@mariano42, I'm using a little hack to help prevent over bolusing using the bolus wizard, on the pump I set the high target much higher (160) and then made modification to the profile tool to use the low target as a high target, it worked great for school since they didn't know anything about Net IOB
for that to work you need to have really good loop uptime
Getting this error
edison@doubled:~/Documents/try5$ openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
Warning: could not parse monitor/carbhistory.json
/home/edison/src/oref0/bin/oref0-meal.js:79
if (glucose_data.length < 36) {
^
TypeError: Cannot read property 'length' of undefined
at Object.<anonymous> (/home/edison/src/oref0/bin/oref0-meal.js:79:21)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)
at startup (node.js:119:16)
at node.js:906:3
reporting monitor/meal.json
This might help. openaps/docs#228
yeah I don't have carbhistory.json. Anyone know how to make this file?
I have an empty file called
settings/carbhistory.json. Works just fone.
isn't the point that AMA needs carbhistory to function properly?
All discussed in that link
basically it can get the carb information from either the pump or nightscout. If from NS, you can create that carbhistory.json file.
If already pulling from pump then no need for the info from NS
@GeoffWyche raspberry pi @dm61 I don't have the CareLink plugged in yet... I feel like I'm not following the docs right then? I was following along the walk through and then looked at the openaps/docs#129 but now looked at the https://github.com/openaps/openaps doc and it seems to be another mostly similar version, but slightly different way of setting stuff up... I'm going to try to sort through/synthesize these, but can anyone suggest a primary location to start?
got it thanks
I'm in the beginning of Phase 2 (without having NS set up), but I've read through and remember reading that IOB can't be calculated for square wave boluses... is that correct? If so, I'm just curious why that is.
Yes that is correct. It doesn't work with sq wave boluses.
It was explained to me for one of two reasons:
1) Just hasn't been of high enough priority for folks to work on it
2) I believe there is an issue in that when you pull the pump history it doesn't tell you about the sq wave until the sq wave has completed giving you the insulin. And thus can't be used in a IOB calculation. Don't know if that is because of #1 or something in the basic design of the pump
1) Just hasn't been of high enough priority for folks to work on it
2) I believe there is an issue in that when you pull the pump history it doesn't tell you about the sq wave until the sq wave has completed giving you the insulin. And thus can't be used in a IOB calculation. Don't know if that is because of #1 or something in the basic design of the pump
Pretty sure it’s #2.
When I do a sq wave, and wait to pull pump history until the sq wave completes, it shows up as two boluses. The initial one and then another one that represents the sq wave but the timing is very close to the initial bolus regardless of the time the sq wave is set over
So not very accurate in the IOB calc
The bigger issue for me is I have it so that my loop won't run while the pump is bolusing to prevent some other issues. But that means when I have a sq wave going, the loop won't run at all.
I don’t think the pump ever accepts a temp basal when a square wave is running...
@jasoncalabrese that makes sense.. Like I said, i am really most interested in having a bit of an audit trail behind my overrides. Basically have it report out the pumps recommendation compared to what a modified bc would recommend.
Paul - you are correct. Looks like you can do it manually but doesn't look like you can set it with openaps while a sq wave is going
I use temptargets to simulate a square wave. It’s been working very well so far.
I see. Slightly related, but also something I'm curious about is whether the pump's calculation is available to retrieve, and if so, if it is used regularly by anything for OpenAPS's calculations?
...the pump
's calculation of IOB is available...
@mariano42 you would have an audit trail there, OpenAPS doesn't do anything to the bolus at all, just pulls the data from the pump history, my target override is just a way to prevent someone that doesn't understand or notice the Net IOB from always over bolusing
@garetis we don't use or even look at the pumps IOB since it's almost always wrong, but if you're uploading all the treatment to Nightscout you can see the bolus only IOB
Net IOB is ~0, the Basal IOB is -0.35, and the Treatments IOB (based only boluses) is 0.34
@thacherT1D you're done with Phase 0, correct? Then, I'd suggest you skip Phase 1, go to Phase 2. In Phase 2, you do need to have your CGM device and CareLink stick plugged in. Phase 2 is very important because that's where you actually start talking to your pump and your CGM device, and where you get to learn basic openaps commands, reports, aliases. Once you complete Phase 2, you'll be in good shape to do the rest. In particular, you can then go back to openaps/docs#129 notes (which should be included in a new version of Phase 1), so you can visualize what's going on, then move to Phase 3 etc.
