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

14th
Oct 2016
Scott Leibrand
@scottleibrand
Oct 14 2016 00:00
yeah, lots of academic research has been done. shouldn't be too hard to figure out what population (and how many samples) you'd need to get a much better estimate than a PWD can do for themself
garykidd
@garykidd
Oct 14 2016 00:05
just an idea...there's always a scale and nutrition.com
but I do have a hunch that it might actually work...i could see a handy app...with some sort of incentive for providing guesses on carbs for meals.
Tim Howard
@tghoward
Oct 14 2016 00:07
@scottleibrand just tried the oref0 setup script (referred to here, but in oref0) http://openaps.readthedocs.io/en/latest/docs/walkthrough/phase-2/oref0-setup.html
love it! But right now it bailed for me when trying to add the openxshareble vendor outside of the openaps env
line 275
so there needs to be a cd in there. I'll see if I can figure out the addition.
Scott Leibrand
@scottleibrand
Oct 14 2016 00:11
I think I've fixed that in the oref0-setup branch
but I got hung up on whether I need to disable looping on raw data by default (which I think I do) before merging that to dev
Tim Howard
@tghoward
Oct 14 2016 00:15
oh, ok. I'll checkout the branch and try. Thanks. Oh - another thing to think about. On shareble, the interactive steps ask about the G4 serial, but not the MAC address, which is important for me (and will be for some others). I can add manually, but thought it might be an option in the interactive steps.
Scott Leibrand
@scottleibrand
Oct 14 2016 00:18
yeah, don't bother using openxshareble on dev. all that stuff is fixed in oref0-setup
but be aware that you're gonna get a config that loops based on raw data (if it's below 150 mg/dL or thereabouts)
Tim Howard
@tghoward
Oct 14 2016 00:19
Got it. Going in!
garykidd
@garykidd
Oct 14 2016 00:19
@scottleibrand something weird definitely going on with nano in the PM i just sent you...no?
Tim Howard
@tghoward
Oct 14 2016 00:41
hmm, seems to getting hung up on Checking oref0 installation
Scott Leibrand
@scottleibrand
Oct 14 2016 00:43
run it again, and if it still fails, install the latest oref0 manually with npm run global-install in ~/src/oref0 (after a git pull of the latest oref0-setup branch)
garykidd
@garykidd
Oct 14 2016 00:44
How do you scroll up in screen terminal on edison?
after initial boot...I noticed a couple things FAILED
on startup
for one....Failed to start Hostname Service
This message was deleted
Scott Leibrand
@scottleibrand
Oct 14 2016 00:52
Ctrl-A Esc and then arrow keys.
Assuming you're using screen.
garykidd
@garykidd
Oct 14 2016 00:53
yup...thx
interfaces(5) file used by ifup(8) and ifdown(8)
auto lo iface lo inet loopback

auto usb0 iface usb0 inet static
    address 192.168.2.15
    netmask 255.255.255.0

auto wlan0
iface wlan0 inet dhcp
    wpa-conf /etc/wpa_supplicant/wpa_supplicant.conf


    # For WEP wireless-essid Emutex wireless-mode Managed wireless-key
    #s:password
# And the following 4 lines are for when using hostapd... auto wlan0
#iface wlan0 inet static
#    address 192.168.42.1 netmask 255.255.255.0

~
~
~
~
~
"/etc/network/interfaces" 18L, 489C                           1,1           All
look ok?
Scott Leibrand
@scottleibrand
Oct 14 2016 00:58
Ya
I now suspect the auto wlan0 iface wlan0 stuff would work as well. It was probably the quote mark in your SSID that was screwing it up.
garykidd
@garykidd
Oct 14 2016 01:01
you think this is a problem...
[  OK  ] Started udev Kernel Device Manager.
[FAILED] Failed to start Load Kernel Modules.
David
@davidkeddydb
Oct 14 2016 01:01
@scottleibrand and everyone who contributed to the new set up, thanks. Just set up new TI stick on PI to get rid of carelink USB. Took under 2 minutes to get it set up. SWEET.
Tim Howard
@tghoward
Oct 14 2016 01:02
drat
importing cgm-loop file
cat: /root/src/oref0/lib/oref0-setup/cgm-loop.json: No such file or directory
@scottleibrand got a lot closer! ^^
Scott Leibrand
@scottleibrand
Oct 14 2016 01:11
Maybe you didn't git checkout oref0-setup && git pull?
garykidd
@garykidd
Oct 14 2016 01:27
damn...no wifi access still
Could the 'Failed to Start Hostname service' be an issue?
garykidd
@garykidd
Oct 14 2016 01:33
wpa_supplicant.conf...
ctrl_interface=DIR=/var/run/wpa_supplicant GROUP=netdev

