Sensor accuracy after restart

My apologies if this is a repeat of other discussions. I appreciate knowing how to get the G6 past ten days, but sometimes it’s worth double-checking with a meter. See photo.


1 Like

I find that odd. Mine is always 100+ points higher than fingerstick after restart and I have to calibrate. I wonder what makes the difference.

1 Like

Mine can be quite a bit off right after restart at first, I try to wait a few hours before I calibrate it. And I try to be around 100 when I restart so if it is off, it doesn’t mess with my reports as being out of range lol…I’m actually ready to pounce with an instant calibration if it’s going to be real wacky.

But for example this last restart my blood test was 86 and the sensor said 129. A couple of hours later it started climbing higher and I blood tested and went ahead and calibrated it. Later on when I checked it again it was only 7 points off. If it is still off a little I might try to fine tune it later. But I’m not unhappy with that reading.

Whereas a new sensor will usually require at least 2-3 calibrations over the first couple of days.

Mine is always between 50-100 higher after restart. I calibrate right away and within 10 minutes its accurate.

1 Like

Are any of you who can get the G6 to restart using a hybrid closed loop pump that relays on the sensor to run the pump?
While the system is working great for me, stockpiling sensors would be great but am not sure if I could trust the sensor to give good numbers to the pump. Any insights?


I’m on t:slim x2 and Control-IQ, and do restarts. I haven’t found it to be an issue. I have been testing a little more since starting Control-IQ, just to make sure. Since I’ve been staying steady at a much lower range now (90-110ish with 24/7 sleep mode and aggressive settings), I worry a lot more and have calibrated more. The discrepancy (except for immediately after restart) is always small, less than the 20% rule… But it might still be a discrepancy I care about. Like, Dex will say I’m 79, but I feel a little fussy, and fingerstick says I’m 68 or something. I’m going to calibrate, because I care about logging those false hypos.

I’m home alone for half the month, since my husband travels for work those two weeks… which is consequently the weeks I post here so much. LOL! He was kinda nervous about me trusting my health to a machine when he’s not here to spot the changes. (He’s way more perceptive of my hypos than I am because of little changes.). I’ve promised to fingerstick twice a day after any restart now to ease those fears. They usually prove pointless, though. I don’t think I see anymore discrepancies than people who don’t restart.

1 Like

That makes me so much better about doing restarts! Do you use just your pump or do you also use your phone? Or do you use the receiver?

My OnePlus 6 isn’t one of the Dexcom supported Android phones, so I use xdrip+ in addition to the pump. Unlike a lot of other people, from what I’ve read, I do all my sensor stops/starts/restarts/calibrations from the pump, though. I like having every interaction logged on an approved device.

Here’s my take on it:

The Dexcom is reporting a glucose value based on a calibration curve for a new sensor. That curve changes pretty rapidly as the sensor equilibrates with the surrounding tissues, then stabilizes with little change for the rest of the session. But that’s with a new sensor. A restarted sensor is already equilibrated so that rapid change isn’t present. That’s why there’s a mismatch. The gap closes after a few hours. That’s based on Dexcom’s patents and mapping I did against finger sticks the first three times I restarted.

For me the difference is always 25-100 points. I recalibrate after two to three hours and it’s smooth sailing from there. I also check a few times the first day just to be sure.

BTW - the same thing happens to me on a presoaked new sensor.

What’s the purpose of waiting 2-3 hours? I don’t wait at all, and have smooth sailing. Have you calibrated right away and had trouble with it?

I get irritated having no data for the 2 hour warmup. Another 2-3 hours of wrong data would drive me crazy

1 Like

Not everyone will experience the same results on either the G5 or G6 when it comes to the first few hours of a new or restarted sensor. That’s just the way it is, Robyn. No sense trying to figure that out from a logic standpoint. :slight_smile:

1 Like

When I ran a test after a 12 hour and 14 hour presoak the offset drop +/- 30 points over 90min. I probably could calibrate at zero and again at 2-3 hours to bring things closer in line during that period, but I’d rather wait. I still see the BG number (which I know is higher) and the trend during that period.