-
Cal posted an update 5 years, 6 months ago
Has anybody seen a voltage spike recorded by the Mayfly? I have 2 Mayflies measuring water level each powered simply by a 6600mah LIPO battery – no solar panel and no other sources of power. I’m using the sleeping option so that each Mayfly wakes up every half hour and records sensor values plus the battery voltage. The Mayflys have been deployed for about 2 weeks. For several days the voltage on one of the Mayflies was just over 4v. Then it jumped up to 5.8v and slowly decayed back to 4v over 7 hours. Is this a battery glitch? or a Mayfly measuring glitch? Software? Other?
Welcome to EnviroDIY, a community for do-it-yourself environmental science and monitoring. EnviroDIY is part of WikiWatershed, an initiative of Stroud Water Research Center designed to help people advance knowledge and stewardship of fresh water. New to EnviroDIY? Start here
Activity
What version of Mayfly do you have? v0.3, v0.4, v0.5, or v0.5b?
All my Mayflies are v0.5b.
I haven’t seen an issue like that. The battery voltage is measured by a simple resistor divider across the battery input, and measured by an analog pin on the micro, so it’s unlikely to randomly drift, unless there is something wrong with the analog reference pin (AREF) or your battery. If it happens again, maybe check the lipo battery voltage occasionally without disturbing the rest of the setup. You can just put your voltmeter leads on the pins of one of the battery JST sockets. That way you can see if the actual battery voltage is wandering, or if it’s a measurement issue with the Mayfly.
OK. If/when I see this happen again, I’ll check the battery with my voltmeter.
Unfortunately I am seeing some issues with the LiIon battery measurement. I think the issue is that the battery measurement uses a 10M/2.7M resistor with a high impedance Analog input, and if its in an RF noisy location (roving cell phones) it can detect some of that energy. Wires can behave in two ways – on the one hand they conduct current, and the other hand they can act as antennas – the difference is on the terminating impedance/resistance.
I have had a Mayfly running next to a stream in an urban location, and it hasn’t shown any +Ve spikes. On my test bench – a noisy RF location I saw lots of funny spikes on V_Batt measurement that I wasn’t able to understand.
I have modified one of my Mayfly 0.5b for R1/R2 to be 1M/270K, and also connected Vbatt to ADS1115/ExtV and had it running, but I haven’t had time to analyze it in detail yet.
Neil – I saw the spike on my Mayfly for the 2nd time last week. I think I have a different problem than you – and I think my problem is battery related. The spike goes from 4v to 5v then tapers back down over several hours. I can’t imagine any stray electromagnetic field causing that kind of jump for that kind of time span in the middle of nowhere northern Michigan. I have 4 of these batteries and only 1 is exhibiting this kind of behavior so far. When the snow clears in June I’m adding solar panels and LTE-M modems to my stations. I expect to see a much different voltage profile then with the XBee drawing amps of power and the solar panel catching brief periods of sun.
Cal – Some thoughts, if you are using the Mayfly’s battery measurement its not actually measuring the LiIon battery voltage. How are you connecting the solar cell and what rating is it?
If the battery is charging, its taking current and the voltage indication is close to that of the battery – depending on how much solar there is. For the voltage indication to go to 5V it could be because the battery charging is not taking as much current as can be supplied by the solar panel.
So what rating is your solar panel?, what connection are you using (ext or USB input) and does the measured voltage go to +5V when the sun could be intense or likely for the the battery to be charged.?
Neil – there is no solar panel on this setup. We swap the battery with a new charged one after 3 months use.