network={
        ssid="########"
        psk="#########"
}
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
~
"wpa_supplicant.conf" 6L, 110C                                1,1           All
Scott Leibrand
@scottleibrand
Oct 14 2016 02:08
dunno, maybe
you could try re-flashing it I guess
garykidd
@garykidd
Oct 14 2016 02:33
thinking of windows setup got my old AMD up and connected now
garykidd
@garykidd
Oct 14 2016 02:41
Nahh back on mac
Last login: Thu Oct 13 22:37:04 on ttys000
iMac:~ g$ cd /Users/g/Desktop/jubilinux\ 2 
iMac:jubilinux 2 g$ sudo ./flashall.sh
Password:
Error: Running Homebrew as root is extremely dangerous. As Homebrew does not
drop privileges on installation you are giving all build scripts full access
to your system. As a result of the macOS sandbox not handling the root user
correctly HOMEBREW_NO_SANDBOX has been set so the sandbox will not be used. If
we have not merged a pull request to add privilege dropping by November 1st
2016 running Homebrew as root will be disabled. No Homebrew maintainers plan
to work on this functionality.
Using U-Boot target: edison-blankcdc
Now waiting for dfu device 8087:0a99
Please plug and reboot the board
last time I just did it anyway...should I be concerned here?
garykidd
@garykidd
Oct 14 2016 02:57
anyway proceeded and got to this...
Welcome to rescue mode! Type "systemctl default" or ^D to enter default mode.
Type "journalctl -xb" to view system logs. Type "systemctl reboot" to reboot.
Give root password for maintenance
(or type Control-D to continue):
thoughts?
last time I did get same output I just hit Control-D
Scott Leibrand
@scottleibrand
Oct 14 2016 03:17
Yeah, I always do Ctrl-D
Why do you keep doing sudo instead of running it without?
I doubt it matters though.
but I'm wondering if that was a remnant from setting up a pi.
thoughts on....
[  OK  ] Started Journal Service.
[  OK  ] Started udev Coldplug all Devices.
[  OK  ] Started Load/Save Random Seed.
[  OK  ] Started Create Static Device Nodes in /dev.
[FAILED] Failed to start Load Kernel Modules.
See 'systemctl status systemd-modules-load.service' for details.
and.....
[  OK  ] Started Serial Getty on ttyMFD2.
[  OK  ] Reached target Login Prompts.
[  OK  ] Reached target Multi-User System.
         Starting Update UTMP about System Runlevel Changes...
[FAILED] Failed to start Hostname Service.
See 'systemctl status systemd-hostnamed.service' for details.
Scott Leibrand
@scottleibrand
Oct 14 2016 03:57
Dunno. This is not openaps related really, so Google might be warranted.
Terri Lyman
@tatigger
Oct 14 2016 05:05
Hi, I am running the loop all by itself, not attached to me, with no changes to oref0. I have a few questions regarding it's behavior. I thought the max temp basal was set to 9, and we have to change the max temp basal when we are ready and trust the system. Is that correct?
^^^oops I mean0
not 9
Terri Lyman
@tatigger
Oct 14 2016 05:14
@danamlewis ? Is that true regarding the loop
max temp rate?
Because my loop running alone just set a temp basal rate of 1.05u/hr when my basal rate is 0.45 units/hour and my bg was 113 with an angled up arrow for 2 tics, and my range is set to 140 to 160. Totally unwarranted. That temp rate ran for 3 tics - which I know is not a ton of insulin, but ????
Scott Leibrand
@scottleibrand
Oct 14 2016 05:24
It's not max basal that is zero. It's max IOB
If you've been low temped a long time and start rising with lots of negative IOB, it will high temp to get your IOB back to zero so you don't have a big rebound.
It was probably showing an eventualBG of 180+
Terri Lyman
@tatigger
Oct 14 2016 05:26
Ah, OK, that makes sense. Yes, it has been negative temping all day. At least it stayed connected all day. Thanks - that helps a lot.
Lots of dumb questions, sorry.
Scott Leibrand
@scottleibrand
Oct 14 2016 05:28
Not dumb at all. Exactly the kinds of things you should be second guessing until you understand why b
(" b" is iPhone typoese for ".") ;-)
fiberfan
@fiberfan
Oct 14 2016 06:25
I haven't had time to track down the carb issue but my informal testing by attempting to mirror what openaps was doing with my vibe showed it was working well enough I switched to the looping pump this afternoon. @tatigger
it took me a bit to grok how the ups and downs in basal rates where equaling such smooth blood sugars. I had 2 meals the normal post meal blood sugar spike and the 2nd one definately should have created a spike.
Should be 2 meals without the normal post meal blood sugar spike :sigh:
PieterGit
@PieterGit
Oct 14 2016 08:08
@bewest (or others): has the idea of sending triggers with oacids (https://github.com/openaps/oacids/blob/master/oacids/triggers/emit.py ) been postponed. i would think sending triggers (e.g. "fresh bloog glucose value") is more efficient than time based loops every minute
PieterGit
@PieterGit
Oct 14 2016 08:39
i am thinking of sending a signal to start the pump-loop when i get a new glucose value of the dex (by usb), similar to http://stackoverflow.com/questions/33830949/sending-trigger-to-my-python-script
PieterGit
@PieterGit
Oct 14 2016 09:16
I tested it on a raspberry pi zero. seems to work, but i think the rpi0 is too slow to be usefull
Sandra Keßler
@SandraK82
Oct 14 2016 09:17
@PieterGit you are welcome, glad that I could help. I have to look into upgrade stuff to get latest ns
@libxmike runs ns on pi0 and loop on another
PieterGit
@PieterGit
Oct 14 2016 09:18
some security remarks: a lot of stuff runs under root user, instead of the pi user, some stuff ended up in /root/src and other in /home/pi/src . I think everything can run in user space
Sandra Keßler
@SandraK82
Oct 14 2016 09:18
Maybe the latest release will reduce pressure as much less data will be transferred
PieterGit
@PieterGit
Oct 14 2016 09:19
i run nightscout on an upboard (Intel Atom, http://www.up-board.org/ ) , so that will beat the pi0 for sure :smile:
Sandra Keßler
@SandraK82
Oct 14 2016 09:20
In src there should be nothing. The only sources I pull are from gcm-remote-monitor into /home/pi/cgm-remote-monitor the daemon script is run by root as required (as far as I remember)
PieterGit
@PieterGit
Oct 14 2016 09:21
aha, than it's a oref0-setup.sh thing
nightscout does not have to run as root, can be run as normal users (port >1024)
Sandra Keßler
@SandraK82
Oct 14 2016 09:22
Then the daemon script needs an update to run as pi
Have to look up how to do it, not springing to my mind atm
PieterGit
@PieterGit
Oct 14 2016 09:24
i needed to add DISPLAY_UNITS=mmol in the my.env and to set the BASE_URL
Sandra Keßler
@SandraK82
Oct 14 2016 09:25
Base URL for what is this needed?
PieterGit
@PieterGit
Oct 14 2016 09:26

Required

MONGO_CONNECTION - Your mongo uri, for example: mongodb://sally:sallypass@ds099999.mongolab.com:99999/nightscout
DISPLAY_UNITS (mg/dl) - Choices: mg/dl and mmol. Setting to mmol puts the entire server into mmol mode by default, no further settings needed.
BASE_URL - Used for building links to your sites api, ie pushover callbacks, usually the URL of your Nightscout site you may want https instead of http

