[neilh20]

So it worked for me. I took @ensign  basic structure, and added it as an extension in the Class Logger, and then attached it to the two SerialCommands LR and L?

LoggerBase SD Class extension

and

LoggerBase SD methods

 

[neilh20]

Hi I’ve been thinking/working on something slightly similar when @ensign posted above.

BTW I found the github.com/ScottEnsign/SDoutput hasn’t got the actual code, I think it probably needs a “git push” to send the code up.

Seems to me the issue with the @selbig  original post #15641 above, is that the SD card is initialized in Logger, so  ShowData() needs to be an extension to LoggerBase.x : Class Logger {}

From  @selbig post transfer-file-from-sd-card-to-ios-android-via-bluetooth  post I’ve been thinking about adding BT ……

So I believe the Software Use Case  is:

User approaches a Mayfly/BT logger, and with the BT enabled phone, connects to the Mayfly-BT logger, and downloads the readings, also verifies the logger time is accurate.

This could be done with developing an App that possibly uses an embedded remote procedure call (eRpc). However its a lot of skills.

Another version might be ; User approaches a Mayfly/BT logger, and with the BT enabled phone,  connects to the Mayfly-BT logger with a terminal app that uses the BT serial pass through, and then commands the Mayfly to downloads the readings with “LR” (list readings) . Another  separate  command can list the time Mayfly date/time (D?), and also set the Mayfly time with D YYMMDD HHMM.

I have a system that I am thinking of this for,   inexpensive-diy-conductivity-sensor   that currently has  Boot Net (walk up and get data).

Some baby steps in getting there, has been developing a Mayfly SerialCommand interface, and its just about there – see Arduino-SerialCommand  This solves a problem of how to check date in the field.

Initially I developed an interface using a simple String to collect the serial user input data.  However using String led to a terrible bug, as the Mayfly Rx input is floating (Mayfly Issue#28) and so can cause random bursts of characters, which with String meant it periodically fractured the heap and lobotomized the ram. See eight-tips-to-use-the-string-class-efficiently.  The hardware fix for the Mayfly is to add a dongle to J2 “FTDI Programmer” port, with a 1M resistor pull down to the Rx pin3

I’ve just switched to a serial command – and added the context for “LR” and “L?” to  tu_serialCmd.h

The next stages are to get the file listing, and then to try the BT context with the Serial Command being processed from the BT serial interface.

One other issue is that with the UART serial port 115200baud, the serial input is misread about 20% of the time – I’m guessing the baud rate doesn’t quite line up. I wouldn’t think this would happen with BT.

 

 

[neilh20]

@zeke-holloman  I’ve been out the office. Just to summarize our offline investigation – I got a good POST using your thingspeak settings. It seem to me that its your modem data plan that isn’t connecting in some way. Maybe there is a better error indication from DigiXBeeCellularTransparent, but its an evolving area to make it easy.

[neilh20]

@vogelrnws nice to se the layout. That battery holder looks pretty solid and as you said 2 *12V * 9Ah gives 216WHr of capacity. a LiIon at 4.4AH @ 4V is 16Ah.
Its great to have a discussion about the mechanical side, what works, what doesn’t work so well.
Seems like on of the challenges is that uSD is right up against the side of the cabinet – doesn’t appear like it can be easily removed. Would require the Teensy 3.5 to be pulled first.
I wonder also if there is any protection for static/lightening on the instrument leads. That’s a problem in certain parts of the country.

The DIN rails take some thinking about. The issue as I see it is how to do assembly simple and reliably. If the screws are inserted from the front, and require a nut at the back to hold it, it can become a challenge for replacement in the field.

https://www.mcmaster.com/din-rails/ shows some of the options for clips that might solve that.

For the Mayfly, in the Arduino style of a low cost environment, I’ve been using a plastic polycarbonate bar, and configuring it for the mayfly holes.   Here is a write up .. Bud Ptq 11050 Enclosure

 

[neilh20]

I’ve had some people ask me how to get hold of boards.
Just wondering, I know there has been a worldwide shortage of some parts, but is there any update of when the boards might be available? many thanks

[neilh20]

Thanks, good for reference – the critical aspect of any plastic outdoor box is I understand polycarbonate and those IP65 seals 🙂 Keep out the falling rain. Those ants like to get in if they can, so it is a challenge. I really appreciate the detailed guide of what to think about with https://www.envirodiy.org/mayfly-sensor-station-manual/ though I choose my polycarbonate box based on the project.

When building a system, I do try and document it for myself so that I can easily repeat it, and also share it. The Mayfly can fit in a small space ~ NBB-15240 with the uSD facing outwards for easy offloading.
Sometimes though its the instrument wires and dessicant that need routing space, so I’ve also been using the Bud PTQ-11050 and for testing PTQ-11050-C.

Still it would be soo nice to have a processor with at least 4 UARTs, an embedded USB host, and a 1M flash, enough ram and timers for totalizing/counting … with a little screen for site provisioning. I’m playing around with the Wio Terminal – https://wiki.seeedstudio.com/Wio-Terminal-Getting-Started/ but how to get the sleep power usage down, and have it take an Xbee format socket with Digi LTE powering, with all the software open source license…. as ever a variety of engineering challenges. 🙂

[neilh20]

Ahh very nice.  I’m sure its worth it to get a standard design for all climates, and able to cope with keeping some sensors warm.   Seems like the cost is going to go up dramatically though. Powering is a lot of cost and of course IP65 space is costly (for those batteries!!).  I wonder if it is open sourced, or a private contract design.  I use a 4.4AH LiIon, which is quite large for the Mayfly, and of course it can only charge at a max of 0.5A/hr – which means the solar panel is only about 3Watts (though easy to oversize to cope with shading) . Thanks for sharing and the references.

[neilh20]

I wonder if the AWS interface can be done through some form of @vogelrnws  python that runs on the Xbee LTE? JMight be worth looking for something that runs on the raspberryPi in Python for AWS, and then seeing if they could be adapted to Xbee LTE python.  (so easy to throw out an idea – but I have no experience with AWS)

[neilh20]

@vogelrnws thanks.  Would you have an overall schematic of it. Sounds like Solar to +12V.  Then guessing +12V battery that is the main power source. Then  Teensy3.5 supply 3.5to5.5V, possibly also Xbee LTE boards

https://www.pjrc.com/store/teensy35.html

Probably a carrier board for the Teensy3.4 and LTE board?

 

[neilh20]

@vogelrnws  sounds great .. the Teensy 3.5 is a great board and amazing support. If you’ve figured out remote upgrade as well, very noce.  I wonder if you would create a separate thread for it ~ perhaps “using the Teensy 3.5”. One of the biggest issues for me is always how to make a system, low power and solar powered.

[Author]

Posts