Temp/RH might be irrelevant for some users, but presumably cheap to add so why not.

[mcrossley]

John, I think they use the T/H internally to apply calibrations to the sensor outputs, so required anyway.

[yan]

I'd be most interested in a sensor with an SDI-12 interface and minimal power requirement [...]

That's an interesting idea. We could technically add that serial port to the device to broadcast raw data. What data platform are you currently using?

The raw data from our station is currently sent through wifi and is calibrated/corrected over the internet. Would access to wifi difficult to handle?

John, I think they use the T/H internally to apply calibrations to the sensor outputs, so required anyway.

That's exactly right. We need measurements of T/H for correction and the closer to the gas sensor, the better. We actually do not perform correction internally but over our internet server and provide calibration coefficients back. The reason for that is that we would like to be able to improve our correction algorithm over time and use the history of the sensor in the calibration. Is internal calibration is important to you?

[weather34]

good morning

this looks promising I have been on the edge for sometime to purchase a AIR QUALITY SENSOR i did order one for our home indoor use which is stuck in customs .However do you have an API or raw data access as I would introduce the option for the template I design . example of using purple air currently alongside weather station data .

do you foresee any problems shipping to Turkey if I decide on one and are there any realtime internet display of these in use .as I said been on the edge for a while but sooner or later I will buy one .

air quality simple display in bottom right hand corner and yes its color coded according to the familiar air quality colour schemes



brian

[spweather]

I would definitely be interested in this if I was given the capability to display the data on my own website. With Ken's Saratoga Templates or a standalone website.

Dennis

[Bushman]

You need to allow a standalone (i.e. not connected to your server) version - too much cloud stuff does not work with remote areas and they seem to have a way of being locked to a vendor.

[yan]

this looks promising I have been on the edge for sometime to purchase a AIR QUALITY SENSOR i did order one for our home indoor use which is stuck in customs .However do you have an API or raw data access as I would introduce the option for the template I design . example of using purple air currently alongside weather station data . do you foresee any problems shipping to Turkey if I decide on one and are there any real-time internet display of these in use .as I said been on the edge for a while but sooner or later I will buy one . air quality simple display in bottom right hand corner and yes its color coded according to the familiar air quality colour schemes

Hi Brian,

For the moment I did not build the data layer interface (our Baltimore experiment has a real time data base but no API). I will probably just have an API to query real-time data in Javascript that could give out JSON files and a turn-key embedded script that could fit in any html page, with color coded icon for various averages.

For sending the device to Turkey, I suggest using services such as Aramex shop and ship that could allow you to have a PO box abroad (in the USA in particular). This can be a bit expansive but alleviate issues at customs.

For our current device, I will need at least 100 beta customer to be able to deliver the outdoor unit (after that I will need 6 weeks of lead time to produce). I am currently focusing on the indoor unit (see https://www.tropo.cc/), which should ship out in 8 weeks.

[yan]

I would definitely be interested in this if I was given the capability to display the data on my own website. With Ken's Saratoga Templates or a standalone website.
Dennis

Yeah I will make it work with any html page. Thank you for Ken Saratoga Template pointer, I'll be sure to make it compatible.

You need to allow a standalone (i.e. not connected to your server) version - too much cloud stuff does not work with remote areas and they seem to have a way of being locked to a vendor.

Hey Bushman,
That's a good point. I have been racking my head over this one. The advantage of the server is in pushing code updates and pushing correction coefficients on the gas sensors to extend the time between two calibrations.

That being said, each device act as a wifi hub, meaning you can directly connect to it and see and download the data with your phone or any wifi enabled device whether or not you have access to the internet. It is also possible to use a usb cable to get the data over serial. The calibration and correction coefficient will just be static and relate to initial calibration.

Yan

[daman]

Hey all,

My name is Yan, I am a mechanical engineer, and I am new to the Wxforum.

I have been working on an open-source air quality monitor for the Baltimore Open Air Project https://github.com/baltimoreopenair, a community based air quality monitoring network. I am now working on a new device to make air quality monitoring less expensive and more available for everyone.

