Easier way to restart dexcom wo removing transmitter

A whole thread on the filament being left in. So I always try to remember to watch for it when I remove the sensor!

1 Like

@Jim10 , I have not done a finger stick in over two years. I have had multiple hospital and MD office labs with glucose determinations. My endo & I look at those values and compare them to the CGM reading at the time of the venipuncture. They are consistenly the expect distance apart one would expect for interstitial vs plasma fluids.

I am only ‘dark’ with my CGM 2 hours out of every 240. Your report of multiple calibrations, leaves me to ponder your circumstances leading down that journey.

Here is my collected information about calibrations:

CGM Calibration Facts & Rationale to Avoid Pitfalls

CGM Calibration Facts & rationale to avoid pitfalls & rabbit holes
NOTE: The term “rabbit hole” comes from the 1865 piece “Alice’s Adventures in Wonderland” by Lewis Carrol 1865 and has come to mean a bizarre, confusing, or nonsensical situation or environment, from which it is difficult to extricate oneself.

Test by finger stick (FS) anytime CGM doesn’t seem to match the way you feel.

Learn & verify acceptable deviation according to FDA approvals, ±20% above 100mg/dL and ±20mg/dL below 100. Examples, CGM =200, then FS can be between 160 & 240 and still be fine, or CGM = 70, then FS should be between 50 & 90 with level CGM.

Wait 60 minutes between tests if testing for CGM for validation.

Remember, CGM measurements are interstitial fluid and finger sticks are whole blood and glucose doesn’t move instantly.

Whole blood & interstitial fluid are physiologically different & 15-20 minutes apart in the movement of glucose.

Call tech support if 2 tests are >20% off, 1 hour apart. You may need a new sensor or have other technical issues.

Only enter calibration with great consideration. This reduces the number of times you will get into the calibrate, calibrate, calibrate rabbit hole. There are many who have avoided or not needed calibration for months after understanding the CGM actions.

If you are calling tech support for a goofy sensor, request replacement, regardless.

Follow tech support about changing a sensor.

Avoid the calibration rabbit hole of the system requesting multiple calibrations for hours and sometimes days.

Not everyone (some doctors) who prescribed insulin fully understands CIQ & how it uses CGM. See Section 3:CGM Features to learn more.

Why FSs & CGMs don’t play nice together.

First, FS & CGM are only expected to be within 20% of each other according to published standards accepted in the USA by the FDA. If your CGM says 200 then ±20% is ±40 mg/dL meaning a finger stick is fine if it is between 160 & 240. At the same time if CGM says 100 the ±20% would be 80 to 120. BELOW 80 it is ±20 points (mg/dL) without the percentage according to US FDA standards.

NOTE: Some individuals expect much narrower values between FS & CGM, even to the point of expecting them to match spot on, every time, all the time. This is unrealistic.

Second, as food is eaten, the food enters the blood stream in various places along the digestive tract. Some enters quickly from the mouth, like granular sugar placed under the tongue. Other foods swallowed quickly may not enter the blood stream until arriving in the stomach. The final absorption occurs in the small intestine.

After food is absorbed into the blood stream, it is ‘checked’ by the liver and then dispatched thru the body in the blood stream.

After glucose is in the blood stream, the glucose moves in the liquid part of the blood (sometimes referred to correctly as plasma or serum - like serum or plasma glucose for the doc’s office or hospital blood test) into the interstitial fluid that moves around all body cells outside the blood vessels. It is this interstitial fluid that CGM sensor wires are in to read the glucose and report it to our measuring devices.

The time for all of the glucose movement to take place is usually 15-30 minutes or more, depending on multiple variables including the amount of carbohydrates, proteins, & fats in the consumed food which changes the way the food is processed in the body.

It is this 15–30-minute interval and body fluid differences that give significant part to the variation in CGM and FS readings. This is also the reason most comparisons are done when glucose values are demonstrated level by the CGM before doing comparison finger sticks.

Wish they would play nice together……

1 Like


Thanks for taking the time to comment on my post. Your point about the differences between testing blood and interstitial liquid are good, as is your emhasizing the differences in rates that glucose moves through those two.

When I got my first CGM (Dexcom G5), I learned I should avoid calibrating when my blood glucose level was changing – best only when it was stable and had been for awhile. That is one way to deal with the differences in move rates.