Sandra Keßler
@SandraK82
Oct 14 2016 09:26
Ah oki
PieterGit
@PieterGit
Oct 14 2016 09:26
as you could have read in the documentation :smile: https://github.com/nightscout/cgm-remote-monitor#required
Sandra Keßler
@SandraK82
Oct 14 2016 09:27
Tried a little bit and copied much together, missed some parts though, like the limits for high and low...
PieterGit
@PieterGit
Oct 14 2016 09:29
i used a custom script to start mongo, that way you don't have to encode spaces with %20, e.g.
#!/bin/sh
cd /home/pi/cgm-remote-monitor
export MONGO_CONNECTION="mongodb://sally:sallypass@ds099999.mongolab.com:99999/nightscout"
export DISPLAY_UNITS=mmol
...
npm start
last thing: your mongodb isn't secured for local users, was surprised by that. it's only accessible from localhost, but i think it's better to require a username/password
Sandra Keßler
@SandraK82
Oct 14 2016 09:32
It is package default. The script does nothing with the mongo setup besides from installing it
This might not be the best solution, however it is a compromise between complexity of the setup and security
Mike
@libxmike
Oct 14 2016 09:38
@PieterGit @SandraK82 I dont run openaps on Rpi Zero anymore, looks like carelink/usb problem, runs for about 6h then no more, I run openaps on Rpi 2 and NS/Mongodb on Zero, switched to Rpi 2 2h ago
PieterGit
@PieterGit
Oct 14 2016 09:38
don't take it as an offence, it's just that i noticed while setting it up.
Sandra Keßler
@SandraK82
Oct 14 2016 09:40
@PieterGit no offense taken. Any feedback is welcome. I see the script as wip to be improved by any feedback. If you want to, I would be happy to include a patch for securing mongo
PieterGit
@PieterGit
Oct 14 2016 09:42
will work on it later and probably sent you some pull requests later (i keep fighting with git and github; i won't give up)
@libxmike : you tried the oref0-reset-usb.sh script ? (i really prefer the ti stick, because it's much better and supports multiple rigs / wait-for-silence)
Mike
@libxmike
Oct 14 2016 09:46
@PieterGit I did not try oref0-reset-usb.sh script , maybe will try, can you point me to it? I wait for the explorer board from Dana/Scott it has everything you need on board
PieterGit
@PieterGit
Oct 14 2016 09:48
yeah, @danamlewis and @scottleibrand also keep me waiting for the explorer board :wink: # WeAreWaitingForExplorerBoard
Matthias
@ceben80
Oct 14 2016 10:11
@/all Q to the new oref0-setup: I'd like to ns upload mmtuneand EdisonVoltageinfo.
Anybody knows how to.... :question:
Thanks in advance!
PieterGit
@PieterGit
Oct 14 2016 10:17
@SandraK82 i forked your repository and am working on a pull request, will give it another run and also add the oref0-setup.sh stuff to it, and make it possible to use the pi as a wifi hotspot (requires hupiwi or wifi usb adapter)
C-Ville
@C-Ville
Oct 14 2016 10:21
@PieterGit cool, I will check on your skript and see in I can scavenge some hints for the edison. I am contemplating to make my Edison a hotspot rather than my cellphone
PieterGit
@PieterGit
Oct 14 2016 10:24
@C-Ville : it's quite simple, see http://elinux.org/RPI-Wireless-Hotspot
Mike
@libxmike
Oct 14 2016 10:25
@PieterGit thx for the oref0 reset, this may work if I include it in crontab and run mayer every hour, will test it with Rasp Zero after my test w/o reset on Rpi 2
C-Ville
@C-Ville
Oct 14 2016 10:25
@PieterGit thanks
PieterGit
@PieterGit
Oct 14 2016 10:27
@libxmike : i think the oref0-reset-usb.sh should also be integrated with the oref0-setup.sh script, it's currently doing a mmtune if stuff fails. perhaps you could give @scottleibrand a help to integrate it (for carelink and ti-stick, unfortuately the ti stick also sometimes hangs)
Mike
@libxmike
Oct 14 2016 12:56
ist it possible with openaps report invoke or any other command to set a bolus? (like with the 522/722 remote control)
Dana Lewis
@danamlewis
Oct 14 2016 12:58
Yes. But highly recommend not building those commands into your loop for safety reasons. And take caution if you use those outside of a loop to ensure you're delivering what you think you are, etc. and don't double deliver if you resend the command by accident, etc. (I.e. I wouldn't use those commands)
Mike
@libxmike
Oct 14 2016 13:01
not for loop, first want to test how it works with the pump, can you point me to doc?
Dana Lewis
@danamlewis
Oct 14 2016 13:01
openaps -h and work from there
Mike
@libxmike
Oct 14 2016 13:05
thx @danamlewis, found it
Tim Street
@tim2000s
Oct 14 2016 13:13
How does oref0 handle a pump change?
The state where the NS data is different from that pulled off the pump?
Dana Lewis
@danamlewis
Oct 14 2016 13:16
Not sure what you mean. We are calculating iob based on insulin delivered, not reservoir count
Tim Street
@tim2000s
Oct 14 2016 13:24
Yes, but when you change the pump for another one, it has a different history.
So the insulin delivered, according to the pump isn't what you've delivered.
Dana Lewis
@danamlewis
Oct 14 2016 13:29
oh. not a site change, a pump swap . it should be running a pump history from the new pump after you update the SN and run the next loop and would pull and recalculate from the new history.
NS might look confusing for a while, but the loop would be doing the right thing based on the right data source once it’s pulled a history from the new pump
Tim Street
@tim2000s
Oct 14 2016 13:31
Except that the history from the new pump isn't what you've done for the last 4 hours, e.g the bolus 2 hours ago , and various changes in basal rates.
Thinking of what happens when a pump breaks and you have to put a new one in. I assume the loop will cope based on sgv but you'll need to keep an eye on it. I suspect you should probably make a pump change first thing in the morning.
Dana Lewis
@danamlewis
Oct 14 2016 13:33
true. although I had a bug very early on where it didn’t track iob correctly (so it thought history was 0) and it did the right thing. I had more ups/downs, but still stayed within targets.
yea keeping an eye on it will be important, although since it’s looping also looking at actual changes in BGs, that will help some.
Dana Lewis
@danamlewis
Oct 14 2016 13:39
Just sent out The Email. (The one about ordering your explorer board, if you filled out our interest form :D). And for those who didn’t fill out a form and are interested now, here’s the link to check it out/order one: http://bit.ly/2e3fkew
Mike
@libxmike
Oct 14 2016 13:42
@danamlewis "There are no shipping methods available for your cart or destination." for Germany, I dont use PayPal, can I pay with CC?
Dana Lewis
@danamlewis
Oct 14 2016 13:43
Huh. Let me share that with Morgan (it’s his site). There should be a CC option
Also just heard from someone in Canada. so stay tuned - hopefully they will fix the international ordering options ASAP!
tazitoo
@tazitoo
Oct 14 2016 13:45
Just pushed the 'buy' button - WOOT!
Dana Lewis
@danamlewis
Oct 14 2016 13:46
:) :clap:
Sebastien Lussier
@SebastienLussier
Oct 14 2016 13:48
@danamlewis Thanks Dana ! :smile:
Dana Lewis
@danamlewis
Oct 14 2016 13:49
:+1: sorry for the difficulties, international friends! I tested with my own order, but obviously didn't test international.
tazitoo
@tazitoo
Oct 14 2016 13:49
@danamlewis - since this is a preorder, what's the anticipated timeline for shipping dates?
Dana Lewis
@danamlewis
Oct 14 2016 13:49
November!
tazitoo
@tazitoo
Oct 14 2016 13:49
(/me is shocked you did not take an international flight to personally test such an important launch)
Mike
@libxmike
Oct 14 2016 13:49
np, at least board is here, can you tell when they will be send?
tazitoo
@tazitoo
Oct 14 2016 13:49
:smile:
Dana Lewis
@danamlewis
Oct 14 2016 13:50
LOL @tazitoo . Next time ;)
@libxmike ETA is November
Mike
@libxmike
Oct 14 2016 13:50
:+1:
Tim Street
@tim2000s
Oct 14 2016 13:52
We aren't getting it in Europe it seems.... :worried:
Dana Lewis
@danamlewis
Oct 14 2016 13:52
You are getting it, after the west coast board guys wake up and see that the site needs fixing :)
that’s my bad for being excited on the east coast and sending out the link before they are awake
Mike
@libxmike
Oct 14 2016 13:54
ahh west cost, 9h differance to us, give the 1h more then wake them:)
sorry , west is 6h
east is 9h
Dana Lewis
@danamlewis
Oct 14 2016 13:54
Yea. I’m heading back there today if I can make it through the tropical storm ;) in Seattle. (that’s a real thing)
Tim Street
@tim2000s
Oct 14 2016 13:55
Got it shipped to my contact address in US. Saves on shipping!
Mike
@libxmike
Oct 14 2016 13:55
in Seattle ? Tropical Storm?
yeah, I wonder how much for shipping to EU
Dana Lewis
@danamlewis
Oct 14 2016 13:56
@tim2000s sorry about the hassle!
Andy Sharrow
@dramageek
Oct 14 2016 14:01
What was the final size of the board? Gonna try and check out the options for batteries.
Dana Lewis
@danamlewis
Oct 14 2016 14:03
