Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
    nigelkeirross
    @nigelkeirross
    @keencave Thanks for the explanation. Will stick to the current version. Where can I learn more about the OOP algorithm?
    Patrick Sonnerat
    @psonnera
    Anybody knows if xDrip+ Wear support was ever developped for LimiTTer? (Original version not RFDuino)
    keencave
    @keencave
    @psonnera Wear has to support orignial LimiTTer protocol for that. As the wear protocol engine is identical to the xDrip+ engine on source code level it should run. But better to look into the wear sources to crosscheck before testing.
    Patrick Sonnerat
    @psonnera
    :+1:
    nfry321
    @nfry321
    Hi All, I'm attempting to build a LimiTTer device but using an Arduino nano 33 BLE as it has the Bluetooth built into the board. The problem is I can't find any info on what format I need to send the data in to get xDrip+ to understand it. The nano 33 BLE uses the ArduinoBLE library and you create Services and Characteristics which can be updated with glucose level etc. There are standard descriptors for these and they include several for CGM (https://www.bluetooth.com/specifications/gatt/characteristics/) so I am assuming xDrip+ should be able to understand these, but cant find out if this is the case or what the exact data format for each one is. Any ideas?
    keencave
    @keencave
    xDrip+ understands two formats for data. First the native LimiTTer format which is an ASCII string containing time, BG and battery level. Second is the old xbridge2 format which emulates an old wixel device. Further information could be find here: https://github.com/savek-cc/wixel-sdk/tree/master/apps/xBridge2. For the native LimiTTer stirng please refer to the sources: https://github.com/keencave/LBridge/blob/master/LBridge_Arduino_V09/LBridge_Arduino_V09.ino#L2680 or refer to the original project from @JoernL .
    nfry321
    @nfry321
    Thanks for that info. For now, I am just trying to use the original hardware, but when I upload LBridge_Arduino_170716_1950.ino I get the error 'BODS' was not declared in this scope. What am I missing?
    keencave
    @keencave
    Which Arduino IDE version are you using?
    nfry321
    @nfry321
    1.8.10 on windows
    keencave
    @keencave
    Please try 1.8.9. The BODS declaration is used in a function which is programming the watchdog registers.
    timocm
    @timocm
    U
    Marek Macner
    @MarekM60
    @gui-dos yes. Sensor RF buffer for NFC is limited to 3x8 bytes of data
    Cybereu
    @Cybereu
    Hi, im orthopaedic surgeon with DM1 with 0 knowledge
    0 programming skills i mean.
    I would like to build one Limitter for myself
    Althought
    Cybereu
    @Cybereu
    I use linux as my main OS, and i have a brother who knows C, python, etc and can help me with all needs.
    Where i should start? With official hardware or another ? Thanks
    A
    dabear
    @dabear
    Easier to buy a miaomiao or bubble or droplet device id you dont have any experience
    If
    keencave
    @keencave
    You have to choose between 1 or 2 months of work and hardware costs of aprox. 60,- EUR for a self built LimiTTer or buy a "commercial" transmitter for about roughly 170,- US$ (= 155,- EUR) which is waterproof.
    tzachi-dar
    @tzachi-dar
    t1devel
    @t1devel
    Could you please provide the sourcecode of L2read? I got thousands of sensor scans on my hard drive I'd like to analyze.
    davidl567
    @davidl567
    Hi, first i am really happy to read your comment on thats community. I am begginer in that world. may i have your comments on this device , and do you now other people that do that. Or can you give me a direction on a other room that talk about that ? Thanks ( and you are hot guys ! Believed i was the only crazy diabetic in this world...) https://youtu.be/qwzOXXWGJxM
    dabear
    @dabear
    That device is unrelated to this channel.
    Imho you should not use diy pumps. There is a lot of safety engineering and certifications on omnipod, medtronic pumps etc that the diy pump neglects
    keencave
    @keencave
    Please dont follow that way. It is much too dangerous to build a pump for your own.
    davidl567
    @davidl567
    Thx
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave I fell over recently, and damaged a limiTTer that I had built about 6 months ago. I resoldered the obvious broken wire, but the limiTTer is not sending readings to xDrip+. I connected it up to the Arduino dev tool to get a trace. It seems that after entering the main loop, I get pairs of messages: " poll SetProtocol_Command not successfull" & then "Protocol Set Command FAIL" repeated 3 times, then "NFC reading not successful" & "NFC read error, wait 3 s". This repeats 3 times & then goes to sleep. Do you think it is the BM019 or the HM17 that could be damaged? I have visually checked the wiring & all looks good.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Having looked at the error messages in the code, it seems that this is the BM019 failing. How can I tell if it might be a bad connection, or a bad device? Thanks.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave I have replaced the BM019 & now working again. I wondered if you have a diagnostic sketh that can be loaded onto the adduino to do a basic functionality check of the limitter hardware?
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave I am building another limiTTer. In my hardware configuration, the wiring would be slightly simpler if I sawapped the arduino pins 10 & 13. If I made the appropriate changes in the code (eg const int SSPin = 13; & const int SCKPin = 10; would it work if I then changed the wiring?
    keencave
    @keencave
    Yes! It should Work.
    keencave
    @keencave
    There is no special diagnostic code as there are only two boards attache: BT and NFC. The error codes should indicate whoch subsystem make problems.
