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

25th
Jun 2018
Scott Leibrand
@scottleibrand
Jun 25 2018 01:06
@juehv good to know about Li-ion voltage. We might need to add a preference for that to modify the shutdown voltage.
garykidd
@garykidd
Jun 25 2018 01:17
@mhaeberli can you let me know display specs so I can search amazon?
loztblaz
@loztblaz
Jun 25 2018 01:20
so this is not really a problem to solve, more of a wondering if it's been successfully done thing. I'm on a 670g currently with guardian 3 sensors, and I've got my Paradigm 723 + rig mostly set up. I've been uploading CGM/treatment data to nightscout with the 670g, and when I switch to OpenAPS in the near future it'll still be my only CGM source. Are there any glaring conflicts with this plan, or has anyone done something similar? Main concern is about having two pumps uploading data to the same nightscout, even with one on suspend only providing CGM data.
eventually i'd move to dexcom, just not able to out of pocket pay for them at the moment
Eric
@ecc1
Jun 25 2018 01:34
Should be possible. But I'd set a basal profile of zero units/hour all day, rather than keep the pump suspended.
Martin Haeberli
@mhaeberli
Jun 25 2018 05:04
@garykidd - nota bene - removing the thru hole solder on the 4 pin connectors has a risk of trashing the vias, and thus killing the Pi Explorer HAT board completely. if you feel comfortable with that risk, go for it! Part is "OLED LCD Display,LANMU 0.96" I2C IIC SPI Serial 128X64 White OLED LCD Shield Board OLED Module for Arduino 51 Msp420 Stim32 SCR” / https://www.amazon.com/gp/product/B01G6RE2DQ/ref=od_aui_detailpages00?ie=UTF8&psc=1 but make very sure that the pin order on the part you order and receive is the same as the pin order on the board you are repairing. Otherwise… Just so you know, the design specifies that jumpers are used to select where the vcc and ground pins are on the display - these look to be surface mount resistors selectively placed on the board.
Patcalfe
@Patcalfe
Jun 25 2018 05:24
Hi everyone, I am just getting started and have already run into trouble just getting jubilinux flashed to my edison. I doesn't seem to take. the command prompt window keeps saying "Using U-boot target: edison-blankrndis
Now waiting for dfu device 8087:0a99
Please plug and reboot the board
Dfu device not found Timeout
Did you plug and reboot your board?
If yes, please try a recovery by calling this script with the --recovery option"
and then it just reboots back to yocto.
any ideas? somthing im missing?
Patcalfe
@Patcalfe
Jun 25 2018 05:32
nevermind, found the rest of the troubleshooting in the docs.
thanks
Martin Haeberli
@mhaeberli
Jun 25 2018 05:48
:+1:
Jens Heuschkel
@juehv
Jun 25 2018 07:15
@mhaeberli I have the same display issue... Is this not a failure which should be replaced by the vendor?
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 07:18
I have the radio problem again. My rig keeps rebooting trying to fix it. Error message is again : do you need to run 'export rfspy_rtscts=0' first ? No more looping, no data pushed to NS.
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 07:28
@N3FM have you found a fix for this bug ?
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 07:54
I don't know if it's related but the only change I made before seeing these errors appear was to try to setup the BT tethering with my phone (unsuccessfully BTW). But it might just be a coincidence though.
Jens Heuschkel
@juehv
Jun 25 2018 08:12
@scottleibrand As far as I understood the Pi HAT manual, the hardware will shut down at 3.2V by the BMS. Thought this is a fixed value and not configurable by the pi ?!
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 08:26
Looking at the charts for the 18650 I use (HE4), I would not loose that much capacityif I shut down at 3.2V. https://lygte-info.dk/review/batteries2012/LG%2018650%20HE4%202500mAh%20%28Yellow%29%20UK.html
Around 5% I'd say (2250 vs 2375mAh)
Jens Heuschkel
@juehv
Jun 25 2018 11:14
@JohnDoeAkira but decent quality 18650 batteries reach capacity of nearly 4 Ah. Usually they have a huge drop at the beginning and a linear range between 4.05 to 2.8 (they can go down to 2.5 with a huge drop again). It really depends on the specific model but if you are unlucky, you "loose" 1/3 capacity.