“small” ;) I don’t remember offhand exact measurements, you can check and see if they’re in the repo: https://github.com/EnhancedRadioDevices/915MHzEdisonExplorer
Mike
@libxmike
Oct 14 2016 14:04
@dramageek from what I see about 3cm x 4cm/4.5cm
or what Dana said :)
Dana Lewis
@danamlewis
Oct 14 2016 14:06
(Check the ‘tiny’ pic, too, to see it next to chapstick)
And I think someone was giving me crap about US chapstick not being the same size as international ;) so take it with a grain of salt or use the pump for sizing ;)
tazitoo
@tazitoo
Oct 14 2016 14:07
someone should help you out with a custom case - that rubber band is so 90's
Dana Lewis
@danamlewis
Oct 14 2016 14:07
I have a soft case, but there’s others working on 3d printed hard cases if you want one for yours. My pancreas is beautiful just the way it is, thank you very much ;) :)
tazitoo
@tazitoo
Oct 14 2016 14:08
what size lipo are you using with it? ...and run time?
Dana Lewis
@danamlewis
Oct 14 2016 14:09
been testing with 2200mah..but haven’t done good enough testing to see if it does as well with the explorer rig as it did with my old edison rig (where it could get 20 hours with a full charge). I’ve had a couple of 12-14 hour days that the 2200mah lasted through, though :+1:
tazitoo
@tazitoo
Oct 14 2016 14:10
sounds like that's plenty sufficient - thanks!
David
@davidkeddydb
Oct 14 2016 14:10
@danamlewis putting the charger on the new explorer board, way sweet
Dana Lewis
@danamlewis
Oct 14 2016 14:11
@davidkeddydb :D this board is made with lots of :heart: and good feedback from the community and a sprinkle of magical unicorn dust.
David
@davidkeddydb
Oct 14 2016 14:12
@danamlewis All of this will also push the big 3 to get it right in their next iterations which is so cool.
Dana Lewis
@danamlewis
Oct 14 2016 14:14
On a plane - if you see anyone ask, please reiterate that we know there's a problem w international orders but will get that fixed and communicate out when that's done. (But the boards are being ordered already so that won't slow the process down, so no worries on that.) Thanks!
Jarred Yaw
@jyaw
Oct 14 2016 14:15
Thanks @danamlewis! Bought the board!
Dana Lewis
@danamlewis
Oct 14 2016 14:16
:+1: :clap: :confetti_ball: :)
Jason Wittmer
@jbwittmer
Oct 14 2016 14:17
Thanks @danamlewis. Ordered 3 (2 for me and 1 for a friend). Excited. Would love to see what people are coming up with for 3D printed cases
Dana Lewis
@danamlewis
Oct 14 2016 14:18
Ditto! :)
David
@davidkeddydb
Oct 14 2016 14:22
Anyone taking Victoza as a T1 and also using Openaps? If so, you have to do anything special to get the meal time boluses right? I am still working on getting my negative iob worked out by dropping sort or early morning basals but I am having to take a ton of insulin to keep my post prandial within normal numbers. Lowered my carb ratio from 1:7 to 1:6 and today as I saw the rise happening this morning about 1hr 15mins after eating, I just manually bolused a unit so I don't go to 200mg/dl. Thanks for any insight.
tazitoo
@tazitoo
Oct 14 2016 14:27
I'm sure it'll be simpler and smaller that than the case for my edison (sled: http://www.thingiverse.com/thing:1780996, case: http://www.thingiverse.com/thing:1827092, and lid: http://www.thingiverse.com/thing:1780944)
Paul Martin
@Spazholio
Oct 14 2016 14:29
I've been using this little guy for my TI+Edison rig: https://www.thingiverse.com/thing:1525618. It's incredibly compact.
kenstack
@kenstack
Oct 14 2016 14:30
@danamlewis great job to you and @scottleibrand and @ps2 and everyone else who worked on the board can't wait to try it
David
@davidkeddydb
Oct 14 2016 14:32
You all are still doing it way pretty than me. I just jammed mine in the box with some velcro tape. https://smile.amazon.com/gp/product/B0002BBQUA/ref=oh_aui_detailpage_o01_s00?ie=UTF8&psc=1
Tim Street
@tim2000s
Oct 14 2016 14:40
Am I right that you have to buy the Edison separately for the explorer board?
Dana Lewis
@danamlewis
Oct 14 2016 14:41
Yes. And battery separate.
David
@davidkeddydb
Oct 14 2016 14:43
@danamlewis Can the Dex be plugged directly into the explorer board?
Dana Lewis
@danamlewis
Oct 14 2016 14:43
Yes.
Tim Street
@tim2000s
Oct 14 2016 14:44
That's what I thought.
PieterGit
@PieterGit
Oct 14 2016 14:49
but to max the Dex work, you need an extra 5V battery, see EnhancedRadioDevices/915MHzEdisonExplorer#3
Tim Street
@tim2000s
Oct 14 2016 14:51
Or you change Loop's Dex reader into a BT broadcast and pick it up directly from an iPhone
David
@davidkeddydb
Oct 14 2016 14:52
@PieterGit Thanks for pointing that out.
Hannah Crabtree
@lollydaggle
Oct 14 2016 14:54
@davidkeddydb @danamlewis @PieterGit thanks, guys. I'm now sold on picking one up. I went back and reflashed my pi so I could have an offline mobile rig... Can't wait to downsize
Dana Lewis
@danamlewis
Oct 14 2016 14:55
:+1:
Mike
@libxmike
Oct 14 2016 14:58
@danamlewis do you know what is the range of the antenna? (Board to pump)
Dana Lewis
@danamlewis
Oct 14 2016 14:59
Similar to whatever you experience with a TI stick already.
Mike
@libxmike
Oct 14 2016 15:00
ok, thx
Matthias
@ceben80
Oct 14 2016 15:02
@scottleibrand both rigs, Edison and pi throw theese errors, any idea :question:
Don't know what to do....loop doesn't work :cry:
rm: cannot remove ‘enact/enacted.json’: No such file or directory
enact/enacted.json  raised  byte must be in range(0, 256)
pump://JSON/set_temp_basal/enact/enacted.json
Traceback (most recent call last):
ValueError: byte must be in range(0, 256)
grep: enact/enacted.json: No such file or directory
enact/enacted.json  raised  byte must be in range(0, 256)
pump://JSON/set_temp_basal/enact/enacted.json
Traceback (most recent call last):
ValueError: byte must be in range(0, 256)
cat: enact/enacted.json: No such file or directory
PieterGit
@PieterGit
Oct 14 2016 15:02
@danamlewis : ordering one now. noticed newsletter subscribe box https://enhanced-radio-devices.myshopify.com/account/addresses gives a bad_request Parameter Missing or Invalid: Required parameter missing or invalid: address"
Sandra Keßler
@SandraK82
Oct 14 2016 15:04
@ceben80 that's the one that bugs me too... just grap the pump and limit temp basal to 6,35/h until I check and release the fix
Matthias
@ceben80
Oct 14 2016 15:08
@SandraK82 @scottleibrand @C-Ville yeah....really a bug, reduced the limit, works...... :worried:
Sandra Keßler
@SandraK82
Oct 14 2016 15:10
@ceben80 i reinstalled OpenAPS 3 times because of it
Chris Oattes
@cjo20
Oct 14 2016 15:11
yeah, the basal rate is implemented as a multiple of 0.025u/hr, so 6.4u/hr is 256 of those chunks, which doesn't fit in to 1 byte
Sandra Keßler
@SandraK82
Oct 14 2016 15:11
Yeah I have prepared a fix, not yet tested
Chris Oattes
@cjo20
Oct 14 2016 15:12
I think decocare assumes the maximum value will be 1 byte, not sure how the pump actually represents it
Sandra Keßler
@SandraK82
Oct 14 2016 15:12
Loopkit can do 2 bytes
We discussed it here a few days ago
Chris Oattes
@cjo20
Oct 14 2016 15:13
2 bytes is up to 1638.4u/hr. I think that should cover most people :P
C-Ville
@C-Ville
Oct 14 2016 15:14
@cjo20 it is way faster than my 715 can deliver
Chris Oattes
@cjo20
Oct 14 2016 15:15
If someone needs 1600u/hr, I think they've got bigger problems than how fast their pump can deliver insulin
Tim Street
@tim2000s
Oct 14 2016 15:15
:+1:
Mike
@libxmike
Oct 14 2016 15:18
@cjo20 where to buy insulin in buckets :D or in 50L drums ? :)
Chris Oattes
@cjo20
Oct 14 2016 15:18
a 7-series pump would last approximately 12 minutes before needing to be refilled :P
Paul Martin
@Spazholio
Oct 14 2016 15:19
So who's working on a new 3D printed case for the Explorer board? Is it @eluhrs? Because that would be SUPER. =)
And - I have some issues where sometimes my rig loses wifi and needs manual intervention to get it working again. Is there a Pushover alert I can set that will ONLY tell me when NS hasn't received data from the rig after X minutes? Pushover tends to be REALLY noisy, so that's the only thing I want to set...
Matthias
@ceben80
Oct 14 2016 15:38
@cjo20 1638.4u/hr ......lol
Sandra Keßler
@SandraK82
Oct 14 2016 15:39
I will see to it this weekend and hopefully present a self tested patch
Matthias
@ceben80
Oct 14 2016 15:40
@SandraK82 highly appreciated :clap:
garykidd
@garykidd
Oct 14 2016 15:46
@Spazholio great idea on the pushover...My rig right now stops working after awhile...ususally witihn 12-24 hours.
C-Ville
@C-Ville
Oct 14 2016 15:47
dann muss der Dr. Bernstein aber schon fast aus selbiger Zeit stammen. :-) BZ zu hause messen ist ja nicht so neu oder
PieterGit
@PieterGit
Oct 14 2016 15:59
@SandraK82 : i created various improvements to the deployscript, see SandraK82/deploy-ns-local-raspi#3 . Can you please review/merge?
Sandra Keßler
@SandraK82
Oct 14 2016 16:07
I will later today, thank you @PieterGit
cococay
@cococay
Oct 14 2016 16:30

