[Shannon Hicks]

We’ve published some sample power usage readings over the years, but it varies by station configuration and board version and a few other settings.  But in general, a Mayfly v1.1 revB board with an EnviroDIY sim7080 LTEbee will use about 8 mA at idle, 0.650 mA in sleep mode, and a max of 110 mA for a few seconds when actively transmitting.  But your most accurate numbers are going to come from doing your own power measurements after you’ve programmed a board to work with whatever hardware you’ve selected.

[Shannon Hicks]

We’ve published some sample power usage readings over the years, but it varies by station configuration and board version and a few other settings.  But in general, a Mayfly v1.1 revB board with an EnviroDIY sim7080 LTEbee will use about 8 mA at idle, 0.650 mA in sleep mode, and a max of 110 mA for a few seconds when actively transmitting.  But your most accurate numbers are going to come from doing your own power measurements after you’ve programmed a board to work with whatever hardware you’ve selected.

[Shannon Hicks]

Your questions are all very subjective to a variety of operating conditions, so I can only give you some broad estimates of what we’ve experienced with the 300+ stations we’ve installed and maintained.

We typically put a 4400mah Lipo battery on the Mayfly boards because on standard logger configuration, they can be charged fully by a 3.5w or 5w panel in just a few hours of direct sun.  It gets more complicated if you’re a shady location or a higher latitude, or by time of year.

The overall power draw of a CTD-only station with LTE cellular module will mostly be the affected by the time it take the cellular module to connect to the network and send the data.  The CTD uses only a few milliamps of power for a couple seconds, and the Mayfly board sleeps when not taking a sample, so if you’re taking readings every 5 minutes, you could operate for several months without a solar panel if you weren’t transmitting via LTE.  At 15 minute intervals, I would guess you could get 3 or 4 months of lifespan, especially if you upgraded to a 6600ma battery.  With the new EnviroDIY LTE cellular module, we typically get at least 3-4 weeks of operation without a solar panel at 5 minute sample rates before the battery starts to get low.  However, cellular signal strength and network availability at the monitoring location determines how long the cell transmitted is on during each transmission.  I’ve seen some stations make their report in 5 seconds or less.  Other will stay on nearly a minute each time before completing the transmission.  (The old Digi brand LTE boards used 4 times more power during operation and took close to a minute for each transmission, so those boards really killed our battery life. We’ve since replaced them all with our own sim7080 EnviroDIY LTEbee which are much more efficient.)

So stations with long transmission times and shady locations will struggle with battery life sometimes, whereas stations in strong signal areas and sun will last almost indefinitely.  We have some stations that were installed in 2016 that are still using the original Lipo battery pack.  The lifespan of the Lipo pack seems to depend more on temperature extremes and moisture.  The blue 4400mah packs we like to use (because they have the compatible JST-connector polarity) have open ends that allow moisture to get into the insides of the pack and cause corrosion/rust on the metal tabs that connect the two cells.  When one of those tabs rusts through, the pack becomes unstable and soon dies.  We’ve had packs last many years in dry conditions, but if a logger enclosure leaks or has high humidity inside that results in frequent condensation on the battery pack (usually at night when temps drop) then those packs will only last a year or two. Keeping a desiccant pack in the logger enclosure and making sure there are no leaks will greatly extend the life of the battery.

The default charge rate of the Mayfly v1.1revB is 500 mA.  The 5w ETFE panel from Voltaic Systems can generate about 900ma in full sun, but unless you change the default charge rate, the Mayfly will be limited to 500ma of input current.  You can adjust this by changing solder jumper SJ15 (and SJ14) as explained on the Mayfly Jumper Settings page, however see the note on that page that says using a charge rate above 500ma will result in very high temperatures of the charging ic chip on the Mayfly, which could damage or shorten its lifespan.

 

[Shannon Hicks]

You’re right, you correctly have your sensor’s signal white wire connected to the Grove cable white wire, which goes to the AA1 pin of the Aux Analog grove jack, so you can disregard my suggestion to change the ADS1115 code to adc0, and just leave those two lines as they were, using adc1.

[Shannon Hicks]

You’d have to check with Meter Group to find out what the sensor communication protocol is.  There’s nothing on their website to indicate what sort of output it has.