How do carbs expire in OpenAPS? I have OpenAPS saying Ella is at 66 COB and NS saying she has 0g and NS is saying the last time she 17G at 3:30pm PST which is correct.
I have seen this before where I think Carbs should be expired but they aren't so OpenAPS won't pull insulin away when it should.
This is carbhistory.json:
Make sure the glucose file you use covers the time period of the carbs
In the latest AMA there's a check for that
WIll check...thanks, @jasoncalabrese
Yep...I've got glucose numbers with no discernible breaks
Or is there a way I can force making her COB 0?
What does the output look like when oref0-meal runs
If you don't run the report and hard code a value
Editing the file
let me track down what I need to run to generate it
and thanks!
Should be in your log output
ah..easy then
let me loop
STARTING GET-BG
ns-glucose://text/shell/monitor/ns-glucose.json
reporting monitor/ns-glucose.json
STARTING ENACT
{"carbs":104,"boluses":3.1,"mealCOB":34}
{"ratio":1.23}
{"iob":0.005,"activity":0.0114,"bolussnooze":0,"basaliob":-0.034,"netbasalinsulin":0.4,"hightempinsulin":0.65,"time":"2016-06-11T01:55:51.000Z"}
{"delta":-2,"glucose":62,"avgdelta":-1.333}
{"duration":25,"rate":0,"temp":"absolute"}
{"max_iob":2,"type":"current","dia":4,"current_basal":0.3,"max_daily_basal":0.35,"max_basal":1,"min_bg":90,"max_bg":100,"sens":100,"carb_ratio":15}
Adjusting basal from 0.3 to 0.37
Adjusting sens from 100 to 81.3
Carb Impact: 2.6 mg/dL per 5m; CI Duration: 11.8 hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 3.1 hours
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":62,"tick":-2,"eventualBG":154,"snoozeBG":78,"predBGs":{"IOB":[62,60,58,55,53,52,52,51,50,50,51,51,51,51,51,51,51,52,52,53,53,54,54,55,55,56,56,57,57,58,58,59,59,60,60,61,61,62,62,63,63,64,64,64,65],"aCOB":[62,67,72,77,82,89,95,101,107,115,122,129,135,142,148,154,160,166,171,177,182,187,191,196,200,203,207,210,213,215,218,220,222,223,225,226,226,227,227,228,228,229,229,229,230],"COB":[62,60,58,56,55,55,56,56,57,59,62,64,66,69,71,74,76,79,81,84,87,90,93,95,98,101,103,106,108,111,113,116,118,121,123,126,128,131,133,135,138,140,142,144,147,149,151,153,154]},"reason":"COB: 34, Dev: 16, BGI: -4.63, ISF: 81.3, Target: 95; BG 62<70, but 25m left and 0 ~ req 0U/hr: no action required"}
No action required
This?
Also, she didn't have 104 carbs that I know of...she was at camp today...but not in the last three hours
Something got jacked at some point...once her carbs expire, everything will go back to normal and be awesome for weeks...but this has happened maybe two or three times and want to track it down...
On our pump of you cancel a bolus wizard the carbs still get entered
Jason, I love you
It was mostly happening at school, I delete them from NS
Ha
Wait...she did have a dual wave bolus with pizza today
that might be it, too
Need something like zoning to filter them out of the pump history
Shoot...Everything look right in NS
Would a dual wave bolus mess with anything?
So six hours ago, she suspended the remainder of a dual wave bolus and then did another dual wave bolus that should have ended 3.5 hours ago...that's the only weird thing we did today
Don't think so, do you see the output for each treatment with carbs?
when you say output do you mean in carbhistory or where?
