For a water/gas pressure monitoring application I would be looking at the following…
10 sec sampling
6 min logging rate
1 Day communications
I know it has been done with a custom designed MSP430 platform and the MangOH Red is advertising 10 years!
Is particles new generation dev boards really going to be this far off?
You can do it, just might need a few batteries to get that long of a lifespan.
Solar could help with daily recharging if you use a long life battery chemistry like LiFePo4.
The MSP430 is super low power.
1 Like
It's a newer generation of processor, but it's not an ultra-low power processor. There are others, even in the STM32 line that are much better at that application.
Or you mean the ones with Nordic semi processors..? Those are probably better but it's not clear if it can get this kind of extreme performance.
I should have mentioned that sorry, would be looking at a 2 cell as a Standard Battery pack and a 4 or 6 cell as an Extended Battery pack.
By recording a Data Header with the current time when logging starts and incrementing the Data Count every time a logging rate is stored, I can calculate the time stamps for each logged value/trend in the top-end, saving data stored and transferred
Do you think Energia would be a better platform for what I am trying to achieve?
It has the same simplicity of Arduino but for MSP430’s…
Would need to figure out how to interface a Ublox or Telit modem with it though.
Nope, all you need is here.
You just need to write your code based on your outline test the power consumption to calculate how much battery you need to reach your run times.
I move in to my new house this weekend, once I get my office sorted etc I can delve in to the coding 
Do you think it would be worth learning how to put a library together first and call parts of code from other sketches etc
My experience is only coding one big sketch and very limited, starting to think this might be beyond me but, I'm not one to give up straight away!
This is only a humble suggestion, but have you considered pushing valuable data to the cloud instead of all the data logging and mass uploads? The key is to determine if the most recent pressure value has changed substantially from the previous value. You can also evaluate the Rate Of Change (slope of the line). I’ve found that when monitoring the pressure of water systems, most of the data is redundant and not actually required. Once a certain number of cycles have passed with no change, you start incrementally increasing the time until the next wake/measurement. You can change your Pressure Differential and Slope Variables with a Particle Function without re-flashing, to fine-tune the program before final deployment in the field.
I use this approach for water and natural gas systems, but not personally w/ battery power only.
This sounds like a good candidate for a FSM to me.
@RWB or several others here can get you on the right track with a Solar Charger. If your Battery Voltage drops to a specified range, your program can always initiate a few levels of power savings.
The code required to evaluate the data and only publish valuable data will be much cleaner and easier to write. But I certainly understand that this deviates from “data logging”, and might not be what you want… just an idea.
3 Likes
Unfortunately because of intermittent comms that won't work