I am one of those a bit uncomfortable with settling for only +/- 20% accuracy. After a new sensor has been installed for 24 hours, I am more used to seeing a variance of 10%. In my learning about the G5 and then G6, my testing consisted of doing 3 meter tests, sometimes more. Typically if there were outlyer results, I’d throw them out and average the others, then take that to compare with the CGM. I would be uncomfortable if my CGM were unable to be in the +/- 10% range from the above meter test averages while fasting and glucose levels are stable over time.

What I have learned about managing diabetes, though, is YMMV – your mileage may vary.

Bottom line for me is I love my G6 CGM. It is a wonderful tool with some small issues that do not bother me much.

1 Like

The key being you are fasting or stable, so 10% would make sense. I do the same.

@Timothy After starting the session with an old, bogus transmitter SN, how long do you typically have to wait before the error?

1 Like

It’s just a few min depending. If it’s real old and not transmitting it will be less than 5 min. If it has some signal maybe 10 min max.
It will never actually pair it. It will give up.

@Jim10, I must be one of the lucky ones. I have not done a finger poke for about 2 years. I have been on various Dexcom models for 20+ years starting with the 7 & 7+.

I have labs every 90 days for a variety of reasons that always include an A1C and a chemistry panel. My physicians are (searching for the best word) ‘entertained’ by the consistency of the Dexcom G6. It is always 13-16 mg/dL higher than serum glucose. The CGM reading is noted by the phlebotomist on the draw slip and entered into the serum glucose remarks for all to see.

BTW, my last A1C was 5.4.

My thinking is you are over checking the CGM. You have validated the technology. What is pondering is your lack of comfort with the variances within the platform. What are you attempting to find with testing?


You are a lucky one. I am someone whose first day is totally unreliable. Usually the sensor alarming that I am Low or in the 60’s when my meter BG is 100-120. Once I get through the first 24 hours of a sensor, I rarely test with my meter. But I calibrate almost every sensor during the first day. Without calibration I don’t think that I could stand the constant alarms.

I periodically insert the new sensor a day before the old sensor expires but it doesn’t seem to make much of a difference. I have used Dexcom since the 7+ and my first day issues have been consistent. Dexcom is cognizant of the fact that some people have first day problems. At the same time, I am lucky that all of my sensors last 10 days while I have D-friends whose G6 sensors fail almost every time at Day 8.

So great that you never have to use a meter. But your experience is not universal.


Ok everyone I have discovered an even EASIER way to restart your sensors. Still not having to take the transmitter out of your sensor.

  1. Wait for your current session to end.
  2. Use your phone or pump to pair your transmitter.
    Just re-pair the same serial number that you already have. You really don’t need to use an old one like I suggested before.
    You don’t need to delete the Bluetooth pairing or anything.

After you do that, the memory is cleared and the system is ready to start the sensor with the code or without. Which ever way you like doing it.

It will go through the 2 hour start up and start working normally.

Crazy simple. I don’t know why this took me so long to figure out.


I am new to Dexcom, having received a sample G6 sensor and transmitter for trial from my endo. I am using it with my iPhone.

After 9 days the readings became erratic, up and down for no apparent reason and sometimes no readings at all over intervals of a few hours, after which they would come back and seem quite steady for a few more hours. On the 10th day the iPhone told me the sensor had expired. Having no other G6 to replace it, I decided to follow Tim’s instructions for re-starting.

I told the Dexcom app on the iPhone that I was putting in a new sensor. It asked for the 4-digit code and I gave it the same four digits that were on the paper tab before the initial insertion. This code was accepted and the sensor went through the 2-hour warm up just as if it was really a new sensor. After the warm up, the sensor behaved erratically just as it had before it expired, but in its saner moments it gives readings I can believe.

I was never asked for the 6-character transmitter serial number and Bluetooth never asked to pair it.

So my re-start was even more “crazy simple” than Tim’s.

Interesting that it allowed you to do that. Did you wait a long time after your session ended.
My guess is if you wait a half hour or more after session ends to restart, it will work.
Sometimes you think it’s restarted but then you end up with a message half way into your warm up.

Still if your sensor was acting wonky on day 9, you might not get good results going forward.

It could just mean you’re dehydrated and there’s not enough interstitial fluid for the sensor to work correctly, but that’s also what they do at the end of the sensor’s life. Not as in the the end of the programmed 10 days the sensor will run, but physically when your body has interacted with the sensor in such a way that it’s usefulness has ended. (Used up all the reagent?)

Dexcom labeled the G6 for 10 days, because the majority of test subjects could get accurate readings for at least that long… But not EVERYONE can. And not all sensors are created equal. Some give up sooner than others.