Would look something like:
Jun 10 19:31:29 indy-e1 openaps-loop: ------+->>>>+=----------+----++->+>>>+++++++---+=---------+++-++-+>>>+>>>++++--+-+---+- 58% of non-meal deviations negative (target 45%-50%)
Jun 10 19:31:29 indy-e1 openaps-loop: Excess insulin sensitivity detected: ISF adjusted from 80 to 100
Jun 10 19:31:29 indy-e1 openaps-loop: 0
You'd see that for each carb entry in the last dia*2.5
I need to throw some timestamps in my log file:
STARTING ENACT
{"carbs":128,"boluses":3.1,"mealCOB":60}
{"ratio":1.24}
{"iob":0.094,"activity":0.015,"bolussnooze":0,"basaliob":0.028,"netbasalinsulin":0.45,"hightempinsulin":0.65,"time":"2016-06-11T01:48:10.000Z"}
{"delta":1,"glucose":66,"avgdelta":1}
{"duration":29,"rate":0.325,"temp":"absolute"}
{"max_iob":2,"type":"current","dia":4,"current_basal":0.3,"max_daily_basal":0.35,"max_basal":1,"min_bg":90,"max_bg":100,"sens":100,"carb_ratio":15}
Adjusting basal from 0.3 to 0.37
Adjusting sens from 100 to 80.6
Carb Impact: 7 mg/dL per 5m; CI Duration: 7.7 hours
Accel. Carb Impact: 10 mg/dL per 5m; ACI Duration: 5.4 hours
determine-basal://text/shell/enact/suggested.json
reporting enact/suggested.json
{"temp":"absolute","bg":66,"tick":"+1","eventualBG":281,"snoozeBG":100,"predBGs":{"IOB":[66,66,68,68,69,69,70,71,71,71,71,71,71,71,71,70,70,70,70,70,70,70,70,70,71,71,71,71,71,71,71,71,71,71,71,71,71,71,72,72,72,72,72,72,72,73],"aCOB":[66,70,75,80,85,90,97,103,110,117,125,133,141,148,156,163,171,178,185,192,199,206,213,219,225,232,238,243,249,255,260,265,271,275,280,285,290,294,298,302,306,310,314,317,320,323,326,329,332],"COB":[66,67,69,71,74,76,81,85,89,93,99,105,111,116,122,128,133,139,145,150,156,161,167,172,177,182,187,192,197,202,207,212,216,221,225,230,234,239,243,247,251,255,259,263,267,270,274,277,281]},"reason":"COB: 60, Dev: 42, BGI: -6.04, ISF: 80.6, Target: 95; BG 66<70, avg delta 1.00>0, temp 0.325 ~ req 0.37U/hr"}
No action required
STARTING TEMP BASAL
STARTING UPLOAD
Uploaded
STARTING GET-SETTINGS
STARTING AUTOSENSE
>>>++>>>>>>>>>>>>>>>>>>>>>++>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>++>+++>+>>>>>>>>>>>>>>>>>>>++>>>+>>+>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>++++++-+-+++++-+++-++--+-=----------++++++++++-------+-----------++-+-++----------++-+++>>>>+++++++++-+++--++++++++--++++++++++++ 38% of non-meal deviations negative (target 45%-50%)
Excess insulin resistance detected: ISF adjusted from 100 to 81.30081300813008
detect-sensitivity://text/shell/settings/autosens.json
reporting settings/autosens.json
STARTING WAITFORBG
That's got some carbs I don't recognize
This is a crazy example...she was 66 and it thought she had 128 or 60 COB when she had 0 and that her eventual would be 281...yoinks!
What is CI Duration? 7.7 hours for her to burn off her carbs?
How is that determined?
And from the NS Profile editor what is a decent Carbs activity / absorption rate: [g/hour]? We went with the default with 20.
CI duration is how long it would take at the current rate of carb absorption.
That's the autosens output, it looks very similar, there serious be something else when running the meal report
Regarding the TI Stick problem: tried literally everything to get my parts to work. Watched that video, went to forums. The cc debugger light just didn't want to turn green. Thankfully after a brief chat session with Digi-Key I'm getting another debugger and TI stick mailed overnight for free. What service...
Digi-Key is :+1: in my book.
@logichammer https://diyps.org/2014/05/29/determining-your-carbohydrate-absorption-rate-diyps-lessons-learned/ has some good info on determining carbs per hour
I'm somewhere in the 35-40 range
@ryanchen26 great Service!