[Shannon Hicks]

If you’re able to swap out a Hydros21 manufactured in 2022 with your hardware and everything works fine, but sensors from 2023 aren’t working, then it’s possible that they’ve made another design change that requires some tweaking of the ModularSensors code library to make them work correctly.  We’ve had a report of one other person having trouble communicating with a brand new Hydros21 that was just purchased, so I’ll have to follow up with Meter Group to see if there’s something we should know.

[Shannon Hicks]

First you should attach the sensor’s clear wire to the GND pin of the screw terminal board along with the black wire.

Second, the sensor needs at least 5 volts of excitation power, and the jumper setting of the Aux Analog Grove jacks in your photo is set to the default 3v position.  You need to move that jumper to the 5v position.  See the section on the Grove jack jumper settings in the middle of this page:  https://www.envirodiy.org/mayfly/hardware/jumper-settings/

Third, the Grove jacks only receive power when the switched voltage regulators are turned on, which only happens when you set pin D22 high.  You’ll need to add a line in your sketch to do that.  If you want to take continuous readings, then it’s fine to leave D22 high, but it’ll use more power than if you turn if off when not in use, which is important if you want to make a sleeping logger that powers all the sensors down between readings.  For constant power in your example, insert the following lines in your setup section after line 8:

digitalWrite(22, HIGH);
delay(1000);

The delay line is to give the sensor a second to warm up.  In general, you don’t want to start sampling as soon as a sensor receives power since various sensors need anywhere from 100ms to 45 seconds before a reading can be taken.  Most sensor manuals will tell you what that number is, but for good measure, it’s good to wait at least a second for simple sensors, and adjust if you notice that it needs more or less time.  There will be a little red LED in the lower left corner of the Mayfly that will be on anytime the Grove jacks are powered, indicating that the switched power is on.  If you want to turn them off at any point, simply set pin 22 low.

Fourth, the ADS1115 aux analog input has 4 channels, and they start counting with 0 (zero).  So your available channels are 0,1,2,and 3.  In your photo, you’ve got the sensor output connected to the white wire of the uppermost Grove jack, which if you look at the labels next to the jack, is analog channel 0.  So your code in lines 13-14 should be:

int16_t adc0 = ads.readADC_SingleEnded(0);
float voltage = adc0 * 0.1875;

And as an aside, those sensor outputs can be pretty variable between readings on a partly-cloudy day or in a forest environment.  In your example you’re sampling once every second.  But if you deploy it for real-world monitoring and want to do 1- or 5-minute samples, we found it works well to do a loop of about 50 readings over 5 seconds to get a smoother representation of the actual light levels.

[Shannon Hicks]

Does the sensor have bare wires or a 3.5mm stereo plug?  Either way, ensure that the sensor is connected properly to the Mayfly with the sensor power wire going to a 3.3v or 5v constantly powered pin during the address changing sketch.  And make sure that the data wire of the sensor is connected to the proper Mayfly pin, which is usually D7 in the examples we write.  And of course the sensor ground wire must be connected to a Mayfly GND pin.  If you’re using a Mayfly Grove jack, it’s best to use either of the 2 jacks labeled SDI-12, and make sure the header jumper block next to the Grove jack is set to 3v or 5v.

Also make sure that you’re using the latest ModularSensors library files on your computer when you compile and upload the sketch.  Changes to the Hydros 21 sensor last year require a slight timing adjustment in one of the library files (line 107 in MeterHydros21.h).  You can read more details about it here:  https://www.envirodiy.org/topic/hydros-21-decagon-ctd-reading-error/#post-17066

[Shannon Hicks]

Yes, they were all bought shortly after re-listing.  I’ll be working to build more as soon as I can.

[Shannon Hicks]

If you’re keeping the Switched Power Out LED on all the time, then that is also powering the Aux 3v regulator, the 5v boost regulator, and the 12v boost regulator, as well as the LED.  I haven’t measured the current draw of all of those things while the board is sleeping, but it is most certainly the reason your battery drains so quickly.  You could put an ammeter in series with the battery and get an exact measurement of the sleep and wake currents and then do the math to calculate battery life.

[Author]

Posts