[Shannon Hicks]

Thanks for the photo. The screw terminal pic looks fine, on pic of the Mayfly, it’s hard to see the jumper pins for the D6-D7 grove jack, but can you verify that there is a jumper (shunt) on the pins as seen in this photo.

Attachments:

1. 

   20160520_135907.jpg
   

[Shannon Hicks]

Can you post a picture of your sensor wiring showing how it’s connected to the terminal board and also a photo showing how it’s connected to the Mayfly? Did you make any changes at all to the address changing sketch? I’ve personally used almost 500 sensors (CTD, soil moisture, conductivity, etc) from Meter group and have never had a bad one right out of the box, so I have a feeling that the problem is likely in the wiring or the board configuration.

[Shannon Hicks]

Check your wiring, the Meter group CTD sensor has a red wire for the data (signal) wire, white is positive voltage, and black is ground. It’s kind of backwards from most common DC sensor wires where red is usually the voltage supply wire. I’ve confirmed with the folks at Meter that the sensor works just fine at 3.3v.

[Shannon Hicks]

Not sure why it’s not working based on what’s visible in the sketch you posted. But I’m guessing there’s something happening in the TM1637Display library that is looking for return data or confirmation from the display in order for the sketch to continue, so without the display plugged into your Stalker, the sketch hangs and stops, rather than ignoring the lost display and moving on.

I’d recommend something like this display: https://shop.switchdoc.com/products/grove-128×64-i2c-oled-board-for-arduino-and-raspberry-pi

It uses much lower power, you can remove it at any time from the board while the sketch is running and it won’t interrupt things. It has a Grove jack on it, so you can just connect it to one of the Grove ports on your Stalker (or our Mayfly) with a basic Grove cable (they come in various lengths).

[Shannon Hicks]

Matt is correct, the pullup resistors on the I2C bus on the Mayfly board are soldered to the board and can’t be disabled unless you want to unsolder them, which will cause issues if you try to use I2C devices on the Mayfly like the DS3231 clock or the ADS1115 aux A/D.

[Shannon Hicks]

I just read through various manuals and datasheets for the K30 sensor, and it is technically possible to change the I2C address of the sensor, but it’s not easy so I wouldn’t suggest it. But the manual says the board also does serial UART output so you can just follow their example code for Arduino UART to modify it for working with the Mayfly’s correct pins and it should work. You’ll also want to power the board from the switched 5V output pin of the Mayfly since it needs 5v at 300ma to operate, which is a lot. If you run the board continuously, the aux supply regulator on the Mayfly will probably get a little warm and if you’re running this on a battery, it’ll drain pretty quickly. Is this going to be powered continuously by a benchtop supply or a battery?

[Shannon Hicks]

As Sara suggested, the issue is probably related to the SDI-12 address of the sensor. Meter Group ships all of their sensors with a default SDI-12 address of “0” (zero). They do this because it’s the channel their sensors need to be programmed to for working with their own brand of dataloggers. But Meter group sensors don’t respond to SDI-12 commands unless they are programmed to anything other than 0. The DRWI_LTE example sketch you’re using is looking for the sensor on channel 1, so the easiest thing is to use the address changing sketch Sara linked to change the sensor channel to 1. Then reprogram your Mayfly with the logging sketch and it should work.

[Shannon Hicks]

The Hologram network was back up and running early this morning, and there have been no further issues with their network today. All of the stations that lost connectivity on Sunday are functioning normally.

[Shannon Hicks]

T-mobile is the only provider of 2G service that works with the GPRSbee modules that many people are using with the Mayfly, including about 40 stations managed by Stroud. T-Mobile plans to deactivate the network by the end of 2020, so anyone wishing to modernize their Mayfly’s cell technology will need to replace the GPRSbee with a 4G LTE board. We currently use the Digi Xbee LTE boards, which are available for purchase from a variety of online electronics distributors. There are no discounts for purchasing large quantities, so there’s no point in doing a group-buy. Since you paid for your stations and own your old 2G board, it’s up to you what you do with it. But we’ve had many of them fail after a few years of use, so they might be getting towards the end of the lifespan anyway, so I would be hesitant to pass used ones on to someone else.

[Shannon Hicks]

This is an equipment question and not related to MMW, so I’ll answer that in your other forum thread.

[Author]

Posts