Bonehead Move

Overnight my G6 sensor was on it’s last legs. False lows and highs along with sensor errors kept waking me up. I had a sensor presoaking for the expiration of my current session this afternoon. Fed up with all the alarms disturbing my sleep I decided to kill it and start the new sensor at around 4am. Groggy from the early wake up, I walked to the bathroom and pulled the sensor to move the transmitter. But oops, I pulled the new sensor!

I put in another new sensor to presoak and restarted the wonky sensor in hopes I could limp through the day. No such luck. The numbers are useless. I’m spending the rest of the day sans CGM. An experience harkening back to the days before CGM’s with just a few finger stick readings per day. It drives me crazy now not having a continuous stream of readings. It didn’t bother me back then. I guess ignorance was bliss.

Moral of the story “Coffee first, sensor start second”


I had a transmitter fail unexpectedly recently, without a backup. Thankfully Tandem was able to expedite faster than the originally quoted 8-10 days, but I was still in the dark without data for 4 days. TORTURE!

1 Like

It shows how reliant we have become on always knowing what our CGM readings are. I even hate the 2-hour restart.


When I was on the old mini med enlite sensors just a month ago, I could trick my pump into skipping the warm up. If you told it you were continuing with an old sensor instead of starting a new one it would give numbers. I also figured out how to calculate the isig into a glucose number.
But I’m happy with dexcom even with the 2 hour because I can put it on and go to bed. Then it warms up and starts collecting data.
My minimed required me to get up and calibrate it and I hated that.

Are new sensors so bad for you that it’s worth restarting a wonky old one just so you can pre-soak, then go through warmup all over again the next day?

When I had a transmitter that wasn’t sitting in the sensor right, I totally went nuts knowing that even though they supposedly were going to overnight or 2 day ship. I knew I wouldn’t get it for a few days at least. I managed to get the transmitter to work. I realized that a weak spot in my back up supply was no back up transmitter. So I went to Costco and bought one that I cycle through to make sure it doesn’t get old. I managed to tape in that transmitter, which was good because it took a week to get me the replacement.

I also hated the 2 hour warm up and probably did more BG tests in that 2 hours of start up period than for the whole 10 days I would wear the sensor.

But one time when I had to switch out a sensor early in the day. I knew I wouldn’t want to change the next one in the morning so I decided to wait until the evening. And it actually felt good having a “day” off of CGM. It sort of liberated me. I don’t presoak because I never know when my sensor will stop working as I restart them.
But now I take my time switching it out sometimes, I have gone whole nights lol without one started or I have turned it off a few nights because it annoyed me as I was floating near my alert number.

And it’s really not so bad… As long as it’s a limited time lol…

For a while, I could count on the older sensor having loose adhesive and being easy to pull off (compared to a new one).

Not so much anymore. Past 8 months the sensors all have great adhesion after being on.

1 Like

Absolutely. The first 24 hours with a new sensor without presoak is shear torture. The wonky sensor might give numbers that are way off, but at least the trends are OK (except when they go into the dreaded “sensor error”). You have to pick your poison.


Why “Pres oak”

With G6 I find the readings good right after the warmup

Only need to go two hours blind

Count yourself lucky, @Dick_engel! It’s almost always risky to generalize from one’s singular experience. I’m on the first day of a G6 sensor without presoaking. Here’s the rough ride I’m on today.


11:15 105–105
12:11 83–103
12:38 46–108
12:47 83–106
1:10 122–114
1:47 54–117
2:18. 44–121
2:38 50–147
3:24. 48–145
3:31. 56–147
4:04 92–118
4:22. 83–114
4:52. 61–107
5:18. 63–104

Since I use the output of the G6 to inform Loop for every-five-minute insulin dosing, I’ve had to open the loop and run in manual mode. As you can see, my blood sugar as indicated by my fingerstick meter has been relatively stable (103-147) yet my first day sensor has been wildly off the mark (44-122) and inconsistent. My only calibrations have been the two required ones at the end of the warm-up period since I chose to use a “no-code” start-up.

I’m not new to using a CGM but have just started using the G6. You’ll find many accounts like mine if you do much reading through this site. I’ll likely experiment with pre-soaking. Dealing with the false low alarms tries my patience!


Why use the no code?

I use the code and get reliable results regularly

Before G6 my first day readings were always goofy. But the G6 with code and no calibration had been awesome for me

I’m just a few months in using the G6. I’m still experimenting to find a stable data stream. I’ve now experienced several G6 sensor starts using the code with poor results.

So this is my experiment with starting with no code. After the data posted above, my finger-sticks and G6 data started to converge last night. The first meter reading this morning was within 3 mg/dL of the G6. This argues for a 12-hour pre-soak.

There are many variables that we cannot control when we compare our experience with others. They include difference in body chemistry and the possible quality variations in the sensor lots we use as well as the various transmitter serial numbers.

I’ve been using Dexcom CGMs continuously since the 7+ in 2009. My best and most consistent data came from the G4 system. If it was possible, I’d still be using the G4. I didn’t pre-soak the G4 but resisting over-calibrating and didn’t depend too much on the first-day data.

I’m glad that your experience has been good. Not all of us have been lucky that way. Hopefully, I’ll find a method that makes the G6 work for me. It is essential for my glucose control. In the meantime, I will continue to experiment.

1 Like


Wishing you luck

It is interesting that our experiences are different

When I first got the G 6 I did not trust the no calibration and ran hundreds of finger stick comparisons

Within a couple weeks I was a convert and find the best results I ever had

Like you I have been using since 2009

I have it down with my g6, it’s only been 6 weeks, but I’ve finally do what works for me. I insert a new sensor on my tricep. I let it soak over night. I keep my other arm with the sensor so I have data at night.

In the morning I remove my old sensor and install the transmitter in the new one. I take a hot shower. To remove the old adhesive but I think it has beneficial effect on my new sensor too.
When the 2 hours are up I generally have accurate results from the start.
I’ve tried this with no shower and I get wonky results for several hours before it starts working right.
My last 3 sensors I have not needed to calibrate at all with this method.
So I only use the backs of my arms because that’s the only place they will stay in for 10 days, and I don’t bump them.

I tried abdomen and it came out while doing sit-ups once. Another time was sheared off going passed a doorway.
Everyone has a little different body chemistry and the strength of your lymphatic system has a big effect.

I think thats why arms work so well, gravity and moving your arms moves the interstitial fluid faster. That’s just a theory however.
I had good results with enlite in the same area.

1 Like

I want to add an update to my G6 experience. After much trouble getting my G6 sensors to be accurate and consistent, I started my current sensor without a code knowing I would have to calibrate daily. For this sensor, at least after the initial 12 hours, I’m liking what I see. Here’s my last 24-hour trace.

Yesterday’s calibration fingerstick was 97 while the G6 read 90. Today’s calibration was even closer: fingerstick @ 100 as the sensor was showing 101. Last night was the first night in many that the G6 didn’t screech, not even once.

I like the idea of using the code and not requiring any finger sticks so I’ll try that again. I don’t really mind fingersticking especially if it reverts to an accurate track like this one has. Perhaps a 12-hour pre-soak is the answer for me. We’ll see!

1 Like