I would like your help to decide which features are most important, for example, measured species, connectivity, and access to the data.

Right now our stations measure PM, T/Rh, Ozone, Nitrogen Dioxide, Hydrogen Sulfide and Sulfur Dioxide. I have put together a website https://forcasters.landinglion.com/mistralbytroposphere/ that shows the specs and design of our new station and optionally allows you to reserve an early customizable outdoor station as well as our new indoor model.

Please put your suggestions in the comments, It will be a big help to us in our ongoing design efforts.

Happy Air Forecasting!

Yan

But then you have it at $600USD? 

Come on you need to do way better then that.

[yan]

But then you have it at $600USD? 
Come on you need to do way better then that.

.

Hey Daman,
As I said earlier that's not the price of the device I am proposing, but $245 for PM, O3 and NO2. Only if you select all the sensors add SO2 and H2S, make all the sensors redundant and use solar power that you reach that price.

The price is mainly driven by sensor cost, low scale production and certification cost, while I will pay myself (assuming I am actually selling something  ) this year 5 times less than what I can make as a researcher to make this project succeed.

At $245, I am reaching a comparable price than Purple air (which contains $50 of electronics and measures only PM) and lower than the air quality egg (that measures one gas, but is now subscription-based). I am also calibrating each and every sensor, which none of the affordable competitor do (you would have to pay 3k for a calibrated device from industrial sellers).

I hope I will be able to give a better pricing in the future, and I would be grateful for feedback on how to do that.

[weather34]

this looks promising I have been on the edge for sometime to purchase a AIR QUALITY SENSOR i did order one for our home indoor use which is stuck in customs .However do you have an API or raw data access as I would introduce the option for the template I design . example of using purple air currently alongside weather station data . do you foresee any problems shipping to Turkey if I decide on one and are there any real-time internet display of these in use .as I said been on the edge for a while but sooner or later I will buy one . air quality simple display in bottom right hand corner and yes its color coded according to the familiar air quality colour schemes

Hi Brian,

For the moment I did not build the data layer interface (our Baltimore experiment has a real time data base but no API). I will probably just have an API to query real-time data in Javascript that could give out JSON files and a turn-key embedded script that could fit in any html page, with color coded icon for various averages.

For sending the device to Turkey, I suggest using services such as Aramex shop and ship that could allow you to have a PO box abroad (in the USA in particular). This can be a bit expansive but alleviate issues at customs.

For our current device, I will need at least 100 beta customer to be able to deliver the outdoor unit (after that I will need 6 weeks of lead time to produce). I am currently focusing on the indoor unit (see https://www.tropo.cc/), which should ship out in 8 weeks.

good morning thanks for the reply , i guess the shipping to istanbul will,deter me there is an easier method however likelihood i will,keep on looking for product releases its becoming a popular addition to weather station hardware .

but yes an api in json format is ideal or raw data access locally . either way is useable and easy to adapt to .

good luck i hope it becomes popular and a useful addition for weather enthusiasts..

brian

[Intheswamp]

Looks awesome, but  over $600 USD with an indoor unit?  Ouch.

Hey Bushman, I hear you. This is mainly due to the fact that good gas sensors are fairly expensive and that for small production devices, the cost of FCC or CE certification is very high. I am also making most of the production in the US to be able to calibrate each device.

If you don’t live downwind from a power plant, a factory, a water treatment plant, a volcano or a swamp you probably don’t need to have H2S and SO2 in the outdoor device, except if you want to know on what day your neighbor is emptying his septic tank. PM, NO2 and O3 are enough in most cases. Assuming that, you end up with a general device at $245 which tells you about T/rh, PM (1, 2.5 and 10), O3 and NO2.

What are your thoughts?

Swamp???? 

[yan]

Swamp???? 
[/quote]

Sorry IntheSwamp   
H2S is produced by bacteria in nutrient-rich standing waters when there is no dissolved oxygen like in swamps and sewers. You can notice it by its rotten egg smell, but you can get used to it if you breathe small quantities all the time. You can test its presence in water using H2S test strips.