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

2nd
Apr 2018
gszakacs
@gszakacs
Apr 02 2018 03:36
Yes, it's BTLE. https://fccid.io/PH29588
Alan
@Alzibiff
Apr 02 2018 05:58
image.png
Looking for some help. My overnight graph was almost a perfect straight line ... but way above my target. I am relatively new to all this and cannot understand why the basal line is so up and down - surely it should have stayed quite flat as well in order to bring my BGs back to within range. Any advice on what changes I need to investigate will be very gratefully received.
James Babcock
@jimrandomh
Apr 02 2018 06:31
It looks at the rate-of-change of your BG together with the amount of extra insulin it's given above recently, to predict how much the already-delivered extra insulin is going to bring you down. In this case, it delivers some extra insulin, predicts that it's going to work to bring your BG down, and then it doesn't work.
Either you were extra insulin resistant for some reason (eg your infusion set has gone bad), or your settings are off (insulin sensitivity factor much too high or basal much too low).
James Babcock
@jimrandomh
Apr 02 2018 06:36
It looks like the problem started a few hours after an infusion set change (assuming the cannula-age in that screenshot is accurate), so that is most likely the root cause
Alan
@Alzibiff
Apr 02 2018 06:37
There was an infusion set change just before my evening meal which may well have been the 'extra insulin resistant' culprit, guess I will have to see how it goes today. If that is the cause, I suspect that this is a common occurrence for others as well as me and if that is the case, are there any workarounds? Thank you Jim.
James Babcock
@jimrandomh
Apr 02 2018 06:38
Different types of infusion sets and different body locations give varying results
Alan
@Alzibiff
Apr 02 2018 06:41
Yes, I have struggled with that problem since I went on a pump 9 years ago. How do successful loopers cope with this though? Different ISF profiles every 3 days or so? Still learning!
James Babcock
@jimrandomh
Apr 02 2018 06:43
Better to identify it as an infusion-set problem quickly and put in another one, rather than try to compensate with extra insulin, because compensating with extra insulin tends not to work.
OpenAPS does have a component called autosens which is meant to detect and compensate for fluctuations in sensitivity. But there's only so much it can do.
Alan
@Alzibiff
Apr 02 2018 06:52
I agree with you saying that infusion set problems need looking at quickly but real problem sets - the ones which simply don't seem to work because of kinked cannulas or similar, manifest themselves with a very steep upwards climb for me and it is these which I change as soon as the problem becomes obvious. In the past, I have used a temporary basal of something like 175% for a couple of hours after a set change and that always seemed to work well.
James Babcock
@jimrandomh
Apr 02 2018 06:53
Out of curiosity, where (body location) do you put them?
Alan
@Alzibiff
Apr 02 2018 06:56
Abdomen. Used an omnipod for 18 months and extended my sites to include arms as well but since going back to Medtronic, have always used my stomach area.
James Babcock
@jimrandomh
Apr 02 2018 06:56
Other possibility: if dinner was both an underestimate of amount of carbs, and slow digesting, that could maybe explain it
(Would have to be quite a large underestimate, though)
Alan
@Alzibiff
Apr 02 2018 06:58
Think you are spot on there - noodles last night and probably not enough bolus coupled with the set change produced that perfect storm!
Did a correction bolus earlier on which is bringing things down nicely so it appears that the infusion set has bedded in. Now need a strategy to avoid this happening again in 3 days time.
cameronrenwick
@cameronrenwick
Apr 02 2018 10:20
@gszakacs thanks for this info re G6.. the pix in particular will be super helpful come battery replacement time!!
mkericson
@mkericson
Apr 02 2018 12:07
image.png
Had this invalid date error this morning.
Actual time in pump and real time is 8:05AM EST
mkericson
@mkericson
Apr 02 2018 12:20
--- Found problem -- I think
Rig is out of memory...
image.png
Dana Lewis
@danamlewis
Apr 02 2018 12:52
@Alzibiff make sure you look at the logs for that time period too. Don't just go based on the pictures. Take a look and make sure you're not bumping into any setting caps that you may want to change if that occurs multiple times.
Also, don't fixate on the idea that the blue basal line should be flat. Your basal is an average of your insulin needs. You will need more or less. Could be because your baseline basal is off, could be because site age, food, sick, site, exercise etc. but basal changes from OpenAPS aren't a negative thing. Take a look at Autotune logs if you wonder if your baseline pump values need to be changing.
mkericson
@mkericson
Apr 02 2018 13:08
Back in business. Besides cleaning out some files in ./log is there other areas that are safe to clean out on the rig. Looks like my rig stops working when it gets to about 1.4G. When I clean out some of the log files I can get it down to 1.1G. That so far seems to last about 2.5 Months.
