These are chat archives for nightscout/intend-to-bolus

28th
Feb 2019
djnoor
@djnoor
Feb 28 15:31
Are there any advantages to looping with a MiniMed 523 vs. a 522?
Eric
@ecc1
Feb 28 15:39
Finer resolution delivery of insulin, probably only matters for very insulin-sensitive people (children, often)
Andy Sharrow
@dramageek
Feb 28 15:40
@djnoor Technically, smaller dosing. Practically, OpenAPS is going to be tweaking things anyway and will balance it out without the smaller increments.
Teamhrach
@Teamhrach
Feb 28 18:43
@danamlewis PR uploaded
Dana Lewis
@danamlewis
Feb 28 18:45
@Teamhrach I don't see it in the PR list (https://github.com/openaps/docs/pulls) yet - might need to hit the final green button of 'create PR'? or send me a screenshot of what you see and I can tell you what to do next? (general steps are edit and save to your own branch; then create PR to OpenAPS branch; I get notifications when they come in but don't see this PR currently)
I see it now! great, thanks! @Teamhrach
Teamhrach
@Teamhrach
Feb 28 18:50
Great - even though it's only a few sentences, it's takes quite a while to get these few words just right. Feel free to change it further. I can only imagine the 1000's and 1000's of hours of writing / editing to get the docs up to this standard!
Dana Lewis
@danamlewis
Feb 28 18:50
:+1: exactly. and it takes a lot of helping hands, so thank you for inputting this content!
Teamhrach
@Teamhrach
Feb 28 18:50
@danamlewis Great - even though it's only a few sentences, it's takes quite a while to get these few words just right. Feel free to change it further. I can only imagine the 1000's and 1000's of hours of writing / editing to get the docs up to this standard!
Glad to make this tiny contribution. I'm loving what the OpenAPS is doing for me so far
Dana Lewis
@danamlewis
Feb 28 18:51
<3 :)
Teamhrach
@Teamhrach
Feb 28 18:52
I hear from Sh and Gr that you may be visiting us in Ireland in May? Hope to meet you then.
Dana Lewis
@danamlewis
Feb 28 18:52
yes!! :) :) I can't wait.
bjarecki
@bjarecki
Feb 28 19:06
Does anyone else feel like they have nothing but problems with the explorer hat? Screen aside, which breaks if you look at the wrong way, and battery life as side. I have had 4 now: one just started making a buzzing and stopped working. On stopped charging batteries and two will no longer run off battery (the rig only worked if plugged in). I know flashing the edison could be challenging but once the rig was up it felt way more reliable. I am interested to hear other people's experiences.
Jon Cluck
@cluckj
Feb 28 19:54
the only issue I've had is with the screen breaking; what kind of case/protection do you have on your rig?
djnoor
@djnoor
Feb 28 20:53
Oref1 question: If I have enableSMB_always set to true, does it matter whether enableSMB_with_temptarget or enableSMB_with_COB are set to true or false?
Does enableSMB_always bring higher sugars at night down faster since it can bolus instead of just delivering basals?
Does it also help with preventing spikes when exercising?
Scott Leibrand
@scottleibrand
Feb 28 22:20
correct: enableSMB_with_temptarget and enableSMB_with_COB are strict subsets of enableSMB_always
yes, SMB can bring BGs down faster by delivering insulin as a bolus and then setting temp basal to zero to avoid overcorrecting.
for exercise, it depends on your settings and how you use temp targets
if you want to discuss specific exercise scenarios, you'll need to describe what typically happens and how you typically deal with it, and we can discuss what other options you might want to try
djnoor
@djnoor
Feb 28 22:40
@scottleibrand Ok, thanks.
jaylagorio
@jaylagorio
Feb 28 23:43
@scottleibrand @cluckj Just seeing this now. If you have a G6 receiver and try to use the G5 data structures the CRC value won't line up, then it'll get calculated on the wrong data, fail due to that, and then your CGM loop log will say something like "invalid crc" over and over again
Same for G6 since the data structure will be off and fail the CRC there too.
So you shouldn't be able to loop on misaligned data.
Question for you though: I have the following log lines:
Preflight OK. Profile less than 60m old; Profile valid. Pump history updated through 2019-02-28T18:26:34-05:00 with 0 new records; meal.json refreshed: {"carbs":13,"nsCarbs":0,"bwCarbs":13,"journalCarbs":0,"mealCOB":12,"currentDeviation":0.84,"maxDeviation":1.9,"minDeviation":-0.1,"slopeFromMaxDeviation":-0.249,"slopeFromMinDeviation":0.179,"allDeviations":[1,1,1,2,2,0],"lastCarbTime":1551396004000,"bwFound":true}
Listening for 10s: .No interfering pump comms detected from other rigs (this is a good thing!)
Continuing oref0-pump-loop at Thu Feb 28 18:36:31 EST 2019
Checking that pump clock: "2019-02-28T18:36:31-05:00" is within 90s of current time: 2019-02-28T18:36:32-0500
Temp refreshed: monitor/temp_basal.json: {"duration":111,"temp":"absolute","rate":0}
{ "carbs": 0, "mealCOB": 0, "reason": "Could not parse input data" }
Note that it says "meal.json refreshed" and shows the carbs that I entered using bolus wizard
And then the last line, which is the contents of meal.json, which says carbs are 0 because it couldn't parse input data (but it parsed it just fine on the first log line)
Where should I look to troubleshoot? I got as far as checking pumphistory, profile, clock-zoned, and basalprofile because those look like the failure triggers here: https://github.com/openaps/oref0/blob/fa395b4b5fa02539ccd892a3d8fea62d4e3564e0/bin/oref0-meal.js#L52
I'm also noticing that bolus dots, carb dots, and basal modifications aren't taking place in Nightscout although the pred lines and CGM dots are being pushed there
Scott Leibrand
@scottleibrand
Feb 28 23:52
which dev commit are you using? 5cb8f9227bf175da183794d70027adc372e46827 has a bug fixed in 220adcf7c1ecf9e14a1e6d3aaee8169834436029 with latest-ns-treatment that might cause the latter behavior
sometimes you can get meal.json parse failures if it refreshes bad data from the pump...
jaylagorio
@jaylagorio
Feb 28 23:59
It says I'm totally up to date
So latest commit