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

26th
Feb 2018
Tim Gunn
@Tornado-Tim
Feb 26 2018 02:31
Getting prepped for Looping and want to pre-order a 900mhz breakout board but I dont have a clue if the New Zealand Medtronic Pumps are the same as the U.S Australia Pumps and operate at that same frequency
If anyone knows would greatly appreciate the info :)
Dana Lewis
@danamlewis
Feb 26 2018 02:35
They're all WW if they're not North America (NA)
So depends on what your pump says?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:40
HI
Tim Gunn
@Tornado-Tim
Feb 26 2018 02:40
OK that makes it a bit easier :)
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:40
Guess who broke their rig!?
actually i was updating it and it said it ran out of space mid-update. Should I read the docs?
Scott Leibrand
@scottleibrand
Feb 26 2018 02:42
the docs has some tips. one of them is apt-get clean
Tim Gunn
@Tornado-Tim
Feb 26 2018 02:43
OK so all WW Medtronic pumps run at 868-869Mhz?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:46
@scottleibrand So many new errors, this is what happens when you drop off the face of the earth for 6 months and instal the dev version! #gooddecisionmakingskills
Dana Lewis
@danamlewis
Feb 26 2018 02:48
@Tornado-Tim yup. And the pump sn either has ww or na in it, for clarity
@Sgoldberg89 the definition of insanity is doing the same thing over and over and expecting different results ;)
But fwiw, I'd install master right now
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:49
Helpful or hurtful?
Dana Lewis
@danamlewis
Feb 26 2018 02:49
Which?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:49
ok, my disk is apparently out of space so... gonna get my broom out
I was just trying to match the sass.
Tim Gunn
@Tornado-Tim
Feb 26 2018 02:50
Nice Thanks @danamlewis , will report back so others will also find out what I find :)
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:51
serial.serialutil.SerialException: Attempting to use a port that is not open
Setting CGM time to Sun Feb 25 21:50:52 EST 2018
Dana Lewis
@danamlewis
Feb 26 2018 02:51
@Sgoldberg89 fair enough :smile: yup. Sweep away. If you weren't on 0.6.0 before, make sure to read the release notes and pay attention to the maxiob change if you use SMB
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:51
feel like this is the issue. I hate when my ports are closed
Dana Lewis
@danamlewis
Feb 26 2018 02:51
@Tornado-Tim great! And yes. Also consider PRing the docs if you wish that was more clear :smile:
Tim Gunn
@Tornado-Tim
Feb 26 2018 02:52
will do :)
Sgoldberg89
@Sgoldberg89
Feb 26 2018 02:52
It's like playing Catan with a horrible friend.
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:04
763.8MiB [##########] /home
591.3MiB [####### ] /var
555.9MiB [####### ] /root
86.2MiB [# ] /lib
Any ideas why these folders are so beefy?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:10
root/openaps1/.git/objects is full of files that are all 1.4 mb and have names like 8a, bf, etc. Can I delete some of these?
fingers crossed the answer to that was "yes."
Scott Leibrand
@scottleibrand
Feb 26 2018 03:16
you're upgrading to a version without git, so you should probably completely delete your ~/openaps1 folder and start fresh with a new oref0-setup
if you have a ~/openaps1/oref0-runagain.sh, save those contents for reference
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:17
ok thank you!
Scott Leibrand
@scottleibrand
Feb 26 2018 03:18
and if your ~/openaps1/autotune directory has autotuned your rig significantly differently from your pump, you might want to save that
or you can just update your pump basals to reflect it, and let the new install go from there
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:18
Will it automatically recalibrate those if I don't?
Scott Leibrand
@scottleibrand
Feb 26 2018 03:31
if you delete ~/openaps1 (and hence the autotune directory inside it) it'll start over from your pump settings, whatever those are
and autotune from there, starting at 4am tomorrow. if they're much different it will take a few days/weeks to get fully tuned again
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:33
Thanks, everything has been pretty messed up on my loop for a while so I think starting with a clean slate is probably best. I keep running into an issue with my nightscout during setup, just tried something, will let you know if it happens again. Not sure what the issue is because my website is fine.
File "/usr/local/lib/python2.7/dist-packages/git/repo/base.py", line 158, in init
raise InvalidGitRepositoryError(epath)
git.exc.InvalidGitRepositoryError: /root/myopenaps-cgm-loop
Could not run nightscout autoconfigure-device-crud
I know it said something about Git there, so I did what you suggested and deleted the OpenAPS1 file hoping that'd resolve it. Maybe I'll just start from ground 0
Scott Leibrand
@scottleibrand
Feb 26 2018 03:36
try deleting /root/myopenaps-cgm-loop too
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:36
k
same error
Scott Leibrand
@scottleibrand
Feb 26 2018 03:40
sounds like you might still have the old version of openaps that wants git
are you running oref0-setup?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:41
I've been using this: cd ~/src/oref0 && git checkout master && git pull && sudo npm install -g oref0
(since I was sternly (and appropriately) reprimanded for attempting to instal Dev.
