Judging by the Silver Oxide battery depletion curve, when you hit a certain point, it will fail rapidly assuming that the 1.5 volts is needed to run the system.
Just a guess, but assuming worst case storage conditions and picking a point 75-80% of the way to the drop-off is probably where they came up with the time to replace, so it makes sense you can get extra time, as long as you are fine with understanding what happens at the cliff.
@Chris I really do not understand electrical engineering very well, but the explanation I got from the xDrip+ developer said that the key indicator of the battery life is the resistance.
I do believe that when it reaches a specific point the app warns that the transmitter is about to fail. The voltages seem to be staying fairly constant, but the resistance is what is changing.
True, the internal resistance is a good measure of where the device is on the depletion curve. The advantage of those silver oxide batteries (if you look at the curve above) is that they give 1.5 volts or more through 95+% of their life, so the circuit gets a very consistent voltage, which is nice. The issue with this is a voltage measurement isn’t helpful to see where it is on the curve, since it is almost always above 1.5 volts and doesn’t decrease at all until the very end. So internal resistance can be measured and will be more predictive.
So as long as we had the Dexcom G5 Mobile app installed on the iPhone, we could still get readings from an old transmitter? Even though the receiver would not accept them? Or is Nightscout/XDrip required to use this functionality?
@ClaudnDaye Unfortunately, the only way you can extend G5 transmitter use is by using xDrip. The Dexcom iPhone app is a virtual software duplicate of the receiver, including the code to stop accepting signals from a transmitter over 105 days old.
k. This is incentive for me to give xDrip another look.
Weekly transmitter check. Now at 137 days, voltage is still great ( but voltage really only drastically decreases when the batteries die, as @Chris pointed out in his graph above).
Resistance is a bit up at 614, but it’s been fluctuating between 590 and 630 all week.
I’m up to 32 days past labeled expiration, fully one month. A G5 transmitter pair costs about $600, about $100/month. So I guess so far I’ve gotten about $100 extra value out of this sensor.
Hopefully next week it will be another good report.
That is pretty amazing. Looks like you will get +6 weeks out of it. Which is a bunch considering they are meant to last 14.
Time for the weekly transmitter update.
All is good at 144 days, 39 days past the105 day mark. Resistance is actually lower, but again, it has been fluctuating between 550 and 625 this week.
Voltages are still very healthy.
We should do an over/under betting pool on how long this thing will last
162 days.
The batteries speak to me.
Ahh, a battery whisperer.
173
What do we win if we guess right?
The prize is a dead Dexcom G5 transmitter 
Dude. I want a hand crafted battery replaced G5 transmitter.
Alternatively, I would settle for a video showing the attempt.
(Actually would find the video SUBSTANTIALLY more amusing…)
That is EXACTLY what I was asking for.
I don’t want Bob the Builder from YouTube. I want @docslotnick with a Dremel !!!
No Dremel needed. Just a box, a check for $79, and a trip to the post office.
I’m really quite handy in that way.
Sure - take all the fun out of it !!!