@danamlewis I'm putting in my order for a board.

I'm trying to figure out if the form is correct. The email and item list on the link say Free shipping in the US, but the form keeps showing $7 shipping, and does not seem to offer an option to change shipping method.

Am I doing something wrong, is it $7, or a glitch?

THANKS for making this available to us, we can't wait!!!

Dana Lewis
@danamlewis
Oct 14 2016 16:31
They may be in the middle of making the fix for international orders
So hang tight until I hear from them to confirm
Chris Oattes
@cjo20
Oct 14 2016 16:31
putting the price up for the £ probably, given the exchange rate :P
cococay
@cococay
Oct 14 2016 16:32
That makes sense - thanks.. I can wait till later to order, just wanted to get it done and now forget lol :)
Tim Street
@tim2000s
Oct 14 2016 16:32
You know the issue that people have been seeing with myopenaps.git filling up? How frequently is that occurring?
Mike
@libxmike
Oct 14 2016 16:34
@danamlewis will the international shipping price change or will stay same, do you know?
Dana Lewis
@danamlewis
Oct 14 2016 16:36
International shipping price will be quoted depending on where it is shipping to
Mike
@libxmike
Oct 14 2016 16:36
ok
Scott Leibrand
@scottleibrand
Oct 14 2016 16:37
@PieterGit I never implemented Ben's oacids stuff, because the every minute cron works great and has been plenty good enough, and I didn't want the additional complexity yet.
garykidd
@garykidd
Oct 14 2016 16:39
@danamlewis so no free us shipping then?
Tim Street
@tim2000s
Oct 14 2016 16:39
US shipping is definitely free, unless they've just changed it...
cococay
@cococay
Oct 14 2016 16:42
From what Dana said above, sounds like a Glitch at the moment.
garykidd
@garykidd
Oct 14 2016 16:42
I got charged $7 for US shipping
taylorfowler
@taylorfowler
Oct 14 2016 16:43
@garykidd same here
Dana Lewis
@danamlewis
Oct 14 2016 16:46
Stay tuned. We’ll sort out the shipping costs/shipping abilities with the Hamshield team and report back on what the solution/fixes/clarify shipping prices.
PieterGit
@PieterGit
Oct 14 2016 16:47
@scottleibrand : ok, thanks for the answer.are there any other triggers than:
  1. fresh-blood-glucose triggers pump-loop
  2. fresh blood glucose triggers ns-upload
  3. pump-loop triggers ns-upload
taylorfowler
@taylorfowler
Oct 14 2016 16:47
No biggie @danamlewis i was expecting $90, soooo still a net gain :)
Dana Lewis
@danamlewis
Oct 14 2016 16:47
haha. all about setting expectations I guess!
Dana Lewis
@danamlewis
Oct 14 2016 16:57
@cococay see PM
Dana Lewis
@danamlewis
Oct 14 2016 17:10
Anyone want to volunteer to test with a US order, if you’ve been holding off? (Asking for one person, so in case it’s not fixed, everyone doesn’t end up w/ same problem). Should be fixed, but want to confirm before we go from yellow light to green light on orders again.
Ditto for an international tester if you want to raise your hand here.
Christer Jensen
@christerjensen
Oct 14 2016 17:12
Just ordered one to Norway - all fine now - used Paypal.
Paul Martin
@Spazholio
Oct 14 2016 17:13
If anyone wants to try out a US order, you're more than welcome to use my address to test it out. ;)
Dana Lewis
@danamlewis
Oct 14 2016 17:13
hip hip hooray! thanks for confirming @christerjensen.
David
@david-t1
Oct 14 2016 17:13
This message was deleted
Dana Lewis
@danamlewis
Oct 14 2016 17:13
Green light for international orders!
David
@david-t1
Oct 14 2016 17:13
I ordered within US using paypal just fine. No shipping.
Dana Lewis
@danamlewis
Oct 14 2016 17:15
@david-t1 just now?
David
@david-t1
Oct 14 2016 17:15
sorry now, a few hours ago
Dana Lewis
@danamlewis
Oct 14 2016 17:15
ok
need someone to confirm now-ish since international has been fixed.
Mike
@libxmike
Oct 14 2016 17:18
international seams to work, but did not finished my order yet
David
@david-t1
Oct 14 2016 17:24
Planning ahead for when Explorer ships, if I use openxshareble to input dex readings, is there any way to visualize openaps on my phone? It doesn't necessarily have to be cloud-based like Nightscout. Looks like HAPP would work except closed looping uses Nightscout (?). I'd love an offline setup.
Dana Lewis
@danamlewis
Oct 14 2016 17:25
do you want visual graphs, or just access to what the loop is doing? and are you talking offline or online?
David
@david-t1
Oct 14 2016 17:25
just being able to see what openaps is doing would be fantastic
offline even better
hotspotting really jumps cost
PieterGit
@PieterGit
Oct 14 2016 17:26
i didn't get international to work. can't select a shipping method for the netherlands
David
@david-t1
Oct 14 2016 17:26
I will check it out. Tx!
Dana Lewis
@danamlewis
Oct 14 2016 17:27
(h/t to the great @mddub for that awesomeness)
PieterGit
@PieterGit
Oct 14 2016 17:27
if i press the "quick checkout" button i didn't get charged shipping, but it returns to the select shipping method page. something really wrong (or it's some kind of caching because the product was in my cart)
Dana Lewis
@danamlewis
Oct 14 2016 17:27
can you try a different browser?
PieterGit
@PieterGit
Oct 14 2016 17:28
will do later, but must go for now
diabeticgonewild
@diabeticgonewild
Oct 14 2016 17:31
Crap! My 522 pump broke! The pump doesn't detect the reservoir!
Luckily I have a backup 522. But still!
Jason Calabrese
@jasoncalabrese
Oct 14 2016 17:43
I think @ps2 any maybe others are looking for broken pumps
diabeticgonewild
@diabeticgonewild
Oct 14 2016 17:44
Yeah
LMK
I literally bought it new for $300 about 1.8 years ago. So kind of irritated. It should have lasted longer.
But since the winter holidays are coming up, there's likely going to be more people selling their old pumps on Craigslist
Somebody came up with a crazy fix for the issue apparently
But I haven't been getting High BGs like the other person who had this error
Mike
@libxmike
Oct 14 2016 17:51
@danamlewis the Explorer Bord has already Edison right?
Dana Lewis
@danamlewis
Oct 14 2016 17:52
No. Edison not included.
taylorfowler
@taylorfowler
Oct 14 2016 17:52
@libxmike Check out Jet.com
I got mine for ~50 bucks
with the mini breakout board
David
@david-t1
Oct 14 2016 17:53
@diabeticgonewild My 515 is 4 years old. Not fun wondering when the shoe will drop!
Mike
@libxmike
Oct 14 2016 17:54
@taylorfowler I'm in EU, need to buy local
taylorfowler
@taylorfowler
Oct 14 2016 17:57
Ah, Good luck. Maybe try robotshop?
woof, 48 euro.
diabeticgonewild
@diabeticgonewild
Oct 14 2016 18:00
Yeah it sucks. Now I need to look for another backup.
Mike
@libxmike
Oct 14 2016 18:02
yeah, it is not cheap
David
@david-t1
Oct 14 2016 18:03
well, If I ever need another pump, I look forward to 522 or higher because my 515 doesn't play with wireless bg meter nor Tidepool.
diabeticgonewild
@diabeticgonewild
Oct 14 2016 18:05
Yeah I shoot for 522s but I will take pretty much anything for a decent price. I don't care that much about wireless BG meter (freestyle lite strips are considered to be the most accurate) and I don't use tide pool often because I'm not a big fan of collecting data and analysing MY stuff in retrospect.
David
@david-t1
Oct 14 2016 18:06
wish new models were crackable. we need diyaps to keep the market innovating
diabeticgonewild
@diabeticgonewild
Oct 14 2016 18:07
I'm upgrading my t:slim for $800 when the opportunity comes about. I hope we can use a packet sniffer or something to figure it out.
David
@david-t1
Oct 14 2016 18:08
a new pump can allow 5-6 years easy. a used model is a crap shoot (as you just experienced)!
kenstack
@kenstack
Oct 14 2016 18:08
@danamlewis just ordered u shipping was free
us
diabeticgonewild
@diabeticgonewild
Oct 14 2016 18:09
I bought this pump BRAND NEW
I don't need to be lectured about the obvious, thanks.
Dana Lewis
@danamlewis
Oct 14 2016 18:10
@kenstack thanks for confirming!
Mike
@libxmike
Oct 14 2016 18:10
@danamlewis which Edison should I buy for the Explorer Board? there is more than one version.
Dana Lewis
@danamlewis
Oct 14 2016 18:11
Green light for US orders again for explorer board! For anyone who DID pay a $7 US shipping charge, stay tuned for info about a refund on that charge.
Some international orders are going through ok; some are still having problems. Not sure what the difference is, but still investigating..
Sandra Keßler
@SandraK82
Oct 14 2016 18:12
Maybe we are onto something for new pumps, at least accu check combo... maybe I need to ask sooner or later for a big community effort in terms of contributing CPU power for cracking some pins for the combo
C-Ville
@C-Ville
Oct 14 2016 18:13
@SandraK82 like SETI?
Sandra Keßler
@SandraK82
Oct 14 2016 18:16
Yeah something like that. Need to review some stuff before I can tell more
Scott Leibrand
@scottleibrand
Oct 14 2016 18:16
@libxmike I'm not aware of multiple Edisons: can you provide a link?
there are multiple breakout boards: you don't need one of those, as the Explorer board serves that purpose too
ok, then I'm looking for wrong Edison
Scott Leibrand
@scottleibrand
Oct 14 2016 18:19
no, I think these are right
there aren't any real differences I can see, just things like "wearable" certifications and internal vs. external antennas (for wifi and BT I guess)
all of them should work with the Explorer board
Mike
@libxmike
Oct 14 2016 18:21
ok, thx, its confusing wenn you see 8 different Edisons
Scott Leibrand
@scottleibrand
Oct 14 2016 18:21
ya, most sites don't play that game
Mike
@libxmike
Oct 14 2016 18:22
they have also different part numbers
PieterGit
@PieterGit
Oct 14 2016 18:33
@danamlewis : i spot the error. I would like to buy two. I get
International Explorer Shipping for $15 when I selected one, but no shipping if i select two
fiberfan
@fiberfan
Oct 14 2016 18:33
With almost 24 hours of looping I am really liking it. I need to fine tune some thngs but even without that fine tuning it is great! Kudos to all who have contributed to this wonderful advancement!
Dana Lewis
@danamlewis
Oct 14 2016 18:33
@fiberfan :confetti_ball:
Dana Lewis
@danamlewis
Oct 14 2016 18:39
@PieterGit investigating, stay tuned.
Mike
@libxmike
Oct 14 2016 18:48
There are 2 major versions, the EDI2.LPON.AL.S is Low Power On-Board Antenna and the EDI2.SPON.AL.S is Standard Power On-Board Antenna. The EDI2.LPON.AL.S is targeted for wearables where the Wi-Fi power is stepped down to avoid heat. So EDI2.LPON.AL.S may need less power? But I thing it is software controlled.
Scott Leibrand
@scottleibrand
Oct 14 2016 18:50
LPON sounds slightly better, if it actually does lower-power wifi with jubilinux
Mike
@libxmike
Oct 14 2016 18:54
looks like, I'll go with LPON, EDI2.LPON.AL.S
Mike
@libxmike
Oct 14 2016 19:01
but I thing the hardware is the same, WiFi power is firmware/Linux controled
Scott Leibrand
@scottleibrand
Oct 14 2016 19:01
yeah, that seems likely
David
@david-t1
Oct 14 2016 19:04
Can anyone confirm that an Edison setup using Explorer radio will still use mmtune? I ask because I have to use a frequency scanning scrip to get TI comms working with my 515 (unknown whether the issue is with TI stick or the 515 ). My current rpi2 rig + TI stick works so long as I run a frequency scanning scrip to get mmtune working.
Scott Leibrand
@scottleibrand
Oct 14 2016 19:14
not sure what you mean by all that, but yes, an oref0-setup loop on the Explorer board uses mmtune
David
@david-t1
Oct 14 2016 19:15
oh, good, that is all the confirmation I need
Mike
@libxmike
Oct 14 2016 19:41
here more info about Edison versions:
there are in fact two different versions, the EDI2.LP.xx SKUs that correspond to the Wearable description is a low power version, not the CPU, but the wireless radio has been limited to keep the thermal properties at a lower level.
For development and general tinkering, is suggested to use the regular one (EDI2.SP.xx) and if you are concerned about the device heating up, run the radio in bursts and keep a low duty cycle.
It's also the standard one that is bundled with the breakout boards.
Scott Leibrand
@scottleibrand
Oct 14 2016 19:44
that sounds like it's more of a hardware limitation
Mike
@libxmike
Oct 14 2016 19:47
yeah, it may, there is not much info or I did not found it
so for openaps / loop, EDI2.LPON (ON is for on board antenna) will be most suitable
Scott Leibrand
@scottleibrand