)
Dana Lewis
@danamlewis
Feb 26 2018 03:50
That's right - but the git error shouldn't show up, may have an old myopenaps default install? Check and see what other directories you have?
Sgoldberg89
@Sgoldberg89
Feb 26 2018 03:52
aye captain1
!
Sgoldberg89
@Sgoldberg89
Feb 26 2018 04:00
I just kept deleting things and trying and it finally worked!
Now we'll see if it actually work works
Thanks as always for your help
Scott Leibrand
@scottleibrand
Feb 26 2018 04:26
heh, that's a solid troubleshooting method. :)
tepidjuice
@tepidjuice
Feb 26 2018 11:24
Is there a way to use openaps to get bgl data of a dexcom receiver?
PieterGit
@PieterGit
Feb 26 2018 12:22
@tepidjuice what do you mean with bgl. If you mean bloodglucose values: When connected to USB with the CGM=G4-upload option, oref0 will read the bloodglucose values from the connected receiver.
tepidjuice
@tepidjuice
Feb 26 2018 12:26
@PieterGit I had a minor bike crash and msashed my xbridge which I can't repair until tomorrow.
So I need to get my bgl to the rig
PieterGit
@PieterGit
Feb 26 2018 12:27
just use a proper microusb cable to connect it to your rig
tepidjuice
@tepidjuice
Feb 26 2018 12:27
I have a spare wixel and an old dexcom receiver.
I only have micro to full
PieterGit
@PieterGit
Feb 26 2018 12:28
you need to have otg micro-usb cable
tepidjuice
@tepidjuice
Feb 26 2018 12:28
Could I set up an openaps environment on my computer and upload the data from there?
PieterGit
@PieterGit
Feb 26 2018 12:28
if you now buy a buy a pi zero kit, you can have the converter. you'll need the pi zero w for your next rig :smile:
tepidjuice
@tepidjuice
Feb 26 2018 12:29
I just want the rig to run overnight until I can repair my xbridge
  • have NS-alarms that actually wake you up.
tepidjuice
@tepidjuice
Feb 26 2018 12:30
All the stores are closed here.
PieterGit
@PieterGit
Feb 26 2018 12:31
or use a rpi2 or rpi3 for one day
tepidjuice
@tepidjuice
Feb 26 2018 12:32
Is it not possible to set up an openaps environment on a laptop which just gets the bgl from the dexcom receiver and uploads it to NS?
As my laptop has full sized usb ports
PieterGit
@PieterGit
Feb 26 2018 12:32
if you have a debian based linux , that might be possible
tepidjuice
@tepidjuice
Feb 26 2018 12:33
I have gentoo based linux so it should be :)
PieterGit
@PieterGit
Feb 26 2018 12:34
it think that might be possible, just select G4-upload and you'll run across some errors because you don't have proper rig-pump comms.
if you have a carelink stick, you can select that. after install disable the pump-loop, because that will only run in errors
if you have a raspberry pi, i'll use that, because it's much more tested than gentoo based linux
tepidjuice
@tepidjuice
Feb 26 2018 12:39
Just need my cgm data. My rig is fine it's just my xbridge that is broken.
Dave Acklam
@dcacklam
Feb 26 2018 13:31
@Sgoldberg89
@Sgoldberg89 I had the same issue. Had to delete openaps python stuff completely and reinstall from scratch (I was able to keep my myopenaps directory, and restore that after the reinstall, to keep my config and data)
Sgoldberg89
@Sgoldberg89
Feb 26 2018 15:09
Thank you @dcacklam
jsw2480
@jsw2480
Feb 26 2018 15:18
hi - I've ordered the Explorer Block from Enhanced-Radio, and website says "500 additional units expected end of February". Anyone have any info on how that's going -- and when they may ship those out to customers? (I tried reaching them, but no luck). Thank you!
A...
@alhodge2_twitter
Feb 26 2018 16:15
@Sgoldberg89 I saw the same thing so I was concerned about ordering. Also, we are new to looped. If we order the Explorer Block, is that the only piece we need to order or do we need the Edison too?