So are there other areas to purge to clean out more space?
Alan
@Alzibiff
Apr 02 2018 13:24
@danamlewis Thank you for the comments. Yes, I get the idea that the basal will be bumping up and down and that I am likely to be on a temp basal for much of the time it was just that my initial thoughts were that it should have been a long higher than 'base' line for the period I was looking at. My max basal set in the pump is just less than 1u so if I understand things correctly, with a standard prefs set up, isn't the max that algorithm will set actually approaching 3u?
Scott Leibrand
@scottleibrand
Apr 02 2018 15:34
It’s easier to look at the IOB at the time (by scrolling NS back using the bottom graph) than to try to eyeball the basal graph
If you had an IOB high enough that it should bring you down to target (based on your ISF), it won’t try to do much more than that.
If you regularly have elevated BG after a site change, you can set a low temporary target to make OpenAPS be a bit more aggressive. There is also a setting to make low targets automatically adjust the sensitivityRatio upward, to generate a stronger reaction for a given low target.
Alan
@Alzibiff
Apr 02 2018 16:53
Thank you @scottleibrand , useful information on how to navigate the NS site. Please cna you point me in the direction of how to automatically make low targets adjust ISF?
Alan
@Alzibiff
Apr 02 2018 17:23
Or is that one of those IFTTT things?
I believe this is it
Kelly Watson
@kellyisworking
Apr 02 2018 17:43
@live4sw Were you able to resolve your initial issues with BT tethering with pixel 2? We're going to work on setting this up fairly soon. Are you still using Google Fi?
Alan
@Alzibiff
Apr 02 2018 17:48
@PeetuHongisto Top! Thank you.
live4sw
@live4sw
Apr 02 2018 19:22
@kellyisworking I was able to resolve my issue. I'm not using Google Fi, but I am using a similar reseller of wireless service.
It took several hours (and calls to Google and the carrier) but I figured out that the issue related to enabling tethering with my carrier. See item 2 of http://openaps.readthedocs.io/en/latest/docs/Customize-Iterate/bluetooth-tethering-edison.html?highlight=tethering#additional-troubleshooting-steps-for-some-carriers
Feel free to ping me/the channel if you have any trouble with it after you give it a try
Kelly Watson
@kellyisworking
Apr 02 2018 21:14
Thanks @live4sw :thumbsup:
Alexander Teepe
@ateepe
Apr 02 2018 22:50
Hi, I'm new here. I could use some help understanding OpenAPS's oref0-determine-basal script. I'm a student developer working on a project and need to call it from an external program (outside of OpenAPS) to get the recommended basal rate but the documentation for the inputs to oref0-determine-basal.js seem to be incorrect or incomplete (referring to http://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/Understand-determine-basal.html?highlight=determine%20basal). Any help would be greatly appreciated, Thanks!
James Babcock
@jimrandomh
Apr 02 2018 22:52
Assuming you have an online OpenAPS rig: look at the invocation in oref0-pump-loop.sh to get the paths of the inputs, and look at what's at those paths for a sample
Also note that oref0-determine-basal is not self-contained; most of the meat is in lib/determine-basal/determine-basal.js
Alexander Teepe
@ateepe
Apr 02 2018 22:59
Thanks. I have OpenAPS installed, but I'm not running it. I'm running Loop instead, and pulling the data from NightScout, and need to format it to use for OpenAPS. I hope that won't be too difficult to get working
James Babcock
@jimrandomh
Apr 02 2018 23:49
OpenAPS already can pull data from Nightscout, I believe
I'm not really familiar with Loop; where does it fit into your planned setup? What devices do you have and which talk to which?
Paul Heltzel
@pheltzel
Apr 02 2018 23:56
Could use some advice. I'm only on day 3 of OpenAPS, but I'm having some trouble with highs. I think I might be misunderstanding some of the terminology.
On the pump, the BG target is set to 100-250.
I've noticed some discussion of wide ranges causing trouble highs for some.
Are these the BG_TARGET_BOTTOM/BG_TARGET_TOP settings? I'm just unclear on where this could be tightened up.
I'm using Autotune and Autosens