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 11:24
Well, mine are also decent (and expensive:)) but they are focused on beeing able to deliver a huge peak current (for my skateboard). Less capacity but 20A. I agree that it's not very useful fur a rig ... but that's what I had at hand.
Jens Heuschkel
@juehv
Jun 25 2018 11:26
@JohnDoeAkira Haha same here with my huge lipo packs for multirotors :D
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 11:29
Yeah ... that problem with the "do you need to run 'export rfspy_rtscts=0'" appeared after I switched from Lipo to Liion but I suppose it's only a coincidence. I hope so because I prefer to walk around with Liion instead of Lipo (even if there is a Lipo in my minilink transmitter ... )
Jens Heuschkel
@juehv
Jun 25 2018 11:29
However ... the take away message is: If we want to use appropriate Li-Ion 18650 batteries we should be able to leverage this low voltage range
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 11:29
Understood :smile:
cameronrenwick
@cameronrenwick
Jun 25 2018 12:45
anyone have troubles with replacing batteries on a G5 and having it not pair properly with your iphone?
N3FM
@N3FM
Jun 25 2018 13:05
@JohnDoeAkira If I remember, you're running 0.7.0; when I try that, I keep getting the message that the CGM data was disrupted (or something like that) -- which means there is NO CGM data getting in. @juehv is working to get the CGM code into 0.7.0 as we speak. I did see the radio message when I installed 0.7.0 over 0.6.2, then tried to go back to 0.6.2 (it also showed out-of-memory when trying to load aliases during install.) I had to re-flash and reload.
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 13:28
@N3FM Thank you. Actually, I have to confess that I blindly followed the bootstrap script. I have no idea what version I am running ... I have to read the docs a bot deeper to understand how to change the version and recompile.
What I don't understand is that it worked fine ... and then began to fail. This morning I started it again (I ran out of battery during the night). It got some BG data from the pump ... and then suddenly stopped working again.
Scott Leibrand
@scottleibrand
Jun 25 2018 13:35
@juehv in addition to the Pi’s voltage shutoff we have our own graceful shutdown based on voltage.
Jens Heuschkel
@juehv
Jun 25 2018 13:37
@scottleibrand sure! But it doesn't help to put that down to 2.8v if the hardware stops at 3.2v. I figured the bms has a i2c interface to read the voltage. Maybe it's possible to reconfigure the shutdown voltage there?
N3FM
@N3FM
Jun 25 2018 13:42
@JohnDoeAkira If you went straight through, you're on 0.6.2 'Master' branch -- In that version, things will run fine, until a "checking pump clock" comes up. Once in a while, it retrieves an OLD clock value instead of the current value -- and then it tries to reset the clock in the pump, AND the CGM. The pump set_clock reports a failure, and the CGM set_clock induces a more significant failure. Once that happens, for me it's time to reboot, as it takes 1/2 hour or more to get past the error (clock, mmtune, old bg, etc). If you installed Papertrail, check the log and see if that is what happened. version information: When things are running correctly, the next-to last line in a loop says 'Your instance of oref0 [version, branch] is up-to-date.'
Jens Heuschkel
@juehv
Jun 25 2018 13:52
@N3FM @JohnDoeAkira I've added some code to reboot automatically when the loop doesn't succeed in a given time frame. It's part of 0.6.2. You can activate it within the ~/src/oref0/bin/oref0-pump-loop.sh file. Documentation how to use it is included.
garykidd
@garykidd
Jun 25 2018 14:32
@mhaeberli thx. My rig does work now, doesn't charge and display doesn't work. it may be worth the risk given its state (definitely worth it if I can make it work :smile:) .
Martin Haeberli
@mhaeberli
Jun 25 2018 14:34
Or ask support@ enhanced radio devices to swap it out. I know they have cheerfully done so for others
And / or order a spare THEN swap the broken one out
peterfulvi
@peterfulvi
Jun 25 2018 15:33
Im exited my parts are coming today for the PI-HAT. I notice the graph on the display screen. Does the rig actually get redings from you dexcom without the need for xdrip or NS? Im using Lookout right now.
Jon Cluck
@cluckj
Jun 25 2018 15:41
it can, you should check with the folks in xdrip-js for tips on how to make it work :)
Patcalfe
@Patcalfe
Jun 25 2018 15:59
hey everyone....having some trouble right off the bat here trying to flash jubilinux to my edison. the edison ID seems to be mismatched with the flashall.sh file, i have tried to change it to no avail. anyone know where i should look for more info? i am very new, just starting and most of this is greek to me.
Patcalfe
@Patcalfe
Jun 25 2018 16:09
the command promt says "dfu device not found"
Martin Haeberli
@mhaeberli
Jun 25 2018 16:36
@juehv - as to ‘replaced by vendor’ - I recommend you contact them nicely. I know that they have in other cases just replaced the failed Explorer HATs in good faith. I’m not taking a position on whether or not they have an obligation to do so...
Martin Haeberli
@mhaeberli
Jun 25 2018 17:10
with respect to the announcement about Dana RS applying for FDA approval and somehow having ‘openaps inside’ - is the assumption that such a pump, once FDA approved, could just work with an Edison or RPi based controller? and or Loop and an iPhone? Or is it something more subtle?
Jon Cluck
@cluckj
Jun 25 2018 17:10
it already works with androidAPS
Martin Haeberli
@mhaeberli
Jun 25 2018 17:11
cool! so then if / once approved here it would be an alternative to Medtronic ‘unicorns'
Jon Cluck
@cluckj
Jun 25 2018 17:11
it's extremely cool :)
Martin Haeberli
@mhaeberli
Jun 25 2018 17:53
and ‘asking for a friend’ where can we learn more about the cool new OpenAPS simulator @danamlewis mentioned over the weekend at ADA and in blog posts?
N3FM
@N3FM
Jun 25 2018 18:10
@juehv Slick!
N3FM
@N3FM
Jun 25 2018 18:23
@Patcalfe First, make sure you installed the Intel Edison device drivers; and reboot after installing. Next, make sure you UNPLUG, REPLUG, and REBOOT the Edison when flashall.bat asks for it. It took me maybe 5 passes for drivers, etc to all be happy and start the flashing process.
Scott Leibrand
@scottleibrand
Jun 25 2018 18:49
The Dana RS would be an “iPump” in the US, which means it could eventually work with any “iController” (like SOOIL’s locked-down version of AndroidAPS that they hope to get approval for). In the mean time, it’d work with the open source version of AndroidAPS (and presumably Loop) too.
Martin Haeberli
@mhaeberli
Jun 25 2018 18:50
and thus likely? OpenAPS controllers based on Edison/Explorer and/or RPi/Explorer HAT?
Scott Leibrand
@scottleibrand
Jun 25 2018 18:50
The simulator is in the simple-simulator branch of oref0. I think there is some additional one-liner magic I’m doing for the autotune backtesting that isn’t committed yet, but the basic manual operation works in that branch.
Martin Haeberli
@mhaeberli
Jun 25 2018 18:51
cool thx
Scott Leibrand
@scottleibrand
Jun 25 2018 18:51
I don’t see any point in using an OpenAPS rig if you can use an Android/Apple phone with Bluetooth and don’t have to do 900MHz
Martin Haeberli
@mhaeberli
Jun 25 2018 18:53
ok I see what you’re saying … so run Loop on iOS, or AndroidAPS on Android… avoid the 900mhz bridges of any kind
Scott Leibrand
@scottleibrand
Jun 25 2018 18:55
Yep. That’s what they do today in the rest of the world with AndroidAPS and the Dana R and Roche Combo.
Eric
@ecc1
Jun 25 2018 19:41
Agreed. Despite the effort I've invested in software and hardware for a "rig", I'll be only too glad to ditch it for a phone-only solution
cameronrenwick
@cameronrenwick
Jun 25 2018 19:47
@ecc1 as much as I too would like to shed some of the components of T1DM management (okay.. I'd just like to be a nonT1DM but I digress) having some sort of "thing" between my phone and pump gives me a weird sense of security. My phone, and lots of peoples for that matter, can act up. I kinda like using it for viewing into my pump/rig as a spectator. Dunno how others feel but having my phone be ultimate controller of my pump gives me a bit of a sketchy feel
JohnDoeAkira
@JohnDoeAkira
Jun 25 2018 20:34
@N3FM Regarding my openaps version that's working a bit erratically, I confirm it is 0.6.2. The only changes it did were : add a wifi network, change the battery from lipo to Li-ion and try to add bluetooth tethering. It works now since noon (and a manual restart).
Scott Leibrand
@scottleibrand
Jun 25 2018 20:35
@cameronrenwick you can always use a dedicated phone for your APS controller that doesn’t run any other apps etc.
Dana Lewis
@danamlewis
Jun 25 2018 21:02
@mhaeberli more coming soon on simulator (docs, etc.) we scrambled to get it up and running w/ good data to present saturday. about to present again, then need a breather for a few days..then we'll catch back up :)
nice @zgohr PR to the docs about PI speed (http://openaps.readthedocs.io/en/latest/docs/Customize-Iterate/usability-considerations.html#improving-the-battery-life-of-your-raspberry-pi) <--chart looks good in github but bonks in RTD, so I'll fix that later but merging now so people have access to the info
Pi Hat users, check that out ^ for longer battery life
achris2000
@achris2000
Jun 25 2018 21:13
Hi, during the update process, I receive following error :

npm install && sudo npm install -g && sudo npm link && sudo npm link oref0

npm WARN deprecated crypto@0.0.3: This package is no longer supported. It's now a built-in Node module. If you've depended on crypto, you should switch to the one that's built-in.
Killed

npm ERR! oref0@0.7.0-dev global-install: npm install && sudo npm install -g && sudo npm link && sudo npm link oref0
npm ERR! Exit status 137
npm ERR!
npm ERR! Failed at the oref0@0.7.0-dev global-install script.
npm ERR! This is most likely a problem with the oref0 package,
npm ERR! not with npm itself.
npm ERR! Tell the author that this fails on your system:
npm ERR! npm install && sudo npm install -g && sudo npm link && sudo npm link oref0
npm ERR! You can get their info via:
npm ERR! npm owner ls oref0
npm ERR! There is likely additional logging output above.
npm ERR! System Linux 3.10.98-poky-edison
npm ERR! command "/usr/bin/nodejs" "/usr/bin/npm" "run" "global-install"
npm ERR! cwd /root/src/oref0
npm ERR! node -v v0.10.29
npm ERR! npm -v 1.4.21
npm ERR! code ELIFECYCLE
npm WARN This failure might be due to the use of legacy binary "node"
npm WARN For further explanations, please read
/usr/share/doc/nodejs/README.Debian

npm ERR!
npm ERR! Additional logging details can be found in:
npm ERR! /root/src/oref0/npm-debug.log
npm ERR! not ok code 0

any idea what that is ?
Martin Haeberli
@mhaeberli
Jun 25 2018 21:16
@danamlewis thx re simulator update
Dana Lewis
@danamlewis
Jun 25 2018 21:18
:+1:
N3FM
@N3FM
Jun 25 2018 21:24
@juehv I created env vars in /etc/profile and confirmed they are active in the accounts after reboot - yet they don't seem to be triggering the functions in oref0-pump-loop... suggestions?
@achris2000 Did you install Jubilinux? I see "System Linux 3.10.98-poky-edison" in your error output... I don't think it will run on poky.
achris2000
@achris2000
Jun 25 2018 21:53
@N3FM I am reflashing it already. Thanks
joe5165
@joe5165
Jun 25 2018 22:23
I have installed 0.7.0-dev on an edison/explorer. I ran setup and had it install the precompiled Go. Is that correct?