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

14th
Jan 2017
dougtakeuti
@dougtakeuti
Jan 14 2017 00:18
blob
I wonder if anyone using MDT Enlite sensors could help me figuring out what is wrong with uploading glucose data to NS. It was working fine until we decided to go offline for 10 days. When we resumed, it started to behave like this:
Several pills are uploading ok, but BG and the openaps pill are not showing. The loop seems to be running fine as far as I can see in /var/log/openaps/pump-loop.log. I checked all the glucose-related reports and they have correct data in the correct folders. The only exception obviously is cgm/ns-glucose.json as there is no data in NS. I updated oref0 and NS (using dev) and redeployed. In Admin tools there are no treatments in the future. I also redid my openaps instance using oref0-setup.
Tim Mecklem
@tmecklem
Jan 14 2017 00:33
I wonder if the code that tries to find ns gaps is confused. What happens if you do openaps first-upload?
Going from memory on the alias
dougtakeuti
@dougtakeuti
Jan 14 2017 00:39
@tmecklem Looks promising now! I see one BG point. Let's see what it does next.
It seems to work! Thanks! I had forgotten about openaps first-upload.
Now waiting to see if the openaps pill will show.
dougtakeuti
@dougtakeuti
Jan 14 2017 00:45
..And we're back!
blob
Tim Mecklem
@tmecklem
Jan 14 2017 01:15
Great!
jburke41
@jburke41
Jan 14 2017 02:49
I switched from android to iphone. I paired my raspberry pi with my new phone. However, on my Nightscout, OpenAPS has a status of unknown and states that the device is undefined. I can see my blood glucose from dexcom share and the raspberry pi is changing my temp basal rates. However, I cannot see the changes in Nightscout. Any advice?
Paul Martin
@Spazholio
Jan 14 2017 03:00
Ok, so let’s say hypothetically I got a buttload of Skin-Tac on my hands. What negates it? I’ve - again, hypothetically - tried water, lots of soap, Goo-Gone...
Chris Oattes
@cjo20
Jan 14 2017 03:05
mineral oil / baby oil
you probably also, hypothetically, want to keep it off your keyboard :P
jaylagorio
@jaylagorio
Jan 14 2017 03:49
@scottleibrand Thanks!
Chris Oattes
@cjo20
Jan 14 2017 05:04
Eugh, turns out the case I made doesn't work. It seems that because the power connection wires run parallel to the wifi antenna, it stops it working
Chris Oattes
@cjo20
Jan 14 2017 05:24
That may also be the cause of my pump comms issues. I'll have to investigate
Scott Leibrand
@scottleibrand
Jan 14 2017 06:02
@tmecklem good memory. :smile:
Sacha M
@coolestkidsever
Jan 14 2017 07:26
Hi all, I think I have all the parts needed and have just started working through the setup of openAPS. Which is going surprisingly well! But I have one questions at the end of the general setup, I expected some way of connecting the carelink to the pi, and wondered if I missed some instructions somewhere (have looked a few times but can't seem to find anything) thanks in advance!
Scott Leibrand
@scottleibrand
Jan 14 2017 07:35
"Plug it in" :stuck_out_tongue_winking_eye:
Sacha M
@coolestkidsever
Jan 14 2017 07:36
Haha, is that all? :smile:
Scott Leibrand
@scottleibrand
Jan 14 2017 07:36
Many people use a USB extension cable, but you can get started without it.
Sacha M
@coolestkidsever
Jan 14 2017 07:36
I have (btw), just felt like I was missing a step
Scott Leibrand
@scottleibrand
Jan 14 2017 07:37
That's why we still have the Carelink in the directions: it's plug and play. Range sucks, though, so you'll want to move on to mmeowlink and/or Explorer board soon.
Sacha M
@coolestkidsever
Jan 14 2017 07:37
Have set up the first loop but just realized I haven't been using the test pump or getting data on it... so now a waiting game.
