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

23rd
Apr 2018
Tim Gunn
@Tornado-Tim
Apr 23 2018 00:45
Any idea why I havnt been zero temp'd when I am on the verge of being low with no COB?
Screenshot from 2018-04-23 12-44-43.png
I usually understand the logic behind the oref0 algo, but this has me stumped.....
Graph.png
Dana Lewis
@danamlewis
Apr 23 2018 00:51
Looks like eventual BG is at target so it's setting a neutral temp
Tim Gunn
@Tornado-Tim
Apr 23 2018 00:56
yea OK, this prob the first time I have seen the algo come across this, will be very interesting to see if its prediction comes true
Tim Gunn
@Tornado-Tim
Apr 23 2018 03:12
I still haven't had any reply from Enhanced Radio Devices via email about the lipo charge function of the board not working on one of my boards, to see if I can fix myself or needs a swapout. Is there any other way to contact them......?
Chickey101
@Chickey101
Apr 23 2018 07:21
@danamlewis With help :) nano pump.ini (thats not right but I don't have the notes infront of me) I had some assistance, thanks Simon, I ended up running the script again and everything starting going to NS. But...., I'm now sitting in front of the rig cursing again. Anyone have any trouble logging into the rig? ssh just goes nowhere, change cables, turn off, unplug, have a beer (only one), no good. Im away from home and wanted to setup the hotspot wifi so NS will update. I'm medtronic cgm. bit painful crabs don't go into the system unless online.
Michael Spradling
@M1Sports20
Apr 23 2018 13:52
I have multiple rigs setup and have initially synced up autotune results. My question is why would autotune results get out of date between the rigs. They are both using the same data to calculate autotune results correct?
Scott Leibrand
@scottleibrand
Apr 23 2018 14:38
Usually that happens when one rig is not powered or doesn’t have Internet at 4am.
Michael Spradling
@M1Sports20
Apr 23 2018 14:56
Hm, they are both powered, however, I forget I put dev on one of the units. Was just looking at dev
Craig Brenner
@cbrenner
Apr 23 2018 17:14
@danamlewis Hi Dana, just wondering if you had a chance to experiment with the IFTTT button sequencing bug we discussed a few days ago? https://gitter.im/nightscout/intend-to-bolus?at=5ada4dbf6bbe1d27390a8322
Eric
@ecc1
Apr 23 2018 17:16
If anyone is using Medtronic CGM in their loop, I could use your help in adding MDT CGM support to the Go code used by the dev branch of openaps (so far it only handles Dexcom). I'd need you to run some commands on your rig to capture "raw" glucose history pages, plus the openaps command that decodes those pages; I can provide more details if you PM me.
Craig Brenner
@cbrenner
Apr 23 2018 17:20
Anyone have any updates on progress with the Tandem t:slim and OpenAPS or the open API initiative that as announced from JDRF? I haven't seen much activity on the Tandem Slack channel and haven't heard anything new from JDRF. We got the t:slim X2 a few months ago as a hedge in anticipation of hopefully moving off Medtronics and back to the t:slim. Would be great to have choices other than Medtronic.
Scott Leibrand
@scottleibrand
Apr 23 2018 17:23
@danamlewis is in a room right now with FDA, JDRF, and a bunch of pump manufacturers discussing the JDRF Opening Protocols initiative. So it's moving forward, but I wouldn't expect it to move quickly enough to produce something useful for any existing hardware that isn't already loopable. (On that note, the Roche Combo is loopable with AndroidAPS, and already has 100 active users. It's FDA-approved in the US, but hard to get here since Roche pulled out of the market.)
This message was deleted
garykidd
@garykidd
Apr 23 2018 17:40
just wondering if folks are starting to receive delivery of pi hats that were scheduled for april delivery...?
Michael Spradling
@M1Sports20
Apr 23 2018 17:49
Another question(Sorry I know I am full of them, just really tring to understand)
COB: 35, Dev: 71, BGI: -8, ISF: 61, CR: 8, Target: 91, minPredBG 158, minGuardBG 105, IOBpredBG 39, COBpredBG 195, UAMpredBG 141; insulinReq 1.1; setting 90m low temp of 0U/h. Microbolusing 0.5U.
TBR: 0
IOB: 2.692
Autosens: 0.97
The predictions don't make sense to me. I see that these line up with the purple lines and they all seem to goto down really quick., but why would zero-temping line go down faster than the IOB purple line?
I have used this system a lot, and its worked wonderful(thanks). Just trying to understand and fine tune it.
Scott Leibrand
@scottleibrand
Apr 23 2018 17:58
the IOB prediction includes an extrapolation of current deviations for the next hour. the ZT line is based on the scenario that current deviations immediately cease
Craig Brenner
@cbrenner
Apr 23 2018 19:01
@scottleibrand Thanks for the update Scott. I'm not sure i understand your comment, "...I wouldn't expect it to move quickly enough to produce something useful for any existing hardware that isn't already loopable." Seems to me that this is primarily a software issue not hardware and I would think an over the air update from Tandem could make the pump loopable if they included the APIs (Open Protocol). I understand that any major update would likely require FDA approval but I don't see any need to make hardware modifications--esp since the X2 is the platform for their future AP software.
Michael Spradling
@M1Sports20
Apr 23 2018 19:05
@scottleibrand Thanks. So this seems normal, i guess
Scott Leibrand
@scottleibrand
Apr 23 2018 19:54
@cbrenner I'm talking more about what I expect they will do than what they could do. We'll see what comes out of the FDA/JDRF meeting, but I expect the path between those discussion and actual software updates to existing pumps, or new pumps, will be long.
James Babcock
@jimrandomh
Apr 23 2018 21:53
Are there documents to give to the FDA and pump manufacturers explaining what's needed for looping, arguing that it's safe, etc? And if not, is help needed in writing them?
Marek
@CarnageMarkus
Apr 23 2018 22:07
Screenshot_2.jpg
hello, I am looping for some while now, but never got nightscout to actually work. I followed this guide https://openaps.readthedocs.io/en/latest/docs/While%20You%20Wait%20For%20Gear/nightscout-setup.html, and actually there is data on the website, exept it throws a lot of console errors and CGM graph is bugged (see picture above) . Any ideas what might be going wrong ? (Edison loop, MDT CGM with 522 pump). please help. I am looping blindly...
James Babcock
@jimrandomh
Apr 23 2018 22:14
So, first--looping blindly is obviously not recommended, you should probably power off the rig until you've done enough troubleshooting of Nightscout that it has to be on
Of all the error messages in the Chrome console, the best hint is going to be the first one, so scroll up and looking there
Also paste a link to the fork you made of cgm-remote-monitor on GitHub
Marek
@CarnageMarkus
Apr 23 2018 22:24
I am using latest dev. https://github.com/CarnageMarkus/cgm-remote-monitor/ (deployed yesterday, in hopes it will work.) https://github.com/CarnageMarkus/cgm-remote-monitor/
James Babcock
@jimrandomh
Apr 23 2018 22:26
Did you start with dev, or switch to it after master didn't work?
Marek
@CarnageMarkus
Apr 23 2018 22:28
I am not sure, but it had same problem, that has been like 3 month ago.
image.png
the errors seem like there is corrupted/bad data arriving ?
James Babcock
@jimrandomh
Apr 23 2018 22:45
Possibly corrupted incoming data, possibly corrupted old data that got stored in Nightscout's database at some point
I think the next thing to try would be deleting everything from NS's database, or pointing it at a fresh database, to eliminate any junk that might've gotten in there during install or on a previous debugging attempt