Can ask to go back to Libre 1 as Libre 2 just not stable enough for me

[sololite]

@sololite I posted this earlier incorrectly to nonethewiser, where I meant it to be seen by you.
Notwithstanding that Abbott got you to clear the cache, was that for LibreLink or your phone in entirety? Bits of erroneous data get trapped in 'corners' and if that is causing your difficulty you may find it worth following the steps above, ideally shortly before activating your next sensor, thus start on a clean page 'so to speak'.

Thanks for these tips. I will do this before I risk starting a new sensor via the phone

 

[nonethewiser]

@sololite I posted this earlier incorrectly to nonethewiser, where I meant it to be seen by you.
Notwithstanding that Abbott got you to clear the cache, was that for LibreLink or your phone in entirety? Bits of erroneous data get trapped in 'corners' and if that is causing your difficulty you may find it worth following the steps above, ideally shortly before activating your next sensor, thus start on a clean page 'so to speak'.

No problem, handy to know anyway.

 

[nonethewiser]

Thanks . Abbott got me to clear the cache and switch it on and off but by then the sensor was toast

Find uninstalling reinstalling apps works more often than not when having issues with any app, bit like restarting phone does.

 

[Proud to be erratic]

Find uninstalling reinstalling apps works more often than not when having issues with any app, bit like restarting phone does.

BUT, be aware as I previously said, if you uninstall you lose the existing data; which might or might not be acceptable.

Also, if there is a gremlin lurking anywhere in the phone's many cache stores then (in harmony with the D fairy) there is a good chance LibreLink or Bluetooth will find it, attract it and then use it to disrupt. Don't underestimate the complexity of modern phone's and the inter-connectivity of the embedded codes in the phone as well as hidden apps. Example: in an email many months ago, I mistyped bosul for bolus; despite various attempts to erase that embedded mistyype, in LibreLink notes (and on this site) that typo still recurs. Its infuriating, more so because I know it exists and can't find the cache that it's coming from.

That said, I agree that uninstalling and reinstalling is a good way to resolve problems. But you still have to clear the cache inbetween; it's the price of progress with smart phones and built in bloatware! There may be apps that properly clear your phone, particularly android phones.

 

[Proud to be erratic]

Libre 1‘s algorithm also attempted to close that interstitial fluid gap (I think they were aiming for 5 minutes), and has the same potential to show spot readings outside of the trace.

Could I clarify, for my interest only since I'm a L2 guy no matter what, are you confirming that L1 has altered their original algorithm in order to close that interstitial gap?

Libre 2 is a new/updated algorithm though. Which I think was intended to primarily improve accuracy.

It’s a while since I have been invited to meet with the folks at Abbott though.

Just wondering if, from when you last met the Abbott folks, do you have more informative tech info from Abbott about upgrades etc? Most of us are probably speculating, from 'tit-bits' we've gleaned somewhere and the somewhat superficial promotional material that Abbott make easily available.

 

[everydayupsanddowns]

Could I clarify, for my interest only since I'm a L2 guy no matter what, are you confirming that L1 has altered their original algorithm in order to close that interstitial gap?

Sorry for the confusion! No, what I meant to say was that the ‘gap closing’ was always part of the Libre algorithm from the very first meeting I attended in 2014, and wasn’t something that was newly added with Libre 2.

I am only aware of one Libre 1 algorithm change - when they sent out replacement readers from the earliest versions (and also upgraded the app). The aim was increased accuracy overall, I think though, rather than further gap-closing?

Just wondering if, from when you last met the Abbott folks, do you have more informative tech info from Abbott about upgrades etc? Most of us are probably speculating, from 'tit-bits' we've gleaned somewhere and the somewhat superficial promotional material that Abbott make easily available.

Ah well the pandemic has scuppered a lot of those meetings, and it’s now a few years since i have been able to attend anything with them in person. I do get occasional updates via email.

In January I learned that Libre 3 was to be made available as a realtime CGM as part of the new Insulin Pumps, Continuous Glucose Monitoring, Closed Loop Insulin Delivery Systems and Associated Products framework. Which as I understand it, means that it will not be a like-for-like replacement / upgrade option for Libre 2 (ie it won’t be a prescription item), but will follow a similar set-up to the provisioning and assessment criteria for insulin pumps. And that it was due to be available in the UK ‘soon’ (it has been available in Germany for a while now).

Though the NICE guidance update (which is due for publication following consultation) did suggest that access to CGM for T1 was going to become a recommendation - if the draft guidance remains unchanged.