@ryanchen26 let me know if you need any help. i did get one bum TI stick early on which TI credited me back for when I asked for a return.
@dm61 Thanks! working through that now
I am on the dexcom g5 -- docs mention that a work around is needed... ?
are you using the G5 receiver or looking to do it wirelessly?
you can send your G5 data to nightscout and then pull it from nightscout (NS) to openaps. Thats what I did and it works great as long as your loop is online
Cool -- yeah, I've been using both my phone and the receiver with g5, I'm using the night scout bridge already, cool -- I was just going through setup and checking connection and wasn't sure if I needed to attach the g5 receiver too (which didn't seem to be working correctly anyway, haha) thanks!
Thanks @Entropy512 ! Checking it out...
@ryanchen26 your pr makes me think you need this page? https://github.com/openaps/docs/blob/master/docs/docs/walkthrough/phase-6/keeping-up-to-date.md
@scottleibrand, @bewest or anyone else: I am repeatedly seeing 0.325 being set on the pump as a neutral basal when 0.350 is the neutral basal. enacted.json is also not consistent as it says it is setting 0.35 u/hr, when actually it is setting 0.325. Any clue what is going on here?
{
"bg": 77,
"temp": "absolute",
"snoozeBG": 104,
"recieved": true,
"predBGs": {
"IOB": [
77,
77,
78,
79,
80,
81,
83,
84,
86,
87,
89,
90,
91,
92,
94,
95,
96,
97,
98,
99,
100,
101,
102,
103,
103,
104,
105,
105,
105,
106,
106,
106,
106,
107
]
},
"rate": 0.325,
"reason": "COB: 0, Dev: 9, BGI: -1.19, ISF: 95, Target: 110; BG 77<80, delta 1>0; setting current basal of 0.35 as temp. Setting neutral temp basal of 0.35U/hr",
"COB": 0,
"eventualBG": 104,
"timestamp": "2016-06-10T23:17:42.725931",
"duration": 30,
"tick": "+1",
"IOB": -0.189
}
enacted.json (END)
We're requesting .35, but the closest the pump is capable of delivering for 30 minutes is .325.
It can probably handle a .35 over the course of a full hour, but since temps are 30m that's the resolution we get.
Is that an anomaly of a dose that low or does it happen with 0.45 as well for example?
it's not much of an issue, because we will see the slightly negative IOB and be more likely to high-temp later.
Assume IOB gets calculated correctly in this instance?
It will occur at any basal that isn't a multiple of what the pump can do.
Yes it should.
got it thanks
Different question, trying to set up a second edison and banging my head against this error code with the networking set up:
root@ubilinux:/# ifup wlan0
wpa_supplicant: /sbin/wpa_supplicant daemon failed to start
run-parts: /etc/network/if-pre-up.d/wpasupplicant exited with return code 1
by the way that was my wife that sent you that picture of Olivia
's sugar at 100 all night on openaps, she is thrilled!
@ddaniels1 that could be a little typo in the wpa_supplicant conf file
@jasoncalabrese thanks I missed a
} at the end of the file.
computers, they are so literal...
great, that happened to me while on vacation right after we first got things working, thought I broke everything
was just 1 extra char like that
Another question about the Edison. If the power dies in the middle of a loop, because for example the battery dies the git gets corrupted and the only way to fix it seems to be
oref0-reset-git any other way around that? the fix-git command doesn't work. Wondering if I should run that in cron once an hour but seems destructive.
the fix git command should call reset if it can't fix it
what errors do you see?
does
git status fail?
I had something happen a few days ago where
git status was ok, but invoking anything failed due to locks, and the error was getting lost in my preflight so it took a while to find
Don't have an example in the log handy. Something about a "loose file". Is there some bash scripting that will do that. I had just been running
oref0-fix-git-corruption in cron hourly. Then if it locks up manually running oref0-reset-git
next time you see it see if
git statusis ok, that's the check that is used in the fix script, but I think there might be some git corruption issues that don't cause git status to fail
OK thanks
How helpful, my RPi got powered off last night and I couldn’t remember
oref0-reset-git so I couldn’t loop last night
Reminds me I need to make a backup of this microSD card
Morning everyone. Have a question about my nightscout site on Azure - set it up (no data pump yet) but the menu will not pop out (where there are 3 lines). Can someone point me to where I should look, maybe I forgot a step (Don't think so) https://muffindad.azurewebsites.net/#
@danamlewis We closed the loop July 7th. Sent PR for names list, submitted requested form, parts for edison version are on the way....things are good!
