Hello guys,
I run a company which main activity is to develop Decision Support System (SaaS) based on weather data.
We want to develop our own datalogging unit.
I need to evaluate if Arduino eco-system could be a way to log sensors and transmit that onto our server via gprs.
Who could help us in such evaluation, is there some volunteers or consultants we could discuss with ?
Thanks for any suggestion.
Edouard
I need to evaluate if Arduino eco-system could be a way to log sensors and transmit that onto our server via gprs.
Arduino eco-system? Please explain what you are talking about.
If by "eco system" you mean the hardware and software environment/tools, then yes I'd say it's suitable. Definitely for a prototype and even a few production versions.
If this is going to be a product then it's an easy matter to build your own "Arduino" board with just the components you need.
What are you logging, how often, how long?
Who could help us in such evaluation, is there some volunteers or consultants we could discuss with ?
We're all volunteers that will help. Some here will also contract to do this sort if thing.
Rob
Hello guys,
'eco-system' means (for me) all this together : the hardware from the core project, and all the forks / shields that can extend it; all the software and services that 'turns' around the project. Like every community this is made of volunteers, but couldn't live if no private companies could benefit of it (and sponsor its further development). Anyhow this is just a question of words, don't focus on it. My main question is 'can I answer my needs with arduino ?'
I guess that volunteers can help me to make my idea on this question, and won't be offended if any company see an opportunity to collaborate on this (and therefore contact me directly).
Basically, the main idea of our project is the have a unit that log sensors, transmit data onto our server, can upgrade its configuration remotely (including firmware) and interact with its environment (relays). Concerning the figures we actually sell about 200 weather stations a year and could expect much more with a product cost-effective and more adapted to the market.
Sensors are high quality sensors from the market : hygroclip from rotronic for the temp and RH, kipp&zonen for radiation, young for wind, watermarks from irrometer for soil humidity, decagon sensors,...
Transmission, we want to give the choice to our client : gprs (if no LAN), ethernet, wifi otherwise. Is it possible to offer it as an option and pick the appropriate shield ? Can the firmware manage an automated recognition of the shield plugged and us it to transmit data onto our server ?
Remote sensors : we also thing about sensors in the field (like watermark humidity probes) but cable are not welcomed in such environment. Can we use Xbee to communicate between main unit ant such sensors. Do we need an arduino board per sensor or does the Xbee have enough 'intelligence' to deal directly with the main unit ?
We also care about memory issues and data losses (if the network fails for a while). Any comment on this ? do we need an extra shield ?
We also care about electric consumption (is the arduino efficient on this point ?) has anybody realized projects far from it computer, with battery (and solar panels) ?
This is my main preoccupation at the moment.
Any comment / proposal welcomed.
Best regards,
Edouard
@Promete:
Where are you located ?
the main idea of our project is the have a unit that log sensors, transmit data onto our server
Can be done with Arduino,
can upgrade its configuration remotely (including firmware)
upgrade a config, can be done over ethernet; sketch can check new config once a day or so, fetch it and store it in its EEPROM or on SDcard
upgrade firmware over internet, cannot be done => X
Sensors are high quality sensors from the market : hygroclip from rotronic for the temp and RH, kipp&zonen for radiation, young for wind, watermarks from irrometer for soil humidity, decagon sensors,...
Please provide as much details as you can about the sensors, that is the only way to check feasability # IO pins needed
TYPE, NUMBER, WORKING RANGE, ACCURACY, FREQUENCY, LOCAL/ REMOTE, DATASHEET
e.g.
DS18B20, 2, -10 .. 50 C, 0.5C, once per minute, remote, www.somewher.com/ ...
Transmission, we want to give the choice to our client : gprs (if no LAN), ethernet, wifi otherwise. Is it possible to offer it as an option and pick the appropriate shield ? Can the firmware manage an automated recognition of the shield plugged and us it to transmit data onto our server ?
ethernet and wifi may be interchangeable. Supporting autorecognition of shields takes quite some resources (you need "drivers/libs" for both) which restricts room for sensors.
Remote sensors : we also thing about sensors in the field (like watermark humidity probes) but cable are not welcomed in such environment. Can we use Xbee to communicate between main unit ant such sensors. Do we need an arduino board per sensor or does the Xbee have enough 'intelligence' to deal directly with the main unit ?
Arduino can support XBEE, if XBEE can talk directly to the sensor depends on the sensor
We also care about memory issues and data losses (if the network fails for a while). Any comment on this ? do we need an extra shield ?
Build a 2 stage system,
read sensors and store values to SDcard
once per hour/day upload to server,
if failed try again in an hour
an sdCard can hold 2GB of data (or more) which is enough for quite some samples (depending on #sensores, frequency, dataformat, ...)
We also care about electric consumption (is the arduino efficient on this point ?) has anybody realized projects far from it computer, with battery (and solar panels) ?
The Arduino itself doesn't use very much, but with every sensor it adds up. The frequency of sampling makes sleep modes possible or not. It is hard to say without the compete sensor list
If you check the forum (and the rest of internet) you see that all kind of remote projects have been done. Most use (CAR) batteries, sometimes in combination with solar or wind.
All ideas you have are possible - except the remote firmware upgrade - with standard Arduino, if all options can be combined and still work is a question that needs far more information before it can be answered.
Rob
Dear Rob,
Promete is located in Narbonne, south of France.
Please provide as much details as you can about the sensors
I am currently working on the 'functionnal specs listing', I will be able to give you a recapitulating table soon. But for some sensors, no definitive choice has been done and we remain open to anything that will 'do the job'.
TYPE, NUMBER, WORKING RANGE, ACCURACY, FREQUENCY, LOCAL/ REMOTE, DATASHEET
I am not sure to understand what you mean by frequency : are you talking about the signal itself or are you talking about sampling / recording frequency ?
Concerning the different transmission way, I think this would still be very interesting on the commercial point of view and memory doesn't cost that much today. Anyhow, it seems that it is better to view large on this point.
Build a 2 stage system,
Do you mean : creating our own design based on arduino ?
Thanks for your support,
Edouard
Me, Netherlands (near Eindhoven)
I am currently working on the 'functionnal specs listing', I will be able to give you a recapitulating table soon. But for some sensors, no definitive choice has been done and we remain open to anything that will 'do the job'.
At least provide type of measurement e.g. Temperature, range, precision, the exact model of sensor can be determined later..
I am not sure to understand what you mean by frequency : are you talking about the signal itself or are you talking about sampling / recording frequency ?
Primary recording frequency. NB note that some sensors are read multiple times and log the average to gain better precision.
Build a 2 stage system,
Do you mean : creating our own design based on arduino ?
No no I mean, the system could log the data on a local disk/memory and once per hour or once per day (config) this data is uploaded to the server. This makes the system more robust in case the server can't be reached. In short the 2 stages are 1) local logging and 2) uploading.
Dear Rob, Thanks for the last replies.
I keep you update as soon as I have finalized the specs.
Edouard