Oct 14 2016 19:49
yeah, seems like it
Sandra Keßler
@SandraK82
Oct 14 2016 20:19
thanks to @PieterGit a brand new and even betete version of the ns-local script is available for Testing and feedback: https://github.com/SandraK82/deploy-ns-local-raspi if you like, it does the setup for oref0 and preparations for setup-script too
diabeticgonewild
@diabeticgonewild
Oct 14 2016 20:52
MUST EAT! GOT THE MUNCHIES! CAN'T RESIST!!!! 😂
Matthias
@ceben80
Oct 14 2016 20:58
@SandraK82 will search the next pi.... to build my nightscout farm :smile:
Sandra Keßler
@SandraK82
Oct 14 2016 21:06
sadly, none of mine components distributors has valuepacks for raspi :/ ...had Hope for substantiel reductions if i order 100 or more :D
Matthias
@ceben80
Oct 14 2016 21:09
Yeah, new idea, pi nightscout cluster hosting @home
PieterGit
@PieterGit
Oct 14 2016 21:10
This message was deleted
garykidd
@garykidd
Oct 14 2016 21:19
This message was deleted
Hardwick1973
@Hardwick1973
Oct 14 2016 21:28
Does anyone have practical advice for where to locate a Carelink USB stick to give maximum coverage throughout the night - my radio link often gets broken and I am wondering if there is an optimal orientation/location for the USB stick?
Darrell Wright
@beached
Oct 14 2016 21:29
I know the TI sticks have amazing range if that is an option
Im accross floors/house with mine
Hardwick1973
@Hardwick1973
Oct 14 2016 21:33
@beached - thanks. Do you have a model number and what pump are you on?
Sandra Keßler
@SandraK82
Oct 14 2016 21:35
I have some kind of trouble with my pi stick... range is max 2m
Dana Lewis
@danamlewis
Oct 14 2016 21:39
@Hardwick1973 which way are you pointing the stick? Underside should be pointed toward the pump
Sandra Keßler
@SandraK82
Oct 14 2016 21:40
where should I start with debugging TI
garykidd
@garykidd
Oct 14 2016 21:41
does anybody know if this is normal or not when attempting to flash jubilinux to edison via imac terminal
[FAILED] Failed to start Load Kernel Modules.
and...
[FAILED] Failed to start Hostname Service.
everything else shows as [ OK ]
Hardwick1973
@Hardwick1973
Oct 14 2016 21:42
@danamlewis - I have the stick vertically orientated (pointing up) with the rear side facing the bed. I think you are suggesting having the stick flat, with the underside uppermost (to the ceiling) and pointing to the bed. Is that right?
PieterGit
@PieterGit
Oct 14 2016 21:48
@SandraK82 : did you try oref0-reset-usb? do you have a us or ww pump?
Darrell Wright
@beached
Oct 14 2016 21:49
@Hardwick1973 754CML
Sandra Keßler
@SandraK82
Oct 14 2016 21:53
@PieterGit tried that and best working is unplugging.. it is an ww pump
if im sitting arounf 50cm from the pump mmtune goes to around -60 db
PieterGit
@PieterGit
Oct 14 2016 21:54
did you patch mmtune, see openaps/oref0@522fc27
    echo -n "Patching mmtune.py for WW frequencies. " # temporary fix for https://github.com/oskarpearson/mmeowlink/issues/16
 +        sed -i "s/def __init__(self, link, pumpserial, radio_locale='US'):/def __init__(self, link, pumpserial, radio_locale='WW'):/g" ~/src/mmeowlink/mmeowlink/mmtune.py || die "Couldn't patch mmtune.py"
 +    fi
Sandra Keßler
@SandraK82
Oct 14 2016 21:56
i "patched" it manually
or else i would not be able to find the pump
PieterGit
@PieterGit
Oct 14 2016 21:57
Yeah, should have known that. I also found out that mmtune.py --subg_rfspy --radio_locale WW --port /dev/mmeowlink --serial 123456 sometimes does miracles (first stop your loops)
replace serial by your pump number of course
Sandra Keßler
@SandraK82
Oct 14 2016 21:58
openaps use pump mmtune does this if I'm not wrong
PieterGit
@PieterGit
Oct 14 2016 21:59
no, not exactly. i use the mmtune.py after i used the ti in the pi0 setup (where it doesn't work yet). Somehow the mmtune.py (not openaps use pump mmtune ) fixes my proble
Scott Leibrand
@scottleibrand
Oct 14 2016 21:59
@hardwick1973 :+1: to the optimal placement of the Carelink being in the spare electronics bin, with a TI stick (or now an Explorer board) talking to the pump. ;-)
PieterGit
@PieterGit
Oct 14 2016 22:00
@scottleibrand :smile:
@scottleibrand @SandraK82 : but there still are some issues, with the ti (i assume with ww pumps / mmtune). @oskarpearson : great news that you started on making my hack to use ww pumps with openaps to disappear, oskarpearson/mmeowlink#51 (i'll try to contribute when i'm able to)
Hardwick1973
@Hardwick1973
Oct 14 2016 22:07
@scottleibrand - nicely put. Do you have a model number for the TI Explorer board?
(No model number, this is the first place putting together a batch of them)
Hardwick1973
@Hardwick1973
Oct 14 2016 22:17
@danamlewis - what is your experience of using this board in terms of range and reliability? Also, how tricky is the set-up?
Darrell Wright
@beached
Oct 14 2016 22:18
So am I correct that openaps only controls the basal adjustments to acheive target glucose ranges but one still puts the bolus into the pump?
Dana Lewis
@danamlewis
Oct 14 2016 22:18
Range is amazing compared to carelink. Reliability is solid. Set up is biased by the fact that we've set up dozens of rigs in 2 years, but the new docs and set up scripts are going to make this rig setup significantly easier to old Pi+carelink+old docs approach, IMO
@beached most people still choose to meal bolus, yes, given the peak time of insulin we have available to us
Hardwick1973
@Hardwick1973
Oct 14 2016 22:20
@danamlewis - great, I'll place my order. When are the new scripts likely to be available?
Darrell Wright
@beached
Oct 14 2016 22:20
@danamlewis and the interface is the nightscout app/website ?
Dana Lewis
@danamlewis
Oct 14 2016 22:21
New scripts are available. Check out the docs as is today. Keep in mind explorer boards ship in November.
Darrell Wright
@beached
Oct 14 2016 22:21
@danamlewis plus urchin
Dana Lewis
@danamlewis
Oct 14 2016 22:21
@beached you can do carb entry however you like - pump or NS care portal depending on preference
Visualization is most common on NS + watch but you can also access the logs via your phone if your rig is online (via wifi or phones hotspot?
Darrell Wright
@beached
Oct 14 2016 22:24

