I have a few for outside use, luckily this one was for inside use 
2 Likes
There seems to be some kind of design issue with lht65.
I have seven lht65s. Two of them failed after a couple of years. Somewhat early, but I opened them up to change batteries. Even though they are not exactly easy to open up it is a pretty safe prying and bending operation, as the risk of accidentally damaging the electronics in the process is low.
Both of them had drained the old batteries down to about 0.3 volt. When replacing the battery and connecting the battery PCBA to the main PCBA the supply voltage would drop down to 1.4 volt. Measuring the current with an externally connected power supply indicated that the main PCBA would draw 42 mA. Less than a Tesla, but something is clearly amiss.
And this was true in exactly the same way for both of the failing devices. (Work done in an EPA setting)
Which suggests they were âexhaustedâ from a practical perspective long before attempting the swap out - a common issue when remote/in the field. So this begs the question how have the batteries/chemistry behaves in the the intervening period? Have you pulled the main pcb and inspected for any outgassing/vapour issues where mosture might get deposited causing corrosion and pcb/connections/tracing âcreapâ leading to near shorts or (relatively) low impedance paths to be formed that then get exercised by the new batteries leading to such pull down and current draw? Its good practice to monitor batteries and schedule change what say 80%, ensuring work done by around 90% max 
Maybe - that will depend on update rate and how ADR has adjusted SF in long term use - what does your (meta)data history say about how SF has varied over the life of the deployment? As GWâs come and go over such a long time not uncommon to see significant variance over timeâŚand of course high up link rates combined with high SF mean fast battery exhaustionâŚ
Plus temperature profiles & location - ambient doesnât have to be hot - direct sunlight can warm up a device, then it gets cold overnight and the chemistry doesnât play out as expected.
Thereâs something going wrong. The supply -voltage must not drop to 1.4 V with a new battery. Hopefully you connected the battery the right way.
Maybe itâs a short-circuit or a defective component on the PCB.