It’s not typical sensor behavior, but it may be for you. Or it might just be a sub-par sensor and not indicative of what you’ll see if you choose to continue with the G6. There are lots of theories about what affects sensor life, but it’s generally accepted that the better the control (flatter trend lines), the better the sensor life… But there are definitely factors outside our control, too.

The good news is that Dexcom guarantees them to last ten days, and in this case you would call in (or submit an online form) to get a free replacement. At least if it were not a trial. I don’t know if free ones come with a warranty. Wouldn’t hurt to try, though.

Robyn: Thanks for your reply. Actually, I had a problem with this sensor at the start (reading too high) and I called the 24/7 help line. The agent told me to calibrate it with a BG and then it came down right away. She said she would send a replacement sensor to arrive in 4-6 days, but it didn’t appear. After 11 days I called again and a second agent found that the order hasn’t been completed, so he fixed that and said I’ll get it next week.

TIm: I did wait more than an hour before that restart.

I have little doubt that the erratic behavior is due to this particular sensor (or as Robyn suggested, perhaps to me, though I don’t feel dehydrated).

After the restart, the same behavior persisted although I was able to get some readings during the sensor’s normal-looking periods. I got three more days of such data before tonight, when the iPhone app told me in no uncertain terms that this sensor is at the end of its life.

At present I am not depending on the G6 for any treatment, so these erratic end-of-life antics are not a concern. My purpose is to compare the Dexcom G6 with the Medtronic Guardian 3 CGM that I’m using with my 670G pump. I hope to post my observations comparing the two CGM systems on some TuDiabetes forum in a few days.

The reagent in the sensors gets used up. If your sugars stay low
It will last longer. I can get 15 days of reliable data.
But I hate fooling with it so I usually just put in a new one.

I’m going to try just waiting an hour and see if it will restart.

I tried this and used old and new transmitter codes, sensor code and no code but no matter what I did, I got the message “no restarts”. Finally I put on a new sensor.

So what I’ve done in the past is. Change the transmitter number
Start sensor. When u get the error no signal or bad sensor.
Change the transmitter back to the real one and start it again.
But I’m going to try waiting an hour and see this next go


Timothy: For the second time, I have not had your experience of being required to do anything to the transmitter’s Bluetooth settings when re-starting a sensor. In detail, here is what happened.

As reported earlier to this Forum, I called Dexcom’s 24/7 Help Line because the G6 sensor I had started using on trial was reading 25% too high compared to BG measurements. The agent told me how to calibrate the G6 and said she would send me a replacement sensor. That was on 8/31/21. The sensor gave readings for the next 10 days. When it expired on the 10th day, it asked for a new sensor. The promised replacement had hot yet arrived, so after a few hours I re-started the original sensor by telling the iPhone app that I had installed a new one, giving the original 4-digit code. The app did not ask for the transmitter’s 6-character serial number; evidently it was still paired. The app showed (sometimes erratic) G6 readings for several more days until on 9/12 the app told me the sensor had failed and I must install a new sensor. So I removed the original sensor and transmitter.

After much delay and another phone call to the Help Line, the replacement sensor finally arrived on 9/16. On 9/17 I installed the second G6 with the same transmitter. Again, the iPhone app asked for the new sensor’s 4-digit code, but didn’t mention anything about the transmitter. After the 2-hour warm-up period, sensor readings were again showing on the iPhone.

@emelvy When you install a new sensor you don’t have to enter a new transmitter code until the transmitter runs it’s 110 day life. The new transmitter code was only to restart an already used sensor for a restart in @Timothy 's method. The new transmitters don’t usually let you just wait and restart without either removing them or Timothy’s method. We used to be able to just wait and restart but then Dexcom changed the transmitters to make it more difficult. It is interesting that maybe there are still some transmitters out there that might work with the old way. Once a sensor is acting funky in readings it doesn’t usually work to restart it.as It stays with it’s funky readings.

@MBW I always restart my sensors, but Timothy’s method didn’t work for me. I tried it twice and just reverted to the older remove the transmitter way. I figured I was doing something wrong and will try again.

But this previous method always works if you want to restart them @MBW
Here’s the link with video.

Marie 20: I was given the first G6 sensor and the transmitter by my endo so I could run a trial (to compare the G6 with my present Medtronic system). Perhaps the transmitter they gave me was an old one, not having Dexcom’s new programming.

1 Like