-
Notifications
You must be signed in to change notification settings - Fork 102
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
LIMITTER #1
Comments
English would be more easier to understand. |
I speak a little spanish - here is a rough translation: Hi, my name is Pedro and I am the father of a Diabetic boy that uses the Freestyle Libre. First I want to give my congratulations for developing the project since we have been waiting some time for something like this. I want you to share the design for us to be able to use it with my boy, above all, for the overnight watch. Thanks for your time and dedication. I am convinced that there will be many of us parents that thank you. I hope we have more updates from you soon. |
@berniemcintosh , you are welcome. |
Congratulations for developing such as great project. |
I dont't know any other working module. |
Jeers, is there any trick to pair with Android? I use same parts as you do. To see the name "Limitter" in Settings-> BT, I had to change the code to: |
Settings -> Apps -> xdrip -> permissions. Enable location services. |
Thanks Mike. I'll try that. I can see the HMSoft device in xDrip without that mod and it pairs, although it didn't transmit any data. Do you have the correct baud rate set? Sent from my iPhone
|
Hello,I am a T1D for six years and user of a free style libre,I just wonder that LimiTTer(the hardware )could interpretate the Tag-data or transmit the raw data to xDrip directly.For I have used several libre apps ,the value they read are different from the official reader. |
Okay, my progress so far is that I've built the LimiTTer, and that all the components communicate. I can confirm the Arduino is speaking to the HM10 as, with the earlier tweak, I was able to get the module renamed correctly. The assembly then connects to xDrip on the phone successfully. At the five minute intervals, the NFC card comes to life and I am assuming it scans, and I see the green LED flash on the Arduino. What I don't get is anything in xDrip. Nothing is showing up. I welcome any ideas on what I might need to change. The baud rate is set to 9600 as is the default on my HM10. Just not picking anything up and transmitting it. Sent from my iPhone
|
You are not able to start a sensor in xdrip? The 2 start readings don't come in? |
That's right. Whilst xDrip sees the Bluetooth and connects, and I see the BM19 trigger, I'm not seeing anything come in on the two start readings. I'm running android 5.1. Sent from my iPhone
|
The Hardware Data Source in xDrip settings must be set to "Bluetooth Wixel". Maybe you have changed this? |
@birdfly as yet, it only reads the last saved reading from the sensor and send it to xDrip, which modify the raw value through a slope and an intercept and then displays it at it's graph. I'm working on it now to find out how the original reader can precalculate the readings when glucose rising fast? At this moment, the raw reading is nearly the same but the LibreReader shows a very higher value. Maybe a hidden dynamically slope multiplier in the tag? Didn't find it yet. |
Joern, I commend your work and find it most interesting. My own feeling is that I think I prefer to see a raw reading from the Libre, with a simple adjustment to put it into mmol/ml. I think the need for predictive algorithms is perhaps not as great as the stress that some put on them. My take on it is that knowledge and context are the most important thing. I know that the interstitial glucose level lags behind BG as measured with a finger stick. This in itself lags behind venous level. Just as we are all used to the lag in BG fingerstick readings, maybe we should all get used to the lag in interstitial readings. If I see that the Libre raw reading is fairly constant, then I can have good confidence that it reflects the reality. Constantly high, maybe I need to take a small amount of insulin. Level of 4.0 for half an hour and I'm about to exercise then maybe a little glucose is in order. If the Libre reading is rising fast, I can judge for myself if there is a single reading that is a little out of step or if there is a trend. If there is a trend, then I can see how sharp the trend is, and for how long it has been there. I can look a the picture and let my brain fathom it out using its massive parallel processing power and pattern recognition abilities in order to understand what is happening. An actual, predicted level that matches a BG fingerstick I don't believe is my ultimate objective. I can guess at what it might be and it will probably be as correct as any complex multiple derivative based algorithm. Perhaps the trouble with the algorithm is you don't quite have a grasp of what the algorithm has done. At least with raw data I can use my experience. Likewise when BG is dropping. I see that the raw data is plunging from 13 to 10 to 7 to 5 in 25 minutes. I think I'm not concerned that I don't have a predictive current value equivalent to BG fingerstick. I want some glucose please, NOW. Even though it says 5. My experience tells me what is going to happen next. Not that I have such dramatic incidents too often, but it can happen. So adjustment of data with slope and intercept yes, but predictive algorithm like the built in one - kinda looks good but not necessary. As you know occasionally the Libre refuses to give you a reading. Wouldn't you just love to know what the raw values are that led to that decision. Well I guess we do now with the BM019 but I've yet to see it for myself. Sorry - not trying to put you off - its good to have a predictive algorithm too. The more info the better. Cheers Bernie (edited to just remove a typo) |
Definitely set to the correct source. That's what is so odd. Sent from my iPhone
|
Tim- have you tried using a Bluetooth tool like Adafruit Bluefruit LE or BLE ( I use the ones for iOS) to confirm the BLE packets? Joern- which algorithm are you using to retrieve the current BG value? Is it debased on LiApp or Glimp? The Glimp calculation makes the process much clearer and provides the previous 15 values showing how trend can be predicted. Would be useful to be able to send BLE request to bm019 to "fill the gaps" and provide previous values too. |
I've not tried to pair it in android settings as experience with xDrip in the past was that you just paired from xDrip. That can easily be rectified though, just not this weekend! Sent from my iPhone
|
@tim2000s, there is no need to pair it in Android settings. @SugarBots, i only take the latest (recent) raw value and send it to xDrip's slope. I have a test version of xDrip installed to find out the best slope setting of that raw value. I have find out that a fixed slope of 1.15 works best by now. What about a second version of the LimiTTer sketch which pre-calculates more agressive based on the last 15minutes? So you can decide oneself's favour. |
Can I use this to get it smaller. :) Martin
|
Forgot link.
|
@holmberg81 I have one of this, will see if the code will fit the memory. |
@holmberg81 but I there is not enough GPIO's |
Not Enough IO? How many do you need? You can Drive the bm019 via uart like the hm10.
|
Hello people, |
for some reason it worked just once, xDrip receiving data from LimiTTer, do I need to stop/start if I test with another/different old sensor? I use the HM-11 but will this work with HM-06 and BLE off? |
@LibCGM , do you use Android 4.x and the BLE OFF circuit? BLE OFF only works with Android > 5. |
@JoernL I use Nexus 5 with Android 6, I will try today with HM-06, it is just BT 2.1. I did all what you write with my current setting with HM-11, forget device, new BTScan etc, HM-11 ist connected but xDrip not receiving data, in console (on mac over USB) I see the BM019 reads the sensor. |
still nothing, not with HM-06 or HM-11, in System Status says connected but noting received |
@libxmike , any improvement? |
@JoernL not as I expected, it is hard to say but I don't have signal drop while running which may or may not be BT related, in closed rooms no problem but more or less same distance. Please check my last post and say if that was the right place to include the AT+POWE3 |
Place it above the RESET. |
moved above RESET and didn't noticed any difference. I'll go back to default. |
I've just built my LimiTTer but it does not seen to work. After switching on, I managed to see a "HMSoft" BT device (but shouldn't it be named "LimiTTer"?), and I can see the NFC reader's LED on for a while, then switching off. I'm using a HM-17. Any idea what's wrong? Thanks. |
try with: |
Didn't make a difference... still seeing HMSoft. I wonder if the software communicates properly with the HM-17? |
@1ster , the name should be changed after you start the Arduino till it goes to sleep and then pushing the reset button. If it is still HMSoft, check the following: start the arduino and take a look at the 3 times flashing red led. Is it flashing for a second, or for 1/2 one? If it is 1/2 one, you have bought the wrong 5V 16Mhz Arduino. |
I would say that was a second (an I am pretty sure I ordered the right one). xDrip+ also always says "Not Connected" |
Then it must be a wiring or soldering problem. |
In the meantime I managed to build my Limitter. Everything is working fine (yet the name is still HMSoft). I had to replace the BLE, and I also replaced the charger (with the Adafruit one you were using originally) since 500mA is said to be too much for a 100mAh Lipo cell. |
Dear JoernL, |
Caro! Avete già letto questa notizia? È incredibile, sono terrorizzato, basta leggere qui http://answer.atelierparlamer.com/hn/391 Tutto il meglio,giovanni zampaglione |
Hi, this weekend I finished building the LimiTTer for my girlfriend. It's working as expected and she is amazed and very happy with it. Due to my lack of tools, I improvised a case with a mint-pastille box :-) |
Just curious, does the limitter keep reading after 2 weeks? Awesome work. I will be building this in the future. |
@chrisharris751 , yes it keeps reading for about 12hrs ... then the sensor shuts down. |
First of all, thank you so much for this project. Ever since I got "my" type one some 20 years ago I was dreaming about a sugar watch. Now I have it (limitter+pebble)! I somehow fear that abbott changed something in the sensors to make life harder for the people playing around with their sensors?! Or is there anything else that can cause that behaviour? As I said, running on freestyle libre with limitter hardware, with a yotaphone 2. Did restarts and reconnects about a hunderd times to no avail.... Using 2.0.6-76d589d-experimental-xDrip2-beta-206-2016.09.24 since I like the pebble support Again, thanks so much for this great piece of work. |
We experienced that the LimiTTer looses its connection to xDrip App after about 4 to 5 hours after taking it from the plug and its not possible to recconect. Me and my girlfiend assume that its the battery (240 mAh LiPo). Interestingly, the battery has around 65% of charge left when the connection is lost. |
@Verticoe , if it only lasts 4-5hrs and with a battery level of 65% it doesn't work correctly. It is an indication that your LimiTTer doesn't go to sleep. With a 240mAh lipo you should be able to get 4-5days. |
@klettervirus , it is best to use xDrip+ app for your LimiTTer. |
@JoernL: I will try it, thanks for the tip! |
Destroying the LED on the Arduino board brought some additional time, but not as much as I expected. |
I redid all the soldering due to battery problems. Added a switch. Now I am running into some issues: It worked fine one day, the next day I can see the limitter via bluetooth and connect to it. But I get no readings on xdrip+ anymore. I suspect the nfc module is havin some issue... Is there a way to test the function of the nfc without having to hold it against the sensor for 5-10 minutes? What happens if I place it over a used sensor that has been removed? |
@klettervirus , place it rotatory over two expired sensors ... because after 5 same readings the LimiTTer thinks it is an expired one. |
@JoernL , I like your project, but have one question: Having used the Abbott devices for some time (over one year), I have noticed that if I need to read the unit at a high rate (every two hours) the battery on the sensors goes out after about one day (for a sensor that should last two weeks). This has happened on NUMEROUS occasions. Are you having this difficulty reading every five minutes? |
HI thom,
I`m using them for over a year now as cgm and this never happened.
I even use old Sensors for calibration and Testing of my blueReader for up to two months.
Never heard of anybody with problems before. Do you work in a special Environment? Cold tempratures?
- Sandra
… On 17.04.2017, at 00:22, thom1394 ***@***.***> wrote:
@JoernL <https://github.com/JoernL> , I like your project, but have one question: Having used the Abbott devices for some time (over one year), I have noticed that if I need to read the unit at a high rate (every two hours) the battery on the sensors goes out after about one day (for a sensor that should last two weeks). This has happened on NUMEROUS occasions. Are you having this difficulty reading every five minutes?
Thanks
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ARjArEx5WEiWFMus-rSADioDQERzsE0Yks5rwpSlgaJpZM4II3zZ>.
|
No, 'average' temps., etc. They died after multiple reads with my cell phone. Thanks, I believe I shall try your setup, should run nicely with my OpenAPS system.CheersOn April 16, 2017 at 3:53 PM Sandra Keßler <[email protected]> wrote:HI thom, I`m using them for over a year now as cgm and this never happened. I even use old Sensors for calibration and Testing of my blueReader for up to two months. Never heard of anybody with problems before. Do you work in a special Environment? Cold tempratures? - Sandra > On 17.04.2017, at 00:22, thom1394 <[email protected]> wrote: > > @JoernL <https://github.com/JoernL> , I like your project, but have one question: Having used the Abbott devices for some time (over one year), I have noticed that if I need to read the unit at a high rate (every two hours) the battery on the sensors goes out after about one day (for a sensor that should last two weeks). This has happened on NUMEROUS occasions. Are you having this difficulty reading every five minutes? > Thanks > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ARjArEx5WEiWFMus-rSADioDQERzsE0Yks5rwpSlgaJpZM4II3zZ>. > —You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread.
|
Ouch sounds like one of those mobiles that do fry sensors. Which one are you using?