@danamlewis ah ok. Yeah I am just curious right now. Im using loop but I could easily have this going either as a primary or a backup method.

How well does it react to straight carb entry without a bolus? does pre-entry really matter at that point or does that play into the carb absorbtion timeing anyways

Tim Howard
@tghoward
Oct 14 2016 22:27
@scottleibrand , continuing from yesterday... yes, I had refreshed the branch but just checked again. I tried starting fresh and it seems to bail at the same spot
importing cgm-loop file
cat: /root/src/oref0/lib/oref0-setup/cgm-loop.json: No such file or directory
Traceback (most recent call last):
  File "/usr/local/bin/openaps-import", line 6, in <module>
    exec(compile(open(__file__).read(), __file__, 'exec'))
  File "/home/edison/src/openaps/bin/openaps-import", line 89, in <module>
    app( )
  File "/home/edison/src/openaps/openaps/cli/__init__.py", line 51, in __call__
    self.run(self.args)
  File "/home/edison/src/openaps/bin/openaps-import", line 77, in run
    candidates = json.load(args.input)
  File "/usr/lib/python2.7/json/__init__.py", line 290, in load
    **kw)
  File "/usr/lib/python2.7/json/__init__.py", line 338, in loads
    return _default_decoder.decode(s)
  File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
    raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
Could not import cgm-loop.json
Here are the top two commits from git log on the Oref0-setup branch
commit f48318613d0a9635e0f8c181c4c3333d77e8f530
Merge: e716dcd c2b6dac
Author: Scott <scottleibrand@gmail.com>
Date:   Wed Oct 5 19:49:05 2016 -0700

    Merge branch 'oref0-setup' of github.com:openaps/oref0 into oref0-setup

commit e716dcd6e2662d8261b3530119e43e73ef50c307
Author: Scott <scottleibrand@gmail.com>
Date:   Wed Oct 5 19:47:44 2016 -0700

    add oref0-setup to package.json
Dana Lewis
@danamlewis
Oct 14 2016 22:30
@beached it matters a lot (regardless of loop type) if you have net negative iob going into a meal. So I use eating soon mode pretty regularly. Some people have been experimenting with unannounced meal testing and finding AMA does quite well keeping up, but YDMV (and my theory is it most varies when people have negative or 0 iob peaking when carbs hit the body)
But oref0 doesn't require entry of an absorption time - it looks at what's expected to happen and observes a minimum carb absorption rate and looks at deviation of what is actually happening given insulin and carbs and the rate of change.
Darrell Wright
@beached
Oct 14 2016 22:33
@danamlewis thanks :)
Tim Howard
@tghoward
Oct 14 2016 22:44
looks like it must be because I'm running with sudo but I can't replicate cat $HOME translating to `root\'
Sandra Keßler
@SandraK82
Oct 14 2016 22:49
blueReader campaign nearly completed: http://unendlichkeit.net/wordpress/?p=277&lang=en
Tim Howard
@tghoward
Oct 14 2016 23:22
@scottleibrand over that hump; submitted a PR
lukas-ondriga
@lukas-ondriga
Oct 14 2016 23:23
I cross my fingers for blueReader. I plan to use it for OpenAPS. It would be great to have a possibility to use it over bluetooth without connection to Nightscout.
C-Ville
@C-Ville
Oct 14 2016 23:32
@lukas-ondriga try to spread the news :-) there are only a few more people needed t o meet the funding goal
ig888ack
@ig888ack
Oct 14 2016 23:48
Trying to make sure I understand the required hardware with the new explorer board...Explorer, Edison, battery, cc debugger (for initial programming). Is that it? Do I need the Edison breakout board for initial setup or can that be done just with the Explorer board?
Darrell Wright
@beached
Oct 14 2016 23:48
Is the cc-debugger even needed?
Dana Lewis
@danamlewis
Oct 14 2016 23:48
You don't need cc debugger.
And no other boards needed.
ig888ack
@ig888ack
Oct 14 2016 23:49
Awesome! Thanks!
Dana Lewis
@danamlewis
Oct 14 2016 23:49
Keep in mind ship time is November, in case you want to build another rig between now and then with another board (and a debugger), will be easy to switch.
ig888ack
@ig888ack
Oct 14 2016 23:50
OK thanks!