Dana Lewis
@danamlewis
Feb 26 2018 16:16
@alhodge2_twitter you need the Edison, too. Edison is the computer chip, Explorer is the 'radio board'. see http://openaps.readthedocs.io/en/latest/docs/Gear%20Up/edison.html for details
I'd go ahead and order so once this batch comes in & starts shipping, you get it asap. I would expect they should ship soon...
A...
@alhodge2_twitter
Feb 26 2018 16:17
Thank you, @danamlewis. It looks like the Edison is no longer available.
Dana Lewis
@danamlewis
Feb 26 2018 16:17
Look on ebay, craiglist, etc. - still many out there for sale
A...
@alhodge2_twitter
Feb 26 2018 16:17
Thank you
Dana Lewis
@danamlewis
Feb 26 2018 16:29
A...
@alhodge2_twitter
Feb 26 2018 16:30
@danamlewis that is great. Thank you!!
Dana Lewis
@danamlewis
Feb 26 2018 16:30
:+1:
A...
@alhodge2_twitter
Feb 26 2018 16:37
@danamlewis is the Intel Edison Mini-breakout expansion kit (board only) correct?
Dana Lewis
@danamlewis
Feb 26 2018 16:38
not if it's board only. you need the actual Edison chip/module
A...
@alhodge2_twitter
Feb 26 2018 16:38
Ok. I will keep looking!
Dana Lewis
@danamlewis
Feb 26 2018 16:38
:+1: feel free to share a link when you find one for folks to eyeball & help you double check
for example: I went to ebay and put in "edison module" and this was the top listing: https://www.ebay.com/itm/Intel-Edison-Compute-Module/152922066859?hash=item239add8fab:g:SkYAAOSwNmNaktbq (and that would work, although you may could find another listing for bidding or cheaper, but just so you have a visual of what the Edison looks like for sale)
A...
@alhodge2_twitter
Feb 26 2018 17:05
Awesome. I did find the second and was going to ask. You beat me to the punch! Thanks for all of the help.
Dana Lewis
@danamlewis
Feb 26 2018 17:09
:)
Samuchco
@Samuchco
Feb 26 2018 18:15
Is there a way to see the amount of time that one is zero temping?
Dana Lewis
@danamlewis
Feb 26 2018 18:15
@Samuchco yes. but say more about what you're trying to find out, and where?
Samuchco
@Samuchco
Feb 26 2018 18:19
I'm trying to find out if zero temping is causing roller coasting. I'm assuming that there should be a delicate balance between programmed basals and zero temping. Too much zero temping should indicate adjustment to your profile
Scott Leibrand
@scottleibrand
Feb 26 2018 18:20
are you running autotune? that is often helpful in figuring out these kinds of things
Dana Lewis
@danamlewis
Feb 26 2018 18:20
not necessarily
Scott Leibrand
@scottleibrand
Feb 26 2018 18:21
it's also important to focus not just on the zero temps themselves, but on the resulting IOB. you can scroll back in NS to see how negative IOB is getting as a result of the zero temps, and how high it's getting from the corrections done as BG rises
Samuchco
@Samuchco
Feb 26 2018 18:22
@scottleibrand @danamlewis I'm thinking of zero temping as an indicator
Scott Leibrand
@scottleibrand
Feb 26 2018 18:23
most people intuitively think zero temping is way more powerful than it actually is, because they ignore all the insulin that was delivered before and after the zero temp (especially via boluses)
Dana Lewis
@danamlewis
Feb 26 2018 18:23
I would run autotune first to evaluate any trends in basal changes that are needed. then you can scroll back and look at IOB. but low temps alone don't necessarily indicate basal changes are needed, given there are basals and high temps that can influence lower temps given later.
Tim Street
@tim2000s
Feb 26 2018 18:25
It’s also fair to say that in most cases it’s not zero temping causing roller-coastering.
Scott Leibrand
@scottleibrand
Feb 26 2018 18:29
yeah, the algorithm is dampened to prevent oscillation. the only things we see causing roller-coastering that extends outside of ~70-160 mg/dL is when a human gets involved and does correction boluses, rescue carbs, and/or aggressive temp targets in response to high or low BG.
Samuchco
@Samuchco
Feb 26 2018 18:35
@scottleibrand "you can scroll back in NS to see how negative IOB is getting as a result of the zero temps, and how high it's getting from the corrections done as BG rises" can you explain, and is there a way to see it in a chart or table, ala autotune, rather than scrolling back?
jsw2480
@jsw2480
Feb 26 2018 18:44
ihmi
ha! sorry for the nonsense typing!!
Samuchco
@Samuchco
Feb 26 2018 19:06
@tim2000s While zero temping is not the cause of the roller-coasting, and temp targets and carb rescuing i.e. manual interventions affect the swings in BGs, I'm thinking of "Time in Zero Temp" as an analytical tool. But perhaps not.
live4sw
@live4sw
Feb 26 2018 19:08
Could not agree more on zero temping. While zero temps are a great tool, and are great (particularly at night) for leveling out, I've found that the ability to halt a low during the day of a zero temp (at least for me) is pretty weak. More of a way of mitigating significant lows (and working with SMB to handle meals well) than an everyday tool.
In my experience, a zero temp alone only singlehandedly heads off a low for me at most, once or twice a month. And I don't think I've ever seen one singlehandedly cause me to have a rebound high.
Tim Gunn
@Tornado-Tim
Feb 26 2018 19:16
hi @danamlewis the NZ funded Paradigm pumps are Canada/Australia so basing it off this: http://openaps.readthedocs.io/en/latest/docs/Troubleshooting/Pump-rig-communications-troubleshooting.html?highlight=frequency%20mhz it looks like i need the 916Mhz correct?
Scott Leibrand
@scottleibrand
Feb 26 2018 19:17
yeah, Canada uses the same frequencies as the US, and I think AU does as well.
Tim Gunn
@Tornado-Tim
Feb 26 2018 19:19
Cool thanks @scottleibrand
Just need to wait for my Animas Vibe pump to expire with less than 2 months to go.... then can get the 722/522
Samuchco
@Samuchco
Feb 26 2018 19:25
@scottleibrand I guess it would be nice to see the negative IOB charted as part of the NS basal render
Scott Leibrand
@scottleibrand
Feb 26 2018 19:27
@Samuchco there is a way to see that in NS reports, FWIW
Samuchco
@Samuchco
Feb 26 2018 19:28
@scottleibrand as a graph?
Scott Leibrand
@scottleibrand
Feb 26 2018 19:28
yeah. check IOB on the day to day graph I think it is
th122
@th122
Feb 26 2018 19:28

@Samuchco if you want the data in a table, rather than scrolling through ns, you could convert the data you want from the logfile of the pump loop:

grep "^{" pump-loop.log | jq -r '[ .time, .iob, .basaliob, .bolusiob  ] | @csv’

for an example that would just list the iob by timestamps. in a comma-separated list.

Samuchco
@Samuchco
Feb 26 2018 19:29
@th122 @scottleibrand thanks, will check both
th122
@th122
Feb 26 2018 19:37
@scottleibrand And again I’m impressed! Following your advice I tried checking “Insulin distribution” as well, and found that it also sums up positive and negative temp basal in addition to the well known metrics. :+1:
reviloogel
@reviloogel
Feb 26 2018 19:49
@jsw2480 had ours on order for nearly 2 months, they say (via email) that delivery to them is expected last week Feb and then they will ship , so hopefully ours arrives in next 2 weeks
garykidd
@garykidd
Feb 26 2018 20:26
@th122 looks like SMBs treated as bolus (which I would expect) on that stat summary...so just keep that in mind. lookin gat today for me on the day to day---the insulin distribution isnt what I expected to see---showed way more negative basal than I would've thought.
th122
@th122
Feb 26 2018 20:30
@garykidd I’m nowhere near SMB yet, so I don’t have any data on it so far. If you’re looking at the smb details in the punp-loop log, does it show how much of the bolus was “borrowed” from the basal to come? I’d assume it to, given that everything I’ve seen so far has been very transparent at detailling out what went where and why.
Scott Leibrand
@scottleibrand
Feb 26 2018 20:32
openaps has no way to reliably differentiate which boluses are manual vs SMB once they've been delivered, so there's no way to report that distinction to NS either
Martin Haeberli
@mhaeberli
Feb 26 2018 22:38
rig temp up to 88 - 90 C!
smells like burning
turned it off of course
wonder if maybe an emergency temperature based shutoff at a temp > say 84 C might be a plan ..
Martin Haeberli
@mhaeberli
Feb 26 2018 22:44
any thoughts about possible root causes appreciated
Scott Leibrand
@scottleibrand
Feb 26 2018 22:53
we already stop doing most things when rig temp is that high
if it smells like burning, then it's not driven by CPU temp: 90C is no where near hot enough to burn anything
might have a hot spot elsewhere on the board
is there physical evidence of that?
Martin Haeberli
@mhaeberli
Feb 26 2018 22:58
I’ll be looking at the evidence later today. Reports so far reliant on testimony via text message and papertrail
Martin Haeberli
@mhaeberli
Feb 26 2018 23:43
@scottleibrand what is the threshold temperature for ‘stopping’ work...