Yeah I had done further reading in the mmeowlink, will look at that once I get this one past the gate
So, in simple terms, for simple me, the openAPS setup (oref) tells the pi to talk to the carelink (and the carelink already knows to talk to any pump in range?)
Dana Lewis
@danamlewis
Jan 14 2017 07:40
The setup scripts in phase 2 is where you'll tell the rig about your particular pump
Sacha M
@coolestkidsever
Jan 14 2017 07:40
Or just the pump serial number entered in the interactive setup?
Dana Lewis
@danamlewis
Jan 14 2017 07:40
So the carelink can read it
Yep
Sacha M
@coolestkidsever
Jan 14 2017 07:41
Sweet, just making sure I get a feel for what I'm doing! :smile: thanks!
Dana Lewis
@danamlewis
Jan 14 2017 07:42
Make sure the pump is out of prime screen, not suspended, etc (check out the subsequent troubleshooting page for a full list of still not working for you)
:+1:
Sacha M
@coolestkidsever
Jan 14 2017 07:46
So checking the log, it's currently saying error port not open, is that likely to be a setup error or just because I have no data in there yet?
Dana Lewis
@danamlewis
Jan 14 2017 07:48
"No data in there yet" - you mean on the pump, or in the system? Is your carelink plugged in, and did you say no to mmeowlink in setup script?
Sacha M
@coolestkidsever
Jan 14 2017 07:49
Sorry, in the pump, I've only just setup a test pump. Carelink is plugged in and I said no to mmeow
There's plenty of nightscout data, but only one pump sensor reading in the test pump
Dana Lewis
@danamlewis
Jan 14 2017 07:56
Ah. It may need some more pump records, but also may just need more time for it to get rolling.
Sacha M
@coolestkidsever
Jan 14 2017 07:57
Ok. Will try out my patience and see how I go!
Dana Lewis
@danamlewis
Jan 14 2017 07:57
:+1:
Matthias
@ceben80
Jan 14 2017 15:09
@/all explorer board loopers - any experiences about a 1000mAH lipo, how long will this battery last :question:
I want to make my new rig as tiny as possible, for me no problem charging during the day...
Tim Street
@tim2000s
Jan 14 2017 15:14
I've been getting roughly 18 hours with 2000mAh @ceben80
Matthias
@ceben80
Jan 14 2017 15:16
actually 8 hours should work @tim2000s with a 1000er lipo :wink:
Tim Street
@tim2000s
Jan 14 2017 15:21
Right then - running the Dev device again, and I see this behaviour.
glucose.json newer than temp_basal.json
Temp refreshed
Could not parse input data:  [TypeError: Cannot read property 'date' of undefined]
It looks like it is having problems reading the glucose data
Toby Canning
@TC2013
Jan 14 2017 16:36
Hi all, can someone tell me which port can be used for charging on the explorer board?
Matthias
@ceben80
Jan 14 2017 16:47
Hi Toby, I haven't yet connected a battery, but should be the OTG port :wink:
Toby Canning
@TC2013
Jan 14 2017 16:48
Nice
Thanks
Matthias
@ceben80
Jan 14 2017 16:50
@TC2013 last night was running my 1st overnight looping with the explorer board, connected via OTG to a power supply...didn't buy a lipo yet
Terri Lyman
@tatigger
Jan 14 2017 16:55
is there anyone available who can help me trouble shoot ifttt and This Button. I think I have googled and searched most things, and I can't figure out what part I am doing wrong. Thanks
@tc2013, no matter which port you plug in, the Explorer Board seems to charge.
Scott Leibrand
@scottleibrand
Jan 14 2017 16:58
@TC2013 @ceben80 either port on the Explorer board is fine for charging.
Matthias
@ceben80
Jan 14 2017 16:59
great Scott, I think I'll buy a 1000mAh Lipo, this rig must be tiny :smile:
Ryan anglani
@ryan_anglani_twitter
Jan 14 2017 17:21
@ejoe132 hello I entered the password and it worked!
Dana Lewis
@danamlewis
Jan 14 2017 17:31
@tatigger what’s up with IFTTT and This Button? :)
Chris Oattes
@cjo20
Jan 14 2017 17:36
Completely offtopic, but probably something a bunch of people here would find interestihng: SpaceX launch live video https://www.youtube.com/watch?v=7WimRhydggo (https://www.youtube.com/watch?v=tTmbSur4fcs for one with hosts talking about stuff)
Terri Lyman
@tatigger
Jan 14 2017 17:38
@danamlewis I get either "action skipped" if I activate from This Button or "maker error" if I check it within the maker app
Toby Canning
@TC2013
Jan 14 2017 17:51
@scottleibrand @danamlewis We've been running dev for a day now. Everything worked pretty much flawlessly, including bt pairing (very nice!). The output recommendation file for autotune is simply amazing. The recommended changes looked spot on from yesterday's activity. Can't wait to see the results after a week!
Scott Leibrand
@scottleibrand
Jan 14 2017 17:52
Thx @cjo20
@TC2013 :+1:
You can also run it retrospectively over recent weeks.
Toby Canning
@TC2013
Jan 14 2017 17:53
Awesome!
Ryan anglani
@ryan_anglani_twitter
Jan 14 2017 17:55
Hello all ! i am at this point in the process >Configure Bluetooth Low Energy tethering [optional does this mean i can connect the pi3 to the motoG that is blue toothed to the dexshare4 and this will give us a connection to the internet rater than relying on wifi ?
Dana Lewis
@danamlewis
Jan 14 2017 18:00
@TC2013 awesome!
@tatigger hmm, not sure what would cause that - have you tried to build a second recipe? Same error?
Terri Lyman
@tatigger
Jan 14 2017 18:02
That is a good idea :) Will try that.
Dana Lewis
@danamlewis
Jan 14 2017 18:16
:+1:
Hardwick2000
@Hardwick2000
Jan 14 2017 18:16
Is the WW locale still in the dev branch?
Dana Lewis
@danamlewis
Jan 14 2017 18:24
@Hardwick1973 dev but still have to add it manually - we need someone to fix it instead of doing the workaround....
Matthias
@ceben80
Jan 14 2017 18:24
@TC2013 :clap: well done
Tim Street
@tim2000s
Jan 14 2017 18:26
@danamlewis I think it's fixed in the dev branch - the pump locale is now set in pump.ini
Having said that, it doesn't talk to the pump properly, and I can't for the life of me understand why. The mmeowlink set-ups seem identical.
stefanakerblom
@stefanakerblom
Jan 14 2017 18:34
I am at the edge. Have all the gadgets but am hindered by the following: tail: raw = self.send_force_read( )
File "/usr/local/lib/python2.7/dist-packages/decocare/stick.py", line 606, in send_force_read
self.link.write(reader.format( ))
File "/usr/local/lib/python2.7/dist-packages/decocare/link.py", line 47, in write
r = self.serial.write( string )
File "/usr/local/lib/python2.7/dist-packages/serial/serialposix.py", line 517, in write
raise portNotOpenError
serial.serialutil.SerialException: Attempting to use a port that is not open
i followed the manual precisely 4 times. now the pi2 is set up without any "optionals"
I am very excited to take the last step. I have NS, Xdrip, PI2, Carelink, Minimed 722
ohbrandon
@ohbrandon
Jan 14 2017 18:42
Need a little help -- I keep getting the following exception
raise AlreadyInUseException("{port} already in use".format(port=port))
decocare.link.AlreadyInUseException: /dev/serial/by-id/usb-0a21_8001-if00-port0 already in use
woodtrevor
@woodtrevor
Jan 14 2017 18:52
@stefanakerblom Is the Carelink plugged in? This is the output from the loop running via cron, right? What happens if you stop cron, kill all of the openaps processes, and try to run one manually (e.g. openaps use pump model)?
stefanakerblom
@stefanakerblom
Jan 14 2017 18:54
yes, it is.
woodtrevor
@woodtrevor
Jan 14 2017 18:55
@ohbrandon Likewise, what happens if you kill all openaps processes and try to run one manually?
stefanakerblom
@stefanakerblom
Jan 14 2017 18:55
how to kill the processes?
i tried manually openaps-use pump model and get the same
woodtrevor
@woodtrevor
Jan 14 2017 18:59
If you look in your cron (crontab - l) there is a command that kills all openaps processes older than like 15 minutes. You would do the same but without the older-than flag. Sorry, don't remember the exact syntax off the top of my head. But you have to keep cron from starting new ones first so probably edit your crontab and comment all of those commands out.
Port not open probably means that decocare can't find the Carelink though. I've only seen that when Carelink isn't plugged in. What do you see when you run lsusb?
Matthias
@ceben80
Jan 14 2017 19:03
@stefanakerblom killall -g openaps
jaylagorio
@jaylagorio
Jan 14 2017 19:04
@jaysonewer @scottleibrand I hate to say it, but I can no longer recommend the 2.4 GHz antenna we were talking about for use with the Explorer board. As you mentioned, closing the can creates a faraday cage so signals are blocked. To fix this I got the antenna we were talking about, but I only got one. When I connected it to the Explorer board things worked well, but I couldn't close the can to test because Wi-Fi would immediately cut out. I now have two antennas, one for the Edison board and one for the Explorer board. With both antennas on the outside Wi-Fi and Bluetooth work solidly when the can is closed, as expected. Unfortunately the RF connection to the pump fails consistently. Opening the can fixes RF to the pump.
This leads me to believe that connecting the 2.4 GHz antenna didn't do anything, the Explorer board's built-in antenna was actually doing the work. The fact that it worked with the can opened on the other side of the room (while the CareLink sticks have trouble six feet away) is just one more data point about how truly bad Medtronic designed the sticks but also shows how incredably well the Explorer board is designed. Since there doesn't appear to be a 900 MHz antenna with a uFL connector on Adafruit it looks like I'm on the look out for a hard plastic container of similar size and/or that kind of antenna from another source.
Matthias
@ceben80
Jan 14 2017 19:04

@stefanakerblom then
first as written below the see wheter the Carelink is properly connected lsusb

cd ~/myopenaps
openaps use pump model

Eric
@ecc1
Jan 14 2017 19:07
@jaylagorio What kind of antenna were you experimenting with? (Sorry, I can't find anything via the search on Gitter.) There are lots of uFL-to-SMA pigtails available, which opens up a lot of (more bulky) possibities.
I was initially skeptical that it was working properly for an Edison board. I guess I was right to be.
But the idea is to have everything fit in an Altoids sized container, or something else of harder toughness than plastic
A uFL-to-SMA connector might take up more room than I'm working with.
stefanakerblom
@stefanakerblom
Jan 14 2017 19:18
pi@raspberrypi:~ $ lsusb
Bus 001 Device 005: ID 0a12:0001 Cambridge Silicon Radio, Ltd Bluetooth Dongle (HCI mode)
Bus 001 Device 004: ID 7392:7811 Edimax Technology Co., Ltd EW-7811Un 802.11n Wireless Adapter [Realtek RTL8188CUS]
Bus 001 Device 006: ID 0a21:8001 Medtronic Physio Control Corp. MMT-7305WW [Medtronic Minimed CareLink]
Bus 001 Device 003: ID 0424:ec00 Standard Microsystems Corp. SMSC9512/9514 Fast Ethernet Adapter
Bus 001 Device 002: ID 0424:9514 Standard Microsystems Corp.
Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub
It doesn't find the carelink, right?
jaylagorio
@jaylagorio
Jan 14 2017 19:20
@stefanakerblom Right here I think: Bus 001 Device 006: ID 0a21:8001 Medtronic Physio Control Corp. MMT-7305WW [Medtronic Minimed CareLink]
Matthias
@ceben80
Jan 14 2017 19:20
@stefanakerblom try the following, 1st sudo tail -f /var/log/messagesthen disconnect the Carelink, afterwards connect the stick, any differences, when plug in
stefanakerblom
@stefanakerblom
Jan 14 2017 19:23
Nope. "openaps-use pump model" returns
ery
return self.process( )
File "/usr/local/lib/python2.7/dist-packages/decocare/stick.py", line 490, in process
raw = self.send_force_read( )
File "/usr/local/lib/python2.7/dist-packages/decocare/stick.py", line 606, in send_force_read
self.link.write(reader.format( ))
File "/usr/local/lib/python2.7/dist-packages/decocare/link.py", line 47, in write
r = self.serial.write( string )
File "/usr/local/lib/python2.7/dist-packages/serial/serialposix.py", line 517, in write
raise portNotOpenError
serial.serialutil.SerialException: Attempting to use a port that is not open
4
jburke41
@jburke41
Jan 14 2017 19:25
I switched to a new phone (android to iphone) and now I am having trouble connecting openaps to nightscout when I am not connected to the home wifi. I have the Raspberry Pi 3 and I have paired it with the new phone. It says that it is connected, however there is no openaps data in nightscout. Any advice?
Eric
@ecc1
Jan 14 2017 19:25
@jaylagorio This and some other sites seem to say that it will only work if you have the "external antenna" version of the Edison compute module: http://electronics.stackexchange.com/questions/135916/connecting-external-2-4ghz-antenna
Sacha M
@coolestkidsever
Jan 14 2017 19:25
@stefanakerblom I followed the setup yesterday and have got to the same issue, if you resolve it - please let me know :smile: I can't get to test mine again at the moment
jaylagorio
@jaylagorio
Jan 14 2017 19:27
@ecc1 I do have that. The Edison board and Explorer boards are different. The Edison board works like a champ with the antenna in the 2.4 GHz stuff. What was thrown out on Gitter a couple weeks ago was that the 900 MHz Explorer radio was improved with the 2.4 GHz antenna. Testing on my end doesn't support that statement.
stefanakerblom
@stefanakerblom
Jan 14 2017 19:27
will tell definetely. and I'll donate a bunch of $s to nightscout when I have this up and running
woodtrevor
@woodtrevor
Jan 14 2017 19:31
@stefanakerblom @coolestkidsever Have you rebooted your pi since running the setup? I seem to remember seeing this early on and I want to say that a reboot after running the setup fixed it.
Sacha M
@coolestkidsever
Jan 14 2017 19:32
@woodtrevor actually I don't think I have. Will try that when I can get back to it, thanks
Hi all, I upgraded to Grilled Cheese yesterday, (whilst starting to setup openAPS) and since then my nightscout webpage isn't receiving any data.. the uploader phone is receiving it ok, but mlab isn't, and neither is azure..- sorry to ask here, but I just wondered if anything I might have done in the openAPS setup would have affected by nightscout setup, so I know where next to start troubleshooting
Eric
@ecc1
Jan 14 2017 19:47
@jaylagorio Oh, I misunderstood because you said you were right to be skeptical about it working on the Edison. But I am not surprised that a 2.4GHz antenna doesn't work well at a very different frequency! I use an "external antenna" on my 900 MHz radio module consisting of just a 3" length of hookup wire, and it works quite well. Might be worth a quick experiment, since you can easily poke the insulated wire through a hole in the tin.
jaylagorio
@jaylagorio
Jan 14 2017 19:48
Woops! You're right, I did. Typo on my part
ohbrandon
@ohbrandon
Jan 14 2017 20:01
I've got everything setup, getting preflight pass.. etc. But the nightscout page is showing Pump: undefined and OpenAPS: unknown
Went through azur/plugin setup. What have I missed? :\
ohbrandon
@ohbrandon
Jan 14 2017 20:36
Looks like its due to the test pump.
Sacha M
@coolestkidsever
Jan 14 2017 20:38
I hadn't set my test pump up to receive any data yesterday, has been gathering data today, so think I'm in the same place as you @ohbrandon
ohbrandon
@ohbrandon
Jan 14 2017 20:39
Yeah, the docs say the pills won't show up without real data, but I think they do show.. but are incomplete.
The GF is waiting on supplies for this pump to come in before we can finish the setup.
Sacha M
@coolestkidsever
Jan 14 2017 20:41
You can get the test pump to look like it's running without any supplies. Yeah mine show up too but undefined etc
Tim Street
@tim2000s
Jan 14 2017 20:41
I withdraw my earlier comment. The Dev branch is now fully working for me on a worldwide pump. Now to set up BLE!
ohbrandon
@ohbrandon
Jan 14 2017 20:42
Her dexcom data has been pushing to nightscout for months now using xdrip, and the pi has the loops running without error.
but I am assuming that without anything connected to the pump this is as far as I can go
Tim Street
@tim2000s
Jan 14 2017 20:42
No manual intervention needed @danamlewis @Hardwick1973
@cjo20 have you reinstalled dev yet?
Sacha M
@coolestkidsever
Jan 14 2017 20:49
@woodtrevor @stefanakerblom rebooting pi seemed to resolve my port not open error
woodtrevor
@woodtrevor
Jan 14 2017 20:56
:+1: @coolestkidsever @stefanakerblom
Melody Andrews-Caron
@melodyandrews
Jan 14 2017 20:58
@Aaybob @tmecklem I did run the timezone set-up & verified that it is set correctly. Gave my transmitter a full charge yesterday, then calibrated this morning - still no luck
Sacha M
@coolestkidsever
Jan 14 2017 20:59
@ohbrandon I forced our test pump to recognise a reserviour that wasnt there (putting little finger in the reservior so it thoguht there ws one in there), if that helps you move further while waiting for supplies
ohbrandon
@ohbrandon
Jan 14 2017 21:07
Thanks -- I used the same trick. Logs are outputting data, ending with 'Completed pump-loop at Sat Jan 14 14:20:47 CST 2017'
Sacha M
@coolestkidsever
Jan 14 2017 21:11
I now have data in my pill boxes!
so excited!
ohbrandon
@ohbrandon
Jan 14 2017 21:12
With a test pump (disconnected?)
Sacha M
@coolestkidsever
Jan 14 2017 21:12
with the test pump
ohbrandon
@ohbrandon
Jan 14 2017 21:12
so its just a soak period then?
Sacha M
@coolestkidsever
Jan 14 2017 21:12
i set the test pump up 12 hours ago to receive sensor data, realsied this morn i had a few errors in setup, but its working now
ohbrandon
@ohbrandon
Jan 14 2017 21:13
Oh right -- I'm missing sensors.
She has them in the mail
NM, stand corrected -- she's using dexcom for CGM, moving to MM pump from Omni
Sacha M
@coolestkidsever
Jan 14 2017 21:15
ah ok, sorry.. i just assumed infusion sets
ohbrandon
@ohbrandon
Jan 14 2017 21:16
So I'll assume I've got it all setup correctly until and the gaps exist because the test pump is just laying on the table without being connected. Congratulations on sorting it all out though!
Chris Oattes
@cjo20
Jan 14 2017 21:17
@tim2000s I re-ran oref0-setup. Taking a day off from working on it for today though. I discovered last night that the way I had my power cables running in the case was interfering with the edison wifi, so something similar might be happening with pump comms. And it'd explain why only some people get it
jaylagorio
@jaylagorio
Jan 14 2017 22:02
@ecc1 I wonder if I take the antenna board off the end of the wire and make sure enough of it is outside the can? Think that'll work similar to your 3" random wire antenna?
Eric
@ecc1
Jan 14 2017 22:12
That wire is probably small-diameter coax cable, with a shield surrounding the center conductor, so no (unless you strip off the shield)
Ryan anglani
@ryan_anglani_twitter
Jan 14 2017 22:13
@ceben80 i am at this point in the process >Configure Bluetooth Low Energy tethering [optional does this mean i can connect the pi3 to the motoG that is blue toothed to the dexshare4 and this will give us a connection to the internet rater than relying on wifi ?
Eric
@ecc1
Jan 14 2017 22:14
Better off just adding 3" of wire to the center conductor at the end of that coax, if you're going to be removing the antenna board anyway.
Or else desoldering the coax from the male uFL connector and attaching your wire there.
Ryan anglani
@ryan_anglani_twitter
Jan 14 2017 22:16
@ecc1 do you know the answer to my question? Hello all ! i am at this point in the process >Configure Bluetooth Low Energy tethering [optional does this mean i can connect the pi3 to the motoG that is blue toothed to the dexshare4 and this will give us a connection to the internet rater than relying on wifi ?
PieterGit
@PieterGit
Jan 14 2017 22:17
@scottleibrand : more autotune updates (mainly excel), see openaps/oref0#333 . Can you review/merge them?
Dana Lewis
@danamlewis
Jan 14 2017 22:18
@PieterGit cool, we'll take a look
Eric
@ecc1
Jan 14 2017 22:20
@ryan_anglani_twitter Yes, pretty sure that's what it means. You enable Bluetooth tethering on your phone, and this lets your rig use that as its gateway to the internet. I can't be of any further help since I've never succeeded with BT tethering and haven't set up openaps recently.
PieterGit
@PieterGit
Jan 14 2017 22:21
@danamlewis : thanks. btw i noticed only the first isfProfile item gets updated in (json: 'isfProfile'/ 'sensitivities' / '[0]'). Is that a bug?
my isf profile from 00:00 to 07:00 gets updated by autotune, but all the other isfprofiles won't update by autotune (not a bit)
jaylagorio
@jaylagorio
Jan 14 2017 22:24
@ecc1 I'll add the 3" of wire to the coax, it'll probably be easier to desolder the board and do that than to take the connector off. Thanks for the suggestion!
PieterGit
@PieterGit
Jan 14 2017 22:25
@danamlewis : i'll do some doc updates for autotune and dexusb later. Too late for now, and i don't know if I have time for it tommorow (not all people devote 24x7 to openaps #AllExceptScottAndDanaAreSometimesSleepingOrDoingOtherStuff
Eric
@ecc1
Jan 14 2017 22:29
You might get even better results if you solder another 3" length to the end of that shield braid, and try to orient them 180 degrees from one another (then you'll have a 1/2-wavelength dipole instead of a 1/4-wavelength monopole)
Dana Lewis
@danamlewis
Jan 14 2017 22:30
@PieterGit right now it does one isf profile; there's an open issue for tackling multiple isf but not sure anyone's working on that yet
And we don't do 24x7 either ;) which is why sometimes it takes a day or two to review PRs; really appreciate all of the work you're putting both into oref0 and with docs! :smile:
jaylagorio
@jaylagorio
Jan 14 2017 22:33
@ecc1 So you're saying add the 3" wire to the existing solder location on the board but aimed back in the direction of the connector?
PieterGit
@PieterGit
Jan 14 2017 22:35
@danamlewis you mean openaps/oref0#326. But I think it's a bug that autotune will create a isfProfile with more than one entry. Only the first entry is right (and works for 24h), the rest of the entries in isfProfile are bogus. Am I right? I think that if openaps/oref0#326 is not fixed you should shorten the isfProfile to 1 entry
With bogus I mean copied from the original profile, and not autotune'd
Ryan anglani
@ryan_anglani_twitter
Jan 14 2017 22:39
@ecc1 ok thank you
Dana Lewis
@danamlewis
Jan 14 2017 22:43
@PieterGit on the go now so will have to look later; I only have one isf, too, so may need to see some example outputs of what you're referring to, to be sure I am understanding.
Eric
@ecc1
Jan 14 2017 22:44
@jaylagorio I mean: disconnect the antenna board from the coax. If there are two usable connection points, great: connect a 3" wire to each of them. If not, you'll have a little more work to access the 2 conductors (shield braid and inner conductor). Something like this might be helpful: http://www.hcarc.us/articles/soldering%20PL-259%20connectors.htm
(ignore the part about the connector)
When you're done, you want something like this:
A6-3EN.jpg
Eric
@ecc1
Jan 14 2017 22:49
Or:
dipole.png
jaylagorio
@jaylagorio
Jan 14 2017 22:50
But connect each conductor in the jacket to the same end (not the center) of the wire, right?
Mike
@libxmike
Jan 14 2017 23:09
@jaylagorio here is a how-to I made for BT dipole, same prinzip http://imgur.com/a/7CseC
jaylagorio
@jaylagorio
Jan 14 2017 23:14
@libxmike That's awesome, thank you!
Eric
@ecc1
Jan 14 2017 23:14
Exactly ^^^ except modify lengths for 900MHz
jaylagorio
@jaylagorio
Jan 14 2017 23:14
Yup, that's the plan. Thanks to you both!
Mike
@libxmike
Jan 14 2017 23:17
915 1/4 wave = 77.9mm
915 1/2 wave = 155.9mm
915 full wave = 327.8mm
jaylagorio
@jaylagorio
Jan 14 2017 23:17
Yup, I was going to go with 1/4 wave, ~3", and see how that works
Mike
@libxmike
Jan 14 2017 23:19
you don't measure this stuff in inches, always in mm :)
jaylagorio
@jaylagorio
Jan 14 2017 23:19
#Murica
Mike
@libxmike
Jan 14 2017 23:19
hehe, yeah
here for 868
868 1/4 wave = 82.2mm
868 1/2 wave = 164.3mm
868 full wave = 345.5mm
for WW pump
ejoe132
@ejoe132
Jan 14 2017 23:45
Need some help on a rpi3 going from a carelink to my TI stick. went to the mmeowlink page entered the first bit "cd ~
git clone https://github.com/oskarpearson/mmeowlink.git mmeowlink-source
cd mmeowlink-source
git checkout master
sudo pip install -e" then hit enter and it give me this"pi@raspberrypi:~ $ cd ~
pi@raspberrypi:~ $ git clone https://github.com/oskarpearson/mmeowlink.git mmeow link-source
fatal: destination path 'mmeowlink-source' already exists and is not an empty di rectory.
pi@raspberrypi:~ $ cd mmeowlink-source
pi@raspberrypi:~/mmeowlink-source $ git checkout master
Already on 'master'
Your branch is up-to-date with 'origin/master'.
pi@raspberrypi:~/mmeowlink-source $ sudo pip install -e" So should I go ahead and do the vendor bit?
seanneedle
@seanneedle
Jan 14 2017 23:48
Hi everyone! I'm currently setting up the loop for the first time using Intel Edison and the TI stick. I believe I have everything set up correctly, but I'm currently trying to make sure my setup can connect with a Medtronic pump. When I use mmtune, I can get a frequency so I'm not too worried about that. However, when I run the command "openaps use pump model", I get the standard error "Could not get subg_rfspy state or version. Have you got the right port/device and radio_type?". However, I double checked the information in pump.ini and the serial number of the pump and the port both are correct. Does anyone have thoughts on how to best troubleshoot this?
Dana Lewis
@danamlewis
Jan 14 2017 23:58
@ejoe132 setup script will be easiest way to do mmeowlink now
@seanneedle welcome! Did you use the setup scripts? What happens when you tail the pump loop log?