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

20th
Nov 2015
John Males
@johnmales
Nov 20 2015 00:11
I wonder whether something like an edison with built in flash will be a bit more robust from a data corruption point of view than a rPi with an SD card
Oskar Pearson
@oskarpearson
Nov 20 2015 00:15
I’m almost certain it will be
Only one way to find out!
John Males
@johnmales
Nov 20 2015 00:16
Very true
Dana Lewis
@danamlewis
Nov 20 2015 00:16
Yep. Put 'em on a plane with me and we'll find out ;)
Oskar Pearson
@oskarpearson
Nov 20 2015 00:17
:)
:airplane:
Dana Lewis
@danamlewis
Nov 20 2015 00:18
Hehe nice.
Oskar Pearson
@oskarpearson
Nov 20 2015 00:19
Easy bit done:
edison@pan:~$ sudo -i
[sudo] password for edison:
root@pan:~# lsb_release -a
No LSB modules are available.
Distributor ID:    Debian
Description:    Debian GNU/Linux 8.2 (jessie)
Release:    8.2
Codename:    jessie
root@pan:~# openaps --version
openaps 0.0.7
root@pan:~#
Scott Leibrand
@scottleibrand
Nov 20 2015 00:39
:+1:
Oskar Pearson
@oskarpearson
Nov 20 2015 00:55
I can’t test fully now - Siobhan’s gone to bed and has openaps working there - but I can successfully send packets from the edison.
Dana Lewis
@danamlewis
Nov 20 2015 00:55
woot!
Oskar Pearson
@oskarpearson
Nov 20 2015 00:56
File 20-11-2015, 00 54 48.jpeg
Comparative size
The edison and battery pack on the right should last for about 3-4 hours
Dana Lewis
@danamlewis
Nov 20 2015 00:56
Nice, thanks for the comparison pic.
Oskar Pearson
@oskarpearson
Nov 20 2015 00:56
File 20-11-2015, 00 55 18.jpeg
Dana Lewis
@danamlewis
Nov 20 2015 00:57
Suggested shopping list?
Oskar Pearson
@oskarpearson
Nov 20 2015 00:58
I’m trying to find the right set of boards that it can plug into. The issue at the moment is that due to design flaws in the edison blocks it’s super super easy to short their “battery block” with the underlying board - which from my perspective is a non-starter. It could easily cause the battery to catch fire, as you hear about in the news!
so I’m trying to find a way to avoid using their battery pack.
Dana Lewis
@danamlewis
Nov 20 2015 00:59
ah
Oskar Pearson
@oskarpearson
Nov 20 2015 01:00
I’d like a few more days to experiment with talking to an a real pump and making sure everything is reliable
Also you’d need to map to USA sites… But the bits that are a starting point are:
Actually I think I’m going to use sparkfun urls if possible - since they are probably the best source and people can find the local versions if they need to
Oskar Pearson
@oskarpearson
Nov 20 2015 01:06
Intel Edison: https://www.sparkfun.com/products/13024
Sparkfun Block for Edison: https://www.sparkfun.com/products/13045
Hardware Pack for screwing things together (I say get two of these…) https://www.sparkfun.com/products/13187
USB OTG and Micro USB cables (for serial console and for talking to the hardware board)
On the wireless side, I’ve used CC1111EMK - http://www.ti.com/tool/cc1111emk868-915 and a http://www.ti.com/tool/cc-debugger
You could use a GoodFET instead of the cc-debugger (may be easier/cheaper to get) but I don’t have that
Alternatively, you could buy the official RFCAT hardware from http://int3.cc/products/rfcat - but I’ve not used them / one of these
Or for excessive range you could try https://greatscottgadgets.com/yardstickone/ (again - I’ve not used them / one of these - but I expect it’d work)
Buying the official rfcat dongle if you’re in the USA is almost certainly a better option than the TI and CC-debugger
The Edison has built in bluetooth and wifi, so you don’t need usb adapters
Oskar Pearson
@oskarpearson
Nov 20 2015 01:16
At this stage, I’d suggest not ordering the battery block https://www.sparkfun.com/products/13037 as tempting as it seems to be
(It turns it into a serious electronics-hobbyist-level thing - involving carefully cutting batteries off from pcb backplates and things like that… and the risk of crazy fires…)
I’d also suggest against https://www.sparkfun.com/products/13025 - since the sparkfun base block has a better form factor. the sparkfun base block also has the advantage that you can run power into the console port and a USB device off the OTG port. The original Intel one can’t do that - it requires you use the pins instead of the USB cable
I’m likely to buy one of these and test to see if there’s a way we can use it - https://www.sparkfun.com/products/13035
Scott Leibrand
@scottleibrand
Nov 20 2015 01:20
is there a combo pack you can get to get a full working Edison system, like with the CanaKit for Pi?
Oskar Pearson
@oskarpearson
Nov 20 2015 01:21
Also: I’d suggest not buying https://www.wirelessthings.net/srf-stick-868-915-mhz-easy-to-use-usb-radio - although it could supposedly run rfcat at one stage, I couldn’t get it to work
All the combo kits I’ve seen include things that I’ve recommended against - like the battery pack or similar
Scott Leibrand
@scottleibrand
Nov 20 2015 01:21
I'm leaning toward trying the YARD stick.
k
Oskar Pearson
@oskarpearson
Nov 20 2015 01:22
There’s this, for example: https://www.sparkfun.com/products/13276
But we don’t need the GPIO Block or battery pack. So it seems a bit of a waste
I’m tempted to buy the YARD stick too from a EU supplier
After I confirm that everything works a-ok from a software point of view
Then of course… we need a box for it :)
Anyway
that’s something for tomorrow - I’m off to bed
(I said that 1:30 ago to someone.. so much for that idea!)
Cheers all
Dana Lewis
@danamlewis
Nov 20 2015 01:26
:) thanks, night!
Scott Leibrand
@scottleibrand
Nov 20 2015 01:26
heh, later, thx!
Oskar Pearson
@oskarpearson
Nov 20 2015 01:26
I’m going to work the whole of Saturday on this stuff, so hopefully more news on Sat evening uk time
Scott Leibrand
@scottleibrand
Nov 20 2015 01:27
nice
Peter Miller
@ochenmiller
Nov 20 2015 03:30
@oskarpearson , have you tried set up the CC1111EMK to use the GPIO pins instead of USB? Might that allow for even lower power consumption on the edison?
FWIW I've now set up my RPi to talk to the wixel using serial rather than USB. Allows me to get it inside the case and save a few mA. Also considering using the wixel to xmit data to my uploader phone, since it's already listening for traffic from wixel. Again, way less power hungry than wifi. Holy grail is still edison though.
Pete Schwamb
@ps2
Nov 20 2015 03:40
I was pretty exited about the edison, but people have been expressing concerns about battery.
eyim
@eyim
Nov 20 2015 08:15

Hmm - after running smoothly for a few hours, now I am getting this error on every loop since 11:45 pm. Looks like it is find a problem when it tries to munge active-basal-profile.json or the list of temp basals. Any thoughts?

munge://JSON/normalize/monitor/mm-normalized.json
Traceback (most recent call last):
File "/usr/local/bin/openaps-report", line 6, in <module>
exec(compile(open(file).read(), file, 'exec'))
File "/home/pi/src/openaps/bin/openaps-report", line 82, in <module>
app( )
File "/home/pi/src/openaps/openaps/cli/init.py", line 44, in call
self.run(self.args)
File "/home/pi/src/openaps/bin/openaps-report", line 75, in run
output = app(args, self)
File "/home/pi/src/openaps/openaps/cli/subcommand.py", line 50, in call
return self.method.main(args, app)
File "/home/pi/src/openaps/openaps/reports/invoke.py", line 40, in main
output = task.method(args, app)
File "/home/pi/src/openaps/openaps/uses/use.py", line 45, in call
output = self.main(args, app)
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/init.py", line 297, in main
tool = NormalizeRecords(args, *kwargs)
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/historytools.py";, line 440, in init
self.add_history_event(event)
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/historytools.py";, line 451, in add_history_event
decoded = getattr(self, "decode{}".format(event["type"].lower()))(event)
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/historytools.py";, line 583, in _decode_tempbasal
{adjustment: event["amount"]}
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/historytools.py";, line 525, in _basal_adjustments_in_range
description=description
File "/usr/local/lib/python2.7/dist-packages/openapscontrib.mmhistorytools-0.1.1-py2.7.egg/openapscontrib/mmhistorytools/historytools.py";, line 510, in _basal_adjustments_in_range
assert (start_datetime < end_datetime)
AssertionError

Scott Leibrand
@scottleibrand
Nov 20 2015 08:58
I'm not familiar with the mmhistorytools code. Does your loop depend on that part working?
eyim
@eyim
Nov 20 2015 09:08
No it doesn't I removed it and it still works. I had seen it in others' implementation and thought that it might be necessary for later with some other tool so I added it in to be safe
Scott Leibrand
@scottleibrand
Nov 20 2015 09:17
It preps data for the openaps-predict visualizations.
eyim
@eyim
Nov 20 2015 09:19
Ahh - good to know. Thank you
Oskar Pearson
@oskarpearson
Nov 20 2015 12:47
@ps2 About the edison - any more details on the battery issues you’ve mentioned? I’m interested in what problems I’m likely to encounter
Milos Kozak
@MilosKozak
Nov 20 2015 21:50
@jasoncalabrese any advantage of using animas pump during day?
still thinking about direction where to go now ...
Milos Kozak
@MilosKozak
Nov 20 2015 22:03
:point_up: 18. listopad 2015 18:10 this you sentence may kick me forward
Pete Schwamb
@ps2
Nov 20 2015 22:04
@oskarpearson Oh, I had asked on the dexdrip channel about making submodules (916.5 and 2.4ghz) for edison, and someone commented that the edison uses too much power. They seemed down on the idea; I guess I am too easily swayed. :)