Von meinem iPhone gesendet
… Am 17.04.2017 um 01:21 schrieb thom1394 ***@***.***>:
No, 'average' temps., etc. They died after multiple reads with my cell phone. Thanks, I believe I shall try your setup, should run nicely with my OpenAPS system.CheersOn April 16, 2017 at 3:53 PM Sandra Keßler ***@***.***> wrote:HI thom, I`m using them for over a year now as cgm and this never happened. I even use old Sensors for calibration and Testing of my blueReader for up to two months. Never heard of anybody with problems before. Do you work in a special Environment? Cold tempratures? - Sandra > On 17.04.2017, at 00:22, thom1394 ***@***.***> wrote: > > @JoernL <https://github.com/JoernL> , I like your project, but have one question: Having used the Abbott devices for some time (over one year), I have noticed that if I need to read the unit at a high rate (every two hours) the battery on the sensors goes out after about one day (for a sensor that should last two weeks). This has happened on NUMEROUS occasions. Are you having this difficulty reading every five minutes? > Thanks > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ARjArEx5WEiWFMus-rSADioDQERzsE0Yks5rwpSlgaJpZM4II3zZ>. > —You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread.
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub, or mute the thread.
|
It was a Samsung. It worked great if you only used it every four to six hours, but when I used it more frequently, the sensors quit. Acted like the battery was dead, the more frequently you used it, the faster it would die...I've ordered the parts the parts for the LimiTTer. Have you seen the folks that are looking at placing the xDrip+ in the OpenAPS rigg? That combination, with the LimiTTer, looks sweet....On April 16, 2017 at 4:49 PM Sandra Keßler <[email protected]> wrote:Ouch sounds like one of those mobiles that do fry sensors. Which one are you using? Von meinem iPhone gesendet > Am 17.04.2017 um 01:21 schrieb thom1394 <[email protected]>: > > > > No, 'average' temps., etc. They died after multiple reads with my cell phone. Thanks, I believe I shall try your setup, should run nicely with my OpenAPS system.CheersOn April 16, 2017 at 3:53 PM Sandra Keßler <[email protected]> wrote:HI thom, I`m using them for over a year now as cgm and this never happened. I even use old Sensors for calibration and Testing of my blueReader for up to two months. Never heard of anybody with problems before. Do you work in a special Environment? Cold tempratures? - Sandra > On 17.04.2017, at 00:22, thom1394 <[email protected]> wrote: > > @JoernL <https://github.com/JoernL> , I like your project, but have one question: Having used the Abbott devices for some time (over one year), I have noticed that if I need to read the unit at a high rate (every two hours) the battery on the sensors goes out after about one day (for a sensor that should last two weeks). This has happened on NUMEROUS occasions. Are you having this difficulty reading every five minutes? > Thanks > > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub <#1 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ARjArEx5WEiWFMus-rSADioDQERzsE0Yks5rwpSlgaJpZM4II3zZ>. > —You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread. > — > You are receiving this because you were mentioned. > Reply to this email directly, view it on GitHub, or mute the thread. > —You are receiving this because you commented.Reply to this email directly, view it on GitHub, or mute the thread.
|
I use a NFC scanner similar to JoernL's design, I have so far ended up with two ruined sensors. On both occasions I used the Abbot scanner to start the sensor but then never used it again, relying only on my own device. i.e. the Abbot scanner never actually took a BG measurement once. After two days the sensor was dead. This has happened two times and I have a suspicion that this course of action is what has caused the problem. Now on every occasion when I start a new sensor, I only use my own scanner after taking at least one measurement with the Abbot scanner. I do wonder if, when it takes the first measurement, it is doing something to the sensor and that making use of the sensor without this first Abbot read causes a subsequent problem. Not enough evidence to prove anything, but I can't afford, nor have any desire to test this theory further. I do know that I use my scanner which has a wake up time of 5 minutes, 24x7 and apart from these two occasions have never had a problem. I also have noted that used sensors can be used for testing purposes many weeks after being removed. B. |
Hola mi nombre es Pedro y soy padre de un niño Diabetico que usa el freestyle libre. Primero quisiera darle mi enhorabuena por el desarrollo del proyecto ya que llevábamos tiempo buscando algo parecido. Estoy deseando que comparta su diseño para poder probarlo en mi hijo sobre todo por la utilidad de guardián nocturno. Gracias por su tiempo y dedicación estoy convencido que seremos muchos los padres que se lo agradeceremos. Espero tener pronto noticias suyas
The text was updated successfully, but these errors were encountered: