Skip to content

Remove get-ns-bg from ns-loop for MDT cgm - #287

Merged
scottleibrand merged 1 commit into
openaps:devfrom
tmecklem:remove-get-ns-bg-with-mdt-cgm
Dec 26, 2016
Merged

Remove get-ns-bg from ns-loop for MDT cgm#287
scottleibrand merged 1 commit into
openaps:devfrom
tmecklem:remove-get-ns-bg-with-mdt-cgm

Conversation

@tmecklem

@tmecklem tmecklem commented Dec 26, 2016

Copy link
Copy Markdown
Contributor

oref0-setup provides a mechanism to fetch NS cgm in addition to device cgm so that the newer of the two can be used in the loop. This change removes that Nightscout entries fetch when using MDT cgm.

Reasons for the change:

  1. NS fetch is unnecessary with MDT cgm. There is a very low likelihood that NS will have newer sgvs than the pump device when using MDT as the cgm source for oref0.
  2. Reliability of NS entries upload. The NS fetch has been observed to cause a problem with the recent-missing-entries report which causes duplicate entries and inconsistency in the NS uploaded entries.

Related PRs:
#288
openaps/decocare#16

It's unlikely that sgv values get to NS from any other process than
openaps, so fetching it doesn't bring any new information down.
Additionally, it seems to interfere with the recent-missing-entries
gap calculation.
@Bender1061

Copy link
Copy Markdown
Contributor

Here is my take on this. if you are also using a mmconnect to upload your bg to nightscout. (i know it's redundant) but what I really like about it is if you BG is over 400, nothing shows up for what your BG is and after 30 mins of being 400, basically the loop stop doing anything. When you have your mmconnect sending BG to nightscout, the BG just stays at 400, and the loop will keep running, thinking you are 400. So it then once it brings your BG back down you are golden. I know this is an extreme situation, but it does work very well for bringing you back down overnight. And brings you back down in a much safer fashion than guessing and giving yourself a giant bolus.
What I would like more, is OpenAPS read over 400 as 400, so it would do the same thing as the mmconnect.

@tmecklem

tmecklem commented Dec 26, 2016

Copy link
Copy Markdown
Contributor Author

@Bender1061 The high and low glucose sensor readings from openaps should be fixed by openaps/decocare#16, commit openaps/decocare@d8946fc.

@Bender1061

Copy link
Copy Markdown
Contributor

Oh Wow, ok, now this I'm going to take a look at! that is awesome. Ok with that change, This is a great idea.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants