You go Tom! :D
nice, that sounds interesting. I see no reason not to give it a try, but no guarantees it'll be possible to do something no one has ever done before. :)
Thank you @danamlewis i will check out the troubleshooting steps first
@djtechman @scottleibrand Thanks! lol I know I'm going into the unknown but all elements are present.. @pazaan has done a great job on reading from CGM via the CNL over USB, I have automated the button presses and have done some tests to emulate the pump's basal pulses so we can supplement them with microboluses, and of course the OpenAPS algorithm's already there. It's mostly a packaging exercise but the first thing we need is a hardware platform that can do GPIO, USB and some kind of wireless comms out to Nightscout...
@Bal00 so was open aps until a group of brains driven by a special need decided to try the unknown.
@Bal00 you might also be interested in openaps/oref0#262 with regard to microboluses
@jcorbett80 😊 @scottleibrand Ooh interesting and I like the idea of checking reservoir volume, I was planning on using IOB but the natural decay was going to be a problem, reservoir volume is a nice absolute check...!
My output now says: decocare.link.alreadyinuse and port0 already, in use after troubleshooting. I'm not sure if this message is also related to bad connection but I thought i'd check just in case it was a different problem. The command 'openaps use pump model' it shows ability to communicate
@Bal00 not sure if anyone has tried to capture it but minimed pumps work off of a pulse type engine driving the plunger. I would guess a straight single pulse with a counter. If you could access starting point and current point that would be "an exact" count.
I'm having trouble with my edison failing to connect to wifi or bt-pan after running for a while. The most recent time I "caught" it, the board was not even responding to UART. The openAPS logs last entry was at 1pm MST today, I just got around to plugging it in to my PC. Might implement watchdogs to try and keep the net alive. Any other logs I should be looking at to troubleshoot this?
I'm curious about cpu temperature / any system health related alerts
@djtechman mine runs very hot. Mine was dropping wifi and so on. Once a day it would shut down. I thought data was filling it but I had it set to reset data and compress daily. So I made a small stand I could clip mine to when sitting. It kept it in the air for better flow. When in the car I would would put it in path of air blowing from the vents. Mine quit dropping wifi and quit shutting down. Night became an issue as far as getting hot unfortunately. It would make it through ok but was very hot in the am. If I go edison again I am getting a small 3/4 fan and a larger battery for whatever case I make.
@jcorbett80 was all this without being in a case?
@danamlewis actually very hot out of case. I built a case with chimney effect and created a flow. That helped. Then making a small stand to clip it to allowed it to get flow nonstop without being blocked by clothes or anything. When I originally flashed it the first time I had to flash. Let cool then install openaps. It ran that hot. I never got openaps to finish install until I allowed the cooling. Then it went straight through without any issue
@danamlewis I have the low power
@danamlewis and when flashing the first time it was without case.
@jcorbett80 thanks for sharing. I'll try and figure out how to log the temperature on this thing & make sure I'm covering the other bases too.
I doubt there's a way to recover from thermal shutdown without manual intervention
are there any antennas people like to use with the edison / explorer to increase range to pump?
@djtechman the little black button of last resolve ;)
@jcorbett80 no one has demonstrated that the low power Edison's are actually lower power...but wonder if we should start comparing them heat wise to see if one model gives out more heat
@mhaeberli I was able to be 10-15 feet away without added antenna. I would set mine in its stand in the middle of the room and wear my pump anywhere in the room I needed.
@jcorbett80 with or without the antenna mod to the Explorer?
@danamlewis I agree. I thought I saw someone post that from the specs they read, it should actually run hotter based on how it had to work. I have been searching for that. I think it was in the cgm g5 room
@mhaeberli without mod. Not sure antenna makes a noticeable difference in explorer, but haven't seen anyone do specific tests to compare how many extra feet you get (and whether it's only direct line of sight or if extra range works through walls too)
@danamlewis k thx
^ if anyone wants to test and document definitively, that would be awesome. Otherwise we go on not-very-specific experiences
dumb question # 127 - when charging battery through Explorer, I assume the charging light goes out when battery is full?
I can't afford a couple more rugs to test. It took me 4 months to pay for the system I had. However a simple 1st level break test would be using a simple line of sight temperature gun at a measured distance for consistency. A simple board holder and gun holder so the distance is stable and consistent. Run them through flashing, operating, and charging. Temps taken at same time in a side by side in same conditions. Not hard nor intensive. Just need to make sure both are recieving same voltage which can be seen in pill on nightscout. Heat is heat. However if someone writes an appointment to measure that's cool too. But must be a direct side by side to eliminate all outside variable differences.
@mhaeberli mine did.
@jcorbett80 there is a built-in temperature sensor in the Edison
apt-get install lm-sensors to install
and then just
sensors to see the temp
I have one Edison that likes to run hot: it's currently 74C core temp. I just swapped it onto a new Explorer board this morning, and that didn't change anything, so I think it's the Edison itself that has the issue
the others are all in the 54-62C range
@scottleibrand that's kinda where I'm at also. My Explorer gets hot but it's under the edison. The edison is really hot to touch 74c is a pretty hot temp for a small computer. It doesn't have internal room for airflow to overcome that heat.
the Edison is ok up to a core temp of 90C, and mine doesn't tend to get that hot, but I'm sure that it would drain the battery fast if we tried to use that one as a mobile rig.
The edison sits right on the Explorer power supply and right above the 2 usb charge ports. So the heat that "feeds" on each other between the 2 boards is a bad scenario. But due to size it has to be. However an extension plug between boards that would allow a 1/2 inch (12-13mm) between the 2 would be a great help to seperated heat from heat and allow flow of cool air. It would create a forced back draft to pull the cool air in.
Even just 3/8" would be a great help.
so on a related note … or unrelated - has anyone tried using a wireless Qi charger with an Edison / Explorer system?
@mhaeberli yep. bad news bears heat-wise on that (see @tim2000s @logichammer’s experience with that to date)
@mhaeberli Tim street is using one
Lol Dana. You beat me to it.
k
:+1: @danamlewis @jcorbett80
@danamlewis pretty fast "for your age" ;)
@jcorbett80 there are some cheap sparkfun blocks if you want to experiment with using one as a spacer.
Has anyone had any experience with connecting a rig to a wifi hotspot at a coffee shop or hotel where you have to click on the sign-on page?
@scottleibrand are they 2 sided to use as an extender plug if I cut away the rest of the board or are they straight up extender plug blocks
@gregtcm Edimax or hootoo is handy for that
Ok. Im having an idea for looping offline. I want to know if its possible. I have been looking at some repositories for running NS and mongo on a locally on a pi3. Now would it be possible to turn the pi into a wifi hotspot and enter it into the rigs wpa_sup file and access it offline. I have a pi board i never used for openaps and i carry around a good size carrying bag for my diabetic supplies. I had the idea i could put it in my bag connected to a battery running a local night scout offline locally connect by wifi signal.
Not sure what those are ... will google them.
Mini routers, so you can log in via it and get on hotel or plane wifi and then connect all your devices to its network
Ah, thanks.
I've heard of the edimax wifi usb dongle. Thanks!
@scottleibrand looks like the base block would work. I assume the plug is a straight pass through if cut away from board. They have 3mm spacers on each side so that would give 6mm space plus thickness of the board if board was cut away. That would give close to 3/8 and allow a lot of air flow. I am going to see if I can get another Explorer boad. Reset edison back to factory and try this out.
IF that solves a lot of issues getting the Explorer guys to also sell a pass through plug to whatever size we need would be an easy fix.
@scottleibrand since there is a way to pull temp from edison would it make since to include it in open aps. Maybe pull once every 30 minutes so not a large use energy but if saves to logs it would give history on every board out there without anyone doing anything special.
I think someone did a version where it gets added to one of the Nightscout pills. Don't remember details though.
I will see if I can find that.
Been trying to get to where I can nod off for the past hour. I think I am able to now. I will Hunt for that in the am and see about an Explorer board and base board block as a spacer. I will do a comparison between no spacer and spacer to see if it helps.
Ya'll have fun.
@danamlewis Thanks to everyone for getting Hamshield to do the preflashed Edison's with Jubi. Got the whole set up done tonight in about 20-25 minutes with rig #4. Way cool. Hopefully I set the BT tethering up right!
:confetti_ball:
@Bal00 - would be very
Interested to see what you come up with.
@Bal00 still waiting for a new 640g in Canada. Release date - mid April. Seeing as we've already tried swimming with the pump, figure a waterproof model would be an improvement ;)
when i got my first Paradigm, they said it was waterproof...
then a few years later sent out a letter saying "ehhh let's call it water-resistant"
when i was little i ran into the atlantic ocean with a 508. do not recommend.
well, straight out of the box it’s water resistant. Normal wear and tear/surface cracks = not so much anymore.
yup, that was the issue
@mhaeberli - think you mean to be over here?
indeed @danamlewis - I appear to have some kind of Gitter feed disorientation
@intend-to-flash-edison - getting Ready for application in PuTTY, but unable to flash edison either through Intel app nor through flashall.bat - re - gittering here ...
@mhaeberli both of the boards we have done trace cuts on are alpha boards, so the exact locations will differ from the production boards (which have the proper cut location printed on the antenna)
@j4cbo we all knew it wasn't waterproof - perhaps we can still call it water resistant as it lives after a week in dessicant... would prefer not a repeat though. At 10yrs old, there are so many things to remember - the pump clipped to the bathing suit is not a great idea, as we learned... very easy to forget...
@scottleibrand :+1:
the ugly cut on the outside-the-case board is one I did with a sharp kitchen knife. the one in the purple and pink case was done by Morgan with proper tools. :)
@JELCRAWFORD i'm paranoid now, i had one have its buttons short out in a rainstorm on christmas eve a few years ago
how do people handle monitoring Edison battery power level and / or clean shutdown as the battery gets low? or is there somehow nothing to worry about ?
it's not a very big deal: the Edison doesn't corrupt its file system on power loss like the Pi does. But just for good measure, we just merged in a fix for that into dev: openaps/oref0#395
@jcorbett80 I'm doing a side-by-side battery-drain test on my always-hot edison and another one that runs cooler. after 35 minutes, the hotter one is not draining the battery any faster than the cooler one. we'll see how they do as they both drain down to zero overnight.
their temps on battery are 44C and 55C
@scottleibrand thx
@jcorbett80 look forward to results ! @scottleibrand I would be in favor of an optional privacy-violating data gathering scheme on the population of OpenAPS Edisons to see how long they last on battery, temps, etc (we would opt in once we were so far) - one could declare battery make and model, capacity, etc., … but no, I’m not in any shape to code it until at least after we get OpenAPS spun up again ...
Where can I buy a case for the explorer + Edison rig? Thanks.
There is an entire section of the docs on cases.
anyone awake ? I set up my edison rig to connect primarily to my home wifi and thether to my iphone6. I also stupidly added the ability to connect to open wifi hotspots. I took the rig for a first time outside my home yesterday and it nicely stays online via my phone, yay . I brought it to work this morning and it probably picked the open wifi hotspot at work. It uloads to NS, but I cannot reach it with Termius, since I can't figure out what its IP is... Do I have to take a walk outside and remove the open wifi hotspots metion on wpa-supplicant? Or is there another way to log into the rig ?
Hi @scottleibrand. I saw the section on cases but none are for sale. They are all free designs for 3D print but I have no idea what to do with these files. (I've never done 3D printing before ).
You're can visit an online printing service like shapeways, or search a local print store or try a local maker group. They can print the files
now NS supports diplaying AMA features fast as before. (IOB,COB,AMA results). use dev branch
@SandraK82 Thanks Sandra.
hello..how could this work please :
git clone -b dev https://github.com/<lucilegarel>/cgm-remote-monitor.git
cd cgm-remote-monitor
git remote add upstream https://github.com/nightscout/cgm-remote-monitor
git fetch upstream
git merge upstream/dev
git push origin dev
cd cgm-remote-monitor
git remote add upstream https://github.com/nightscout/cgm-remote-monitor
git fetch upstream
git merge upstream/dev
git push origin dev
just a newby question ! (how the prompt command terminal could be able to run this command ?)
@scottleibrand are yours in cases that to carry with you all day? Or is your hot one a home rig your able to leave laying?
with all this talk of heat from the edison I'm wondering if anyone has made/used and aluminum case to help with heat sink?
sorry, I type a wrong 'git' command in my putty screen, so I can't succeed any more in running cd && ~/src/oref0/bin/oref0-setup.sh
How could I delete it ?
How could I delete it ?
@danamlewis when your up, awake, and not at work. Please send me detailed info about the openaps session on the7th if you dont mind. I may be able to come. Thanks in advance.
@danamlewis that is of course if all 3 of those things ever happen at the same time. :)
Dana is always awake
On the issue of Edison running hot, when we were running our "offline" Edison rig with attached supplemental battery (via UART port) and Dexcom (via USB OTG) in the Grand Canyon last week I found that it ran significantly hotter than normal. Didn't measure any actual temperatures but one day got hot enough that it melted through the electrical shrink wrap overlying the Edison chip and the rig shut down. It was in a backpack exposed to the sun, so that exacerbated the situation but we found it was running much hotter than usual otherwise as well. Much hotter than when on charger (for comparison). I'm going to try to figure out how to make the Dexcom stop charging when attached in this way to see if that makes a difference when I have a chance to do some more testing. Anyone else using a version 1.2 board in this manner and seeing similar results?
@j4cbo - I think there is a fair bit of paranoia here now too :)
@jbwittmer the problem could be the shrink wrap (correct me if I'm wrong, did you put the rig/edison in a shrink wrap tube?)
and no air flow caused by shrink wrap
@libxmike I'm sure you're right and that could contribute. Yes, it's in a thin shrink wrap tube (open on both ends and a slit for the charger ports) but otherwise no case or other containment. Typically worn in a thin nylon athletic pouch. But for what it's worth that's the way all of our rigs are configured (4+) and none has ever had prior problems with heat. Edison chip faces out from battery. Normally never gets anything more than slightly warm to touch. So I think that there is something about the combination of supplemental battery and Dexcom that was making it run hotter than usual.
@cameronrenwick the problem with aluminium case is that there is no direct contact to the chip on edison, just a aluminium case won't do much w/o heatsink attached direct to the chip, in that case just air flow, a micro fan may help if overheating is the problem
libxmike: it might do something. Although it'd need to be in direct contact for most efficiency, it'll still do better at absorbing / radiating heat from hot air inside hte case
@cjo20 yes, absorbtion will do its job and transport outside the case but then again, size and where the case is, if i nthe jeans pocket not much
I'm designing a case for Edison/Explorer Board and 18650 battery where there may be place for a micro fan but it may be a overkill, will be available on http://www.thingiverse.com/UltiMike/designs
@lsandini You could connect to the Edison via USB port by wire connection to another computer and ssh in (PUTY or Terminal, etc) and edit wpa_supplicant.conf to remove the offending wifi network.
@scottleibrand Thanks for the edimax link. (Just now seeing it.)
@libxmike I totally hear you. I haven't "really" thought this out but was wondering if there'd be some way of having the chip touching the case (not glued etc but simply contacting) in order to help radiate heat. I look at macbook pros for example and the metal case does help to cool the chips due to proximity. Just kinda rolling ideas around really.
@cameronrenwick then copper will be better in this case as has better heat absorption (thermal conductivity: copper 0.99 aluminium 0.5)
copper heatsink touching top of edison and attached to aluminium outside casing
@libxmike excellent point... fortunately I've got access to copper, alum, steel sheet stock so manufacturing the case, while not 3d printed, is possible. I think I'll look a little deeper into this.
well, I'm limited ............ to 3d printed cases , in the '90 I designed lot of stuff in sheet metal for husqvarna, honda, stihl and yamaha and manufactured dies for it
@jbwittmer I started using a heat shrink tube the past week. I've only been looping for the past two, but I haven't noticed a difference in heat between the two weeks, ie no tube vs. tube. I'm using this (recommended by another looper) : https://www.amazon.com/gp/product/B009IILEVY
That's the same product I use. We've found it to be very durable and lightweight without adding to bulk of rig.
:+1:
anyone have any issues caused by placement of explorer board right up next to lipo with the explorer antenna....thinking this could have been causing some comm issues for me.....?
I've been getting very spotty temp basals set....and openaps and pump pills load inconsistently...
@jbwittmer That was one of these days... didn't have a chance to even try that option at work, but as soon as I left the hospital and its "offending" wifi, I was bluetooth-tethered again and all is fine. I came home and removed the network={key_mgmt=NONE priority=-999 } entry from the wpa_supplicant.conf. I guess the Edison drivers must be installed on the PC I would use to connect to my rig ? I doubt our IT folks would like to see me do that :smile:
@lsandini what is supposed to happen is: if the rig connects to an open wifi and can't get online, it will disable wifi and connect with Bluetooth instead. You'd have to look back at your network.log to see why that didn't happen.
The easiest fix, though, is usually to turn on wifi tethering on your phone, assuming that is configured in your rig. That will cause it to jump off the open wifi onto your phone's, and then you can get in to troubleshoot / reconfigure.
I'd be interested to know what it is about your work wifi that causes the oref0-online captive portal detection to break.
My rig connected to the open wifi and was on a different IP range than the other computers. I thought my rig would prefer my phone's hotspot since it specified priorities in the .conf file, but still it preferred the open wifi . And it was online on the open wifi for sure, since it uploaded to NS.
@jcorbett80 after a full overnight run, I am seeing a difference in battery drain now:
687 {"batteryVoltage":3646, "battery":45} +45.0°C
687 {"batteryVoltage":3542, "battery":30} +55.0°C
687 {"batteryVoltage":3646, "battery":45} +45.0°C
687 {"batteryVoltage":3542, "battery":30} +55.0°C
@lsandini ah, that explains oref0-online's actions then. the wifi was working fine then, so no reason to switch off wifi for hotspot
not sure why it wouldn't switch to the higher-priority network, though. if you wanted you could've left the hotspot up and power-cycled the rig though
FWIW, I have my rigs configured to automatically establish ssh tunnels to a server so I can access the rig from there if I can't get to it locally
@ptinson and I were testing out an OpenVPN config that would accomplish something similar, but that hasn't happened yet as we've been working on higher priorities.
@jcorbett80 the first number is the number of minutes since I unplugged power. so that's after 11.5 hours.
That SSH tunneling sounds like a good solution. I'll ask again If I cannot figure it out by myself, but in the meantime if there is an how-to anywhere, I'd be happy to read it...
the meat of it is:
/usr/bin/ssh -fN -R $1:localhost:22 you@yourserver.net, where $1 is the port number you want to use for that rig.
thank you Scott !
you'll need to set up a passwordless ssh key on the rig that is in authorized_keys on the server, so you'll want to do this on a server that you're ok with being wide open to anyone who can log into your rigs.
Using Mac to flash Edison. The flashing process failed and now won't mount. S MC reset fail to change the situation. Any thoughts on how to get my computer to recognize the Edison again?
I also have some other stuff in my script to
/usr/bin/killall -g --older-than 10m sshwhen necessary so the tunnel can re-establish automatically, like so:if ! /bin/ps -ef | /bin/grep ssh | /bin/grep $1 | /bin/grep -v grep | /bin/grep -v launch; then
/usr/bin/ssh -fN -R $1:localhost:22 me@myserver.com
else
#/usr/bin/killall -g --older-than 12h ssh
if ! /bin/ps -ef | /bin/grep sshd | /bin/grep pts/ | /bin/grep -v grep; then
/usr/bin/killall -g --older-than 10m ssh
fi
ping -c 10 myserver.com || /usr/bin/killall -g --older-than 10m ssh
fi
taking notes :)
@msrcgm first step is to get your UART serial console working again with screen, so you can see what state the Edison is in
if you get stuck with that, scroll back in this channel: I've walked a few people through this in the last few days (which means we really need someone to PR in to the troubleshooting section of the docs whatever is missing from there)
i used boluswizard on my minimed for my lunch about half an hour ago. that data showed up in the ns chart, but its showing 0 COB and openaps thinks i'm going to plummet. any tips?
@andrew-anguiano do you have AMA enabled?
@scottleibrand im honestly not sure - just got my rig started last night so its all pretty new.
ok,
cat ~/myopenaps/oref0-runagain.sh and see what it has after --enable
@scottleibrand makes sense. I would guess the battery heated up more also. From my studies (which is very open to "you fell and hit your head" type comments without offending me) is that a battery is best and safest at room temerature. By best I mean lower drain. By safest I mean as it heats up (especially a LIPO) the more chance of "bad behavior" or as I call it...the Samsung effect. So you add the extra draw along with being paired in a case with a hot board and the battery had much less ability to function as it should and the added "do i belong in a galaxy note 7? It seems to be getting hot in here" However I don't think we are talking too much heat here but definitely enough to drain a battery for our purposes much faster. Then add summer and you will lose more. We're those boards in a case where heat would be even greater at times?
ok i'll check now. thanks!
if you don't have
meal there, you won't get AMA
@jcorbett80 I think you might be (mostly) confusing cause and effect. while battery efficiency is a function of temperature, I think that's a much smaller effect in the 20-90C range than you're assuming. I think what's going on here is simply an energy in = heat out effect, and of course more heat out = higher temperature (for a given case's heat dissipation)
@scottleibrand your probably correct. I think through situations out loud to get knowledge back.
yeah, that is a very effective method of learning. :-)
@scottleibrand i have no enable flag at all. i'll rerun with that enabled. thanks for the tip!
:+1: and before you do, be sure to read up the part in the docs about the advanced features you're enabling, so you know what they're going to do
will do!
Haha @cjo20 @jcorbett80 I am not always awake ;p
@jcorbett80 (and anyone else interested in meet ups) here's the email I sent out with links to the relevant meetup rsvps: https://groups.google.com/d/msg/openaps-dev/cLaboQwPrbo/x6GOatYdAgAJ
San Diego/Boston/Huntsville/maybe Ft Walton, etc.
@cjo20 in fact, @danamlewis is a champion sleeper. Hence the original need for DIYPS louder alarms.
Also, per your Q about hot vs less hot rigs - they're all in the soft cases I use. @jcorbett80 (not actually @scottleibrand 's rigs at all ;) :))
Wow, a meetup in Huntsville.
Im in.
@applehat :smile:
@danamlewis with things I do i can do a soft case without worry if bending the lipo more often than not.
Can not do
@jcorbett80 you could always get an 18650 holder and use that. Those cells are metal and can handle the abuse better then a soft lipo.
Not trying to convince - just telling where mine live for comparison/context in hear discussions.
@scottleibrand is there a place where i can see all the enable flags and configurations? the advanced features page doesn’t show the actual flags to use when running
oref0-runagain.sh and i’m still not seeing COB in NS after updating and eating a snack.
@andrew-anguiano not seeing it in OpenAPS pill?
(Fasted way to see all the config options would actually to be to look at oref0-setup.sh code, fwiw, if you want that level of detail)
@danamlewis hovering over the pill I see
COB: Undefined
But instead of run again it's probably easier to run the regular setup.sh vs runagain to add advanced features in
ok, i’ll take a gander at that. i just want to make sure i actually have the correct flags.
:+1: I'd just run regular setup vs runagain. Then you'll get all the flags in your runagain for future
i was wondering if running from scratch would be a better idea. i’ll give that a shot as well.
thanks so much @danamlewis!
:+1:
@applehat good thought. I will look for that. Do they also put out a 3.7v with the same plug?
Or are you talking about a battery case?
@applehat I just found this doing quick research to figure out the 18650's. Might be good for people to look at. I am also going to look at a simple 18650 case to use with lipo. Thank you for the info. Here is the link for the group. https://batterybro.com/blogs/18650-wholesale-battery-reviews/88881030-5-common-lithium-ion-battery-myths-explained
any idea why if I tail loop log it seems to conk out periodically....though I can see that rig is still "up" as I can poke around the directories, etc...
We don’t know what you mean by “seems to conk out” - can you paste logs/show what you’re talking about?
@danamlewis ....literally stops reporting anything from terminus.......and shows
last line and then stops
will paste in output later---unfortunately cant do it now.
been getting very spotty enacting of temps.../reporting of BGs via share apparently.
@scottleibrand @lsandini I will write a how-to with the script for setting up openvpn this weekend. Initially this will be for those that want to run their own vpn server, its actually very simple:)
Been busy on too many things
:+1: @ptinson
@jcorbett80 Hey sorry was away from chat. Yes, 18650s (any many other Lithium-Ion cells) output 3.7 Nominal (range of 3.2-4.2v, like a LiPo). They are arguably safer then LiPo cells as well. They are the cell found inside most "USB Battery Packs", Power Tool Battery Packs, and a few other things.
You can buy sleds for them from many places already (since they are a standard)
I've even seen a few Edison / OpenAPS 3d printed case designs based around them
A potential other benefit of using an 18650 would be the ability to store some extra charged ones in a bag (in a case!) for long times away from a charger.
I use 18650s for a few things and have had good luck with https://www.imrbatteries.com/
Oh - and any circuit designed to safely charge a lipo cell can safely charge a li-ion cell as well (assuming your not breaking the C-rating of the cell).
Ok, I'll stop rambling about batteries now.
Joining late on the topic of edison overheating... I was brainstorming on integration of a heatsink into my 3D Printed case when I thought - why not just make the whole case out of aluminum and bond the edison's shielding to the case with heat-sink paste?
I have the machining resources to carve something out, looking at sourcing materials now
That would dissipate some heat away from the edison, but without fins / fans on the case to dissipate that heat, would you not just be slightly slowing an overheat, instead of actually cooling?
If completely encapsulated, that is.
well I thought about that, it's in my pocket, so the aluminum will be about 98.6F all the time
but... the edison's thermal shutdown is going to be well over 100F, and if it conducts to the aluminum I'm certain it would equalize the heat
I think the more material we have to sink theheat into, the faster it will dissapate into the air
@djtechman that’d be great for dissipating heat, but pretty bad for connecting to wifi. :)
Or the pump
ah pete, you got me there :) Would need external antennas :P
I wonder if some sort of heat pipe embedded into the case would work.
Looking into it just now, Im not sure if a heat pipe is really something possible to DIY
on the heatsink idea I could just integrate a huge vented chunk of aluminum into one of the case walls
(only covering the edison's metal shield)
What about a Peltier module?
Tho the whole "hot on one side, cold on the other" could lead to issues with it being crammed in a pocket.
And I assume the probably pull a decent bit of power.
peltiers are super inefficient
IMO we probably want to be focusing on reducing energy use to save battery life and reduce heat, rather than just radiating the heat away better
+1
@garykidd does the terminal still respond when the log stops? does it start back up again happily if you Ctrl-C and type
l (that's the lowercase letter l) again?
@scottleibrand I was also thinking of it this way. I probably need to focus on getting my networking over to BTLE/bt-pan. Wifi is killing my phone and the APS rig in less than 12h
I have a question. I think I may be the one that brought a heat discussion to this level when I spoke about my heat issues last night or the night before. Lol forgot when. So my question is. Are this many people having heat issues? Or we all reacting to my single known issue thinking it is bigger than it may be? If a lot seem to be having issue we need to find out for sure how big the issue is. How bad the issue causes more issues and at what temp point do we see a true issue. I did a lot of break testing while working in I.T. and with front end developement. I have seen boards rated for 100C just to find out that is the total failure point. Issues start at a much lower point. If we are just bouncing knowledge to help me I love it and thank you for your help!!
it seems that there are a relatively small number of Edisons that run hot, and a small number of pre-1.2 Explorer boards that got too hot during charging. the latter problem has been fixed, we believe, in the v1.2 Explorer board run
I wonder if mine fell in the previous 1.2 run.
unless you just got it in the last 2 weeks, it did. do you see it run hot all the time, or just when charging?
Pre. I hate android word correct. It hits as it sends
Just to add to that. I created a small script to run every 15 mins and create a record of Edison core heat. The Edison cores in my pocket are running typically at 41-45 Celsius.
the Explorers have their date or version number printed on them: if it doesn't say 1.2 and just has a date, it's older
Both. However when charging and when flashing it pretty well cooks. As I told Dana when I first flashed I couldn't get it to install openaps. I notice how hot it was. So I let it cool. Flashed it with Jubinux to make sure it was a clean try. I then let it cool and installed openaps without any issues.
And I got mine more than 2 weeks ago.
if it's too hot even when running on battery and not plugged into USB, it's probably the Edison, not the Explorer board, that is faulty
Mine just has 2016
I wonder if I have the Explorer issue and a hot edison combination both causing my what seem to be worse issues.
I've contacted the hamshield group and let them know what has happened and asking for guidance. I haven't heard back. I assume I will have to purchase a new explorer board. But holding out for hope. I only was able to run full out openaps for a few days before total port failure.
i can recommend this smartipi touch case for a home / fixed rig. Very nice, to just touch a screen to see Nightscout, see https://www.kickstarter.com/projects/799765240/smartipi-touch-hat-cover/?ref=kicktraq
@danamlewis who were you talking to yesterday that wanted to ride with someone to the meet? I can't find his post.
@PieterGit - I'm new to the openaps world - is it common for people to have "home only" rigs? I always assumed it would be something i'd really want to have 24/7, but I suppose I could see the benefit of it even for just overnight (since I can go to bed with a BS of 100 and wake up at 170 without eating anything).
because of redundancy and development most people have more rigs. the explorer board/edison boards are the best for mobile, but there are quite some pi3 users with a TI USB stick
I suppose thats a good reason many people run off data from Nightscout? So that multiple rigs can be used at different times without dealing with any data loss.
@jcorbett80 was that @webwiztn, I think?
Yes, it was me! I have RSVP'd the meetup. @jcorbett80 if you want to drive down to Nashville, we can meet and take my car. My preference would be to be in HSV by 10:30 so that we can be back in Nashville by supper time.
BTW, my new Explorer board just arrived but the new Edison won't be here until Friday. I ordered it the day before you encouraged folks to order the already-flashed version! Curious to see which component is bad!
@danamlewis What's the secret to disconnecting the battery without pulling out the wires?
what openaps command is doing the upload to nightscout in a G4-upload config?
@webwiztn I use fingernails to grip small flange on white plug. However with Explorer board you don't have to unplug to charge.
And yes we can meet and drive together
I guess i need to find out what time are you calling supper. I am guessing about 4 hour drive +/-.
Sorry meant 3 hour
From my house in West Nashville, it's 2 hours and 11 minutes. Supper, like 6PM? I am flexible, just have to take care of a dog! We can figure out a good place to meet.
Thanks for following up @danamlewis
@jcorbett80 I need to disconnect the battery to test the new boards.
And, yes, thanks @danamlewis!
:+1: looking forward to meeting ya'll!
@danamlewis you say that now.
I don't think I've ever regretted meeting an OpenAPSer :smile:
@webwiztn By the way my name is james. I don't know but I am guessing yours isn't webwiz.
It actually is. His parents wanted a unique name that would represent the career that had picked out for him.
I think its a nice name.
Or her. Didn't mean to assume.
Lol
scott@scott-mbp ~/Desktop/docs $ git blame ./docs/docs/walkthrough/phase-0/loops-in-progress.md | grep webwiztn
e306101d docs/docs/walkthrough/phase-0/loops-in-progress.md (webwiztn 2016-06-19 17:38:55 -0500 43) - Arlene Samowich (Nashville, TN)
e306101d docs/docs/walkthrough/phase-0/loops-in-progress.md (webwiztn 2016-06-19 17:38:55 -0500 43) - Arlene Samowich (Nashville, TN)
Although I do like the idea of Webwiz Samowich ;-p
I am trying to do BT tethering. I realized though that I have the iPhone hotspot settings in my wpa_supplicant file. Do I have to remove that? I am betting that I need to, but wanted to confirm.
no, you can leave them
if you activate your hotspot on the iPhone but then quit out before anything connects to wifi, it will only leave BT hotspot activated
if you hold the Personal Hotspot screen open, it'll allow the rig (and anything else with the network configured) to connect via wifi
@jcorbett80 Not sure why my name doesn't appear as it does in github but it's Arlene Samowich. @scottleibrand I actually had a cat named Tuna and did not realize the irony until one day at the vet's office when the VA called Tuna Samowich!!!
lol
@webwiztn Never used it before today but I am sure slack allows pvt messages. Please send address of where you want to meet and what time. I am assuming @danamlewis Dana will have fresh coffee waiting for us when we get there.
@applehat and by the way...he is a she. :)
@jcorbett80 that’s what Panera employees are for ;)
All - please take a look at openaps/docs#732 and please provide any input (esp. if you have any objections) or a thumbs up if you support this getting merged over to master. This is a proposed template (facelift, aesthetic only) change to RTD. Worth checking it out on mobile, too, as it greatly improves navigation/visability on various sized devices.
Oh no @danamlewis that northern life has ruined you.
@jcorbett80 you would think that moving to Seattle would make me like coffee, but nope. still no go on caring about coffee.
A good southern person never invites anyone without serving coffee and cake or pie
I serve IT help instead ;)