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

9th
Feb 2018
Martin Haeberli
@mhaeberli
Feb 09 2018 01:49
getting LOTS of cgm-loop.log: AttributeError: 'NoneType' object has no attribute ‘iter_records’ - ideas to identify root cause and prevention sought
Scott Leibrand
@scottleibrand
Feb 09 2018 02:20
Do you have a G4 or G5 plugged in via an OTG cable?
If not, that is normal.
Martin Haeberli
@mhaeberli
Feb 09 2018 03:44
No- not plugged in. Ok I’ll chill
phoenixflame611
@phoenixflame611
Feb 09 2018 04:18
Just a general curiosity, what does everyone use as a max basal on your pumps? curious if i should kick that up a bit too. I know i need to update my profile from some autotune runs, but since it hits the cap i set most times i am curious to see if my setting is too low bc of that or if im not setting max rate correctly
Jake Punshon
@JakePunshon
Feb 09 2018 14:15
@kallnap @woodtrevor pump didn't respond to two days in rice (maybe due to it being saltwater), but responded very well to a few squirts of electrical contact cleaner : ) It's back up and looping. Thanks for the replies.
kallnap
@kallnap
Feb 09 2018 14:23
@JakePunshon great news
Craig Brenner
@cbrenner
Feb 09 2018 21:34

@danamlewis Thanks again for your help. I guess the biggest question I have is comparing meal bolusing between 0.5.x and 0.6.x. (with and w/o Fiasp). If a typical meal bolus with 0.5.x (w/o Fiasp) was setting eating soon (via an IFTTT button) an hour before and then entering carbs and bolusing the entire meal 15min prior to eating (my son was/is using the bolus wizard with 0.5.x). How does this change in 0.6.x.?

It sounds like the steps in 0.6.x w/o Fiasp would be:
(1) Setting eating soon (via an IFTTT button) an hour in advance (same as 0.5.x?);
(2) Calculate meal bolus manually (disable bolus wizard) and use easy bolus button to enter a pre-bolus of x% of meal 15 mins in advance;
(2a) Should x% above be 25%/50%/75%, etc. and assume 15mins in advance is still needed
(3) Enter all the meal carbs (probably via IFTTT buttons -- will need to create approx. carb size buttons for meals as my son doesn't make it a habit of using the NS app)
(3a) Should carbs be entered right before eating or in advance?
(4) Let eSMB handle bolusing for rest of meal (i.e., 100% - x% from above).

w/Fiasp
(1) is eating soon still needed?
(2) do you need to pre-bolus?
(3) enter carbs upon eating or earlier?
(4) assume this is the same

Anything else to consider? Thanks!

Dana Lewis
@danamlewis
Feb 09 2018 21:35
you don't necessarily have to change your bolus behavior (i.e. the calculation), unless you're also deciding to turn on SMB in 0.6.0
YDMV for deciding how much and when to meal bolus. but if you want to take the incremental approach - do the same size bolus (so for you, full)
you only need to easy bolus if you also want to SMB, or to enable the a52 risk preference
w/Fiasp - also YDMV. but when you switch to fiasp, you probably don't want to prebolus and even the eating soon might be too much depending on what's going on.
so: I would switch from 0.5 to 0.6 and keep doing the same stuff
Craig Brenner
@cbrenner
Feb 09 2018 21:37
@danamlewis pretty sure we are going to turn on SMB as I think this will work better for him as he is pretty much SWAGing it now for his meal carbs and letting OpenAPS try to high or low temp him back in range
Scott Leibrand
@scottleibrand
Feb 09 2018 21:37
many people have noticed that Fiasp is "too fast" if you bolus for the entire meal up front: It's often better to do a partial bolus and then let AMA/SMB do the rest
Dana Lewis
@danamlewis
Feb 09 2018 21:37
when you're ready, then choose SMB or settings change, or Fiasp as the next step. I wouldn't do both at once. if SMB is next, give that some time to figure things out
then add in Fiasp, which means you probably will stop ES-ing and pre-bolusing given the time change
Scott Leibrand
@scottleibrand
Feb 09 2018 21:37
:+1:
Dana Lewis
@danamlewis
Feb 09 2018 21:38
yea, SMB will probably help. but you still might want to go a day or two on 0.6. before enabling so you feel confident that nothing else is changed. otherwise you might overly extrapolate other noise into "0.6. is hard/bad/worse/whatever"
Craig Brenner
@cbrenner
Feb 09 2018 21:39
@scottleibrand @danamlewis He also has major issues with high fat meals that require using carb equivalents. Seems like eSMB would work much better in this case
Dana Lewis
@danamlewis
Feb 09 2018 21:39
likely will help, yes
Craig Brenner
@cbrenner
Feb 09 2018 21:47

@danamlewis Okay, so the plan will be to start 0.6.0 w/o SMB for a few days, then turn off bolus wizard and turn on SMB. Once we turn on SMB (w/o Fiasp), I am a bit confused. Are you saying we should do the same thing as before with 0.5.x minus the bolus wizard (i.e., use easy bolus) or use the steps I outlined above in 1 - 4. I was under the impression that with 0.6.x and eSMB you do not want or need to input the entire meal bolus upfront as SMB will handle any remainder (certainly better than temp basals).

Also, I recall some discussion about 'announcing you meal" so the system will respond. Can you provide additional context here?

Dana Lewis
@danamlewis
Feb 09 2018 21:49
YDMV - some people still do their normal timing and size meal bolus, and SMB picks up as needed. others do a partial and let SMB do more. it's based on how it works for you, and preference and comfort level with what it's doing. but many people find a partial with SMB gives them better outcomes than a full manual bolus up front.
announcing the meal is just inputting carbs without the bolus - can be done with IFTTT or wherever you want to enter the carbs
Craig Brenner
@cbrenner
Feb 09 2018 21:51
Okay, entering carbs w/o any bolus mean you are leaving SMB to do 100% of the bolusing?
Dana Lewis
@danamlewis
Feb 09 2018 21:51
yep.
Craig Brenner
@cbrenner
Feb 09 2018 21:52
hmm, guessing that only could work with Fiasp
Jon Cluck
@cluckj
Feb 09 2018 21:52
it works under a certain number of carbs too (but YDMV with that)
Dana Lewis
@danamlewis
Feb 09 2018 21:52
most who do that have Fiasp. a few people on Humalog or Novolog do that, though, for small scale meals. varies person to person
^ :+1:
Craig Brenner
@cbrenner
Feb 09 2018 21:52
and announcing a bit earlier than at the onset of eating..
Craig Brenner
@cbrenner
Feb 09 2018 22:00
Okay, so when we do start 0.6.0 w/SMB (w/o Fiasp), probably begin with entering a full meal bolus upfront and see how that works and perhaps incrementally reduce the upfront manual meal bolus by 10% or so to see how SMB works at taking over for my son. I'm guessing we will find a sweet spot that works for him. Curious what % of people are using a full upfront bolus with 0.6.0 and SMB (w/o Fiasp) and how many are using x% less upfront.
Oskar Pearson
@oskarpearson
Feb 09 2018 22:03
Hey folks. Long time no see. :)
Scott Leibrand
@scottleibrand
Feb 09 2018 22:03
welcome back! :)
Jon Cluck
@cluckj
Feb 09 2018 22:04
good idea taking it slow :) SMB can be (very) surprisingly effective!
Oskar Pearson
@oskarpearson
Feb 09 2018 22:04
:) Thanks!
Dana Lewis
@danamlewis
Feb 09 2018 22:18
@cbrenner :+1: good plan.
@oskarpearson :wave: :)
Martin Haeberli
@mhaeberli
Feb 09 2018 23:26
in the past, ordered batteries via Amazon / Atomic Market : https://www.amazon.com/Lithium-Rechargeable-connector-Atomic-Market/dp/B01KJ07A2Q and via AdaFruit - https://www.adafruit.com/product/328 - per their new specs, neither fit in the nice Robert Silver et al cases. The one that fits says: PKCELL LP785060 3.7V 2500 mAh but is smaller than the current AdaFruit one is spec’d to be … any suggestions for alternate vendors?