+1 ! @loudestnoise I recently bought a 16G micro SD card. Where is the best place to look to see how to make a copy? I am such a noob.
@muffindad for just the website, there isn't a '/#' after, but have you tried a restart on the azure?
It's best to call oref0-reset-git in cron. It won't do anything unless git status is failing: if so, it will try oref0-fix-git-corruption first, then if that fails it will mv the .git directory to a backup location and re-init. That is non-destructive: current file state is unaffected, and history is preserved in a backup location for manual recovery if needed. Someone needs to update the docs to make that more clear.
Hi @flutelynne_twitter yes tried a restart. I'm wondering if it doesn't like the "free" plan... I don't add the "#" sign, when I hit the site it adds it.
@muffindad Have you tried asking on the Nightscout page? Either FB or another gitter room?
Just launched the tweetstorm about the #OpenAPS poster. first tweet (the rest are linked in this thread) is here with a link to the blog post I put on OpenAPS.org that has all same content as the poster
@danamlewis I just listened to the Juicebox podcasts that you did both taking about OAPS and your husband. Great story!
Dana, I'm REALLY hoping a pump manufacturer pays attention to your presentation!
We had some great conversations with several manufactures yesterday!
Thanks @dominickgallegos!
I'm finally considering making the switch to a pump, but switching to a pump and trying to immediately obtain a really old one is not something I like. :(
@bewest @scottleibrand and I (plus some other loopers) are here at 2016ADA, and having some great conversations with clinicians and industry
It's too bad Medtronic blocked remote control instead of just requiring local pump interaction when establishing a new pairing relationship - or does the TI protocol not have a concept of "pairing" and just a "target device address" for commands?
Amazing abstract at 2016ADA!
Thank you! Full content from the poster is here for anyone who wants to see it: http://bit.ly/1U1hYeX
So enjoyed reading that today, @danamlewis. Thanks!
Was also looking at the 670G poster, does anyone know if it’s targeting 150mg/dl or 120mg/dl? The pivotal trial description said 120mg/dl, but the graph made it look like it was 150.
Hey -- how much data does everyone use on a monthly basis... I'm just confirmed fully mobile, but wondering how much to up my data plan to account for using the hotspot for openAPS?
also, very cool about 2016ADA :)
@PazolesBrooke_twitter :)
I know the belt clips break easy on the Medtronic pumps. This morning I created a new one to replace the broken ones. I printed it and used my old spring and axle rod to fit it to the pump holder. It works well and I didn't even use ABS. Simple to print (no raft or supports required) and simple to install, although it requires a vice or other press to press the shaft in (hammering will break things!).
Link is here:
http://www.thingiverse.com/thing:1621502
Link is here:
http://www.thingiverse.com/thing:1621502
@danamlewis , great poster at #2016ADA!
Using the Carelink stick, but looking at the TI stick. Will the TI cc1111 work with a Medtronic 515 pump?
@dm61 :) thanks! And thanks goes to everyone who shared their data and experiences and helped with abstract prep :)
Evening all - trying to move to AMA and I am getting an error that I don't know how to debug:
> openaps report invoke monitor/meal.json
meal://text/shell/monitor/meal.json
/usr/local/lib/node_modules/oref0/bin/oref0-meal.js:71
if (typeof basalprofile_data[0].glucose != 'undefined') {
^
TypeError: Cannot read property 'glucose' of undefined
at Object.<anonymous> (/usr/local/lib/node_modules/oref0/bin/oref0-meal.js:71:36)
at Module._compile (module.js:456:26)
at Object.Module._extensions..js (module.js:474:10)
at Module.load (module.js:356:32)
at Function.Module._load (module.js:312:12)
at Function.Module.runMain (module.js:497:10)
at startup (node.js:119:16)
at node.js:906:3
reporting monitor/meal.jsonAny suggestions?