Not sure of that helps?

 

[Proud to be erratic]

Sorry for the confusion! No, what I meant to say was that the ‘gap closing’ was always part of the Libre algorithm from the very first meeting I attended in 2014, and wasn’t something that was newly added with Libre 2.

I am only aware of one Libre 1 algorithm change - when they sent out replacement readers from the earliest versions (and also upgraded the app). The aim was increased accuracy overall, I think though, rather than further gap-closing?


Ah well the pandemic has scuppered a lot of those meetings, and it’s now a few years since i have been able to attend anything with them in person. I do get occasional updates via email.

In January I learned that Libre 3 was to be made available as a realtime CGM as part of the new Insulin Pumps, Continuous Glucose Monitoring, Closed Loop Insulin Delivery Systems and Associated Products framework. Which as I understand it, means that it will not be a like-for-like replacement / upgrade option for Libre 2 (ie it won’t be a prescription item), but will follow a similar set-up to the provisioning and assessment criteria for insulin pumps. And that it was due to be available in the UK ‘soon’ (it has been available in Germany for a while now).

Though the NICE guidance update (which is due for publication following consultation) did suggest that access to CGM for T1 was going to become a recommendation - if the draft guidance remains unchanged.

Not sure of that helps?

Thank you. It helps in that it clarifies some aspects and confirms that we (me included) are to some extent speculating.

One of the things I find interesting is that my android app, Diabox, frequently continues to take readings from the L2 sensor and provide readings to my phone, even when the sensor is giving the Error 373 message 'try again in 10 mins'. Such readings look realistic and on the few times I've finger pricked the Diabox readings have been consistent with actual BG.

This doesn't go on regardless; if the sensor stalls for a longer time, say 30mins plus, then Diabox eventually declares sensor error and sometimes picks up where it left off when L2 resumes normal service and sometimes needs its own restore process. But looking back at the graphs Diabox is generally complete where L2 sometimes isn't.

I guess, my speculation, that the L2 flash scan fails at one given moment but the sensor is working to some extent and its signals every minute are being collected continuously by Diabox - whose algorithm seems less fazed by the changes that cause L2 to stall. Whenever Diabox finally succumbs and stalls because of L2 sensor error, I find that its calibration to actual BG also flounders and recalibration can only be done when everything is back to steady state; for my brittle DM that is sometimes challenging in itself. Initial calibration to actual BG has to be done twice, with a 2 hr gap. Recalibration sometimes restores after one further attempt and sometimes needs the second recal. I can't find a logic or pattern to when this is going to apply: so far!!

I also found that other apps using Bluetooth, as well as some notifications, would sometimes interfere with my continous BG collection. So I now have a 2nd phone for daily smart phone activity, with my older smart phone stripped down of many apps and I've heavily restricted what I do with this dedicated Diabox phone. I appreciate this won't be ideal for many people; but since I made this change before Xmas I've had very, very little instability with either LibreLink or Diabox. I keep LibreLink disconnected from Bluetooth, depending on Diabox to read the sensor and transmit the results, including alarms.

Sorry, I've digressed a bit.

 

[everydayupsanddowns]

One of the things I find interesting is that my android app, Diabox, frequently continues to take readings from the L2 sensor and provide readings to my phone, even when the sensor is giving the Error 373 message 'try again in 10 mins'. Such readings look realistic and on the few times I've finger pricked the Diabox readings have been consistent with actual BG.

My understanding of the ‘try again in 10 minutes’ message is that it’s a sensor signal self-check safeguard.

Obviously Abbott have to put precautions in place for and sensor that is beginning to come unseated, or did not insert properly, and may be giving an erratic feed of data because it is not reading interstitial fluid correctly, the filament isn’t at the right depth, or the sensor itself is just faulty. So the software has to do a sort of ‘does this seem about right?’ check on the feed of data coming in and can’t just turn any old garbage into an apparent glucose value.

If the software sees unexpected levels of variation in the raw data, sometimes it decides to take a break to see if the data begin to settle down to more expected values after a little while.

Of course, this can mean that if the sensor is reading fine, but that glucose values are changing rapidly and erratically, the ’timeout’ will get triggered then too.

 

[helli]

I have experienced a situation with xDrip where it is receiving data after LibreLink declared the sensor to be invalid. XDrip plotted what it received which was a graph bouncing between HI and LO on every test. I guess xDrip does not do any validity checks on the data it receives.