    Nigel Ross
    @NigelKeirRoss_gitlab
    OK, thanks for both replies. I was thinking that a diagnostic to tell if connectivity was broken, or if the BT or NFC boards were not working correctly. But, as you say, probably not worth the effort.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Missing pin assignments? Hi, having swapped the pins as described above, I realised that I could go much further. I could eliminate much more wiring and directly connect all the BM019 pins to arduino pins by overlapping them (perhaps metal pins to connect vs wires), if I hanged to arduino pins connecting to the BM019 from [7,8,13,11,12,10] to [2,3,5,6,7,8] respectively, and 9 remaining the same. (I would use arduino pins 10-13 to connect to HM19). But I would need to change all these assignment in LBridge. However, in the cose I can only see assignments for pins 2,3,7,8,9,10,11,13. ie 5,6 & 12 are missing. I also noticed that pin 4 is assigned, but this is not connected (at least in the original joern limitter hardware schematic that I have been using). Could it be that pins 5,6 & 12 are assigned in some lower level software library? If so, how is it possible to modify those, or would that break stuff? Thanks
    keencave
    @keencave
    Let me check on the weekend. I have to look into the schematic again.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Hi, I have built a LimiTTer with pins 13 & 10 swapped, both in the code, and which pin on the BM019 they connect to. First it powered up (without battery connected) OK & I got the correct name on BTscan from Xdrip+. But no initial reading - the trace message was Battery low! LEVEL: -25% repeated 3 times, then "calibrate PWR DOWN sleep timer (2048 ms) ..." & "calibv = 1026.21, tt1 = 54712, tt2 = 56651", but then "poll SetProtocol_Command not successfull" & "poll Inventory_Command not successfull" 3 times, sensor out of range etc. So I soldered the battery on, and re-tried. Now it just loops with "AT+ command not processed, we have to reset the Arduino ..." and resets. Also the LED on the arduino next to pin 9 comes on and stays on. I can't find out what this means?? It looks to me like I have probably damaged the HM17, but before I replace I wondered what the LED indicated, and if there is any chance this could be caused by the 10/13 pin switch? Thanks
    keencave
    @keencave
    @NigelKeirRoss_gitlab Sorry, I am busy today. Will look into that topic later. The HM17 is very robust. Better check the cabeling before replacing components. The batterie warning is suspicious. Please check the voltage accordingly. The poll errors indicate a connection problem with the NFC chip. The onboard pin 9 LED should be shut off short after reboot. But I am not sure as I habe removed that LED with pliers.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Thanks. I changed the jumper on my FDTI progammer link, and when I rant again with the battery disconnected, it got to the "entering main loop; ... calibv" stage, but now NO battery warning level messages. However, still get the polling error. Repeating same with battery attached behaves exactly as yesterday. In both cases pin 9 LED still on, which is unlike all my other limitter experience. I have rechecked all wiring & seems OK, so my suspicion is that me new wiring for BM019 (swapping pin 10 & pin 13) may be to blame, as they are not identical (pin 10 can do pwm, not pin 13 etc). So I think I will re-wire & reload "original" code...
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Switched the wire/code back to the original, and a) the poll error messages go away, and so does the Arduino led on, and so far (2 readings), it seems to work. So, the switching pins idea doesn't work. Thanks for your help. It would be still interesting to understand about the references to pin 4 in the code, but it is not connected in the hardware...
    keencave
    @keencave
    Ok, correct, pin4 it is not in the schematic. But for feeding enough power to the BM019 the three pins 4, 7, 8 has to be connected. They all deliver 10 mA and that is the longterm save solution. Using only one port the BM019 would use more than the maximum current from one port. We know that a LimiTTer will run with but the system could break after a certain time. There fore please connect these three pins. When you reassign the pins you have to connect other three ports for that purpose. I wasnt aware that the original schematic has a fault here.
    But I htink you made all things right with reassigning. I guess you made some errors on that. Please try it again, either HM17 and BM019 should be very robust.
    And for more informations you should go back in the gitter chat 2 years. There is much more information written there for power consumption, reducing power for LimiTTer and so on. To ensure backward comatibility we havnt changed the wiring all the time. It may have made somehow diffrent (like your approach) to enable better/small form factors. But anyway, compared to the MiaoMiao, Droplet ... devices these days the LimiTTer ist still astable and reliable platform.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave OK thanks for following up. My latest one is now up & running on my arm, having fitted into the 12.7 thick case. I now have 5 versions, and all are missing 4 being connected, and all have 8/7 only connected to BM019 pin 9. 2 of them are about 2 years old, and are used as backups when the more recent ones (300mAh batteries) break. These faults have mainly been that the name of the device changes to "HMSoft", and no longer gets readings (in these cases I let the battery drain down & then power up after a few days) & all is fine again. 2 others (12.3 mm) have the 300mAh battery, have been working since October, unless I fall over & physically break something (which I did to bboth in January). I wonder if the HMSoft thing (which used to happen a lot when my phone was android 4.x) could be related to power. The HMSoft fault usually happens about once every 6 months, but happened twice in January while I was skiing. Maybe the cold meant that the cmos circuits were taking more current, and that triggered the fault if the power supply was marginal. I will go back on this gitter chat & see what I can learn.
    @keencave I am reluctant to have another go at the switched 10/13 wires, since it was only by changing that back to the original & re-loading original code that my most recent version worked. Maybe there is also some underlying library that uses it, or the difference in the pins 13 & 10 are significant in some other way if it isnt PWM. Since it also fixed the pin 9 LED issue, it does seem wrong. So far I haven't been able to find a datasheet that explains what the LED indicates.
    keencave
    @keencave
    According to the Pro Mini description The LED is internally connected to pin 13. This could be a reason why the BM019 SCK pin is not driven correctly. Please remove the LED with a plier.
    Nigel Ross
    @NigelKeirRoss_gitlab
    @keencave Thanks for explaining that, it sounds like to root cause. I think I will breadboard that configuration to test it.
    birdfly
    @birdfly
    do you want to read single block or muti blocks ?