Shop OBEX P1 Docs P2 Docs Learn Events
ELEV8 XBee Link Issue - Page 2 — Parallax Forums

ELEV8 XBee Link Issue

2»

Comments

  • Got my off the shelf XBEE telemetry kit installed. Did not do Firmware upgrade on the XBEE's.

    Downloaded fresh 1.0.2 from ELEV8 site. Downloaded ELEV8 firmware from C code via SimpleIDE.

    1. XBEE USB plugged into computer.
    2. Plug battery in ELEV8.
    3. Start GS 1.0.2
    4 GS reports Firmware 1.1.6
    5. No glitches for 30 minutes.

    I think I had glitches with 1.0.3 but I will try that next.

  • PublisonPublison Posts: 12,366
    edited 2016-08-21 19:39
    1. XBEE USB plugged into computer.
    2. Plug battery in ELEV8.
    3. Start GS 1.0.1
    4 GS reports Firmware 1.1.0
    5.Stops after 10 seconds, picks up after ~30 seconds, stops for ~30 seconds, repeats.

  • 1. XBEE USB plugged into computer.
    2. Plug battery in ELEV8.
    3. Start GS 1.0.3
    4 GS reports Firmware 1.16
    5.Stops after 10 seconds, picks up after ~30 seconds, stops for ~30 seconds, repeats.
  • I was using the Cpp firmware, most recent, and the most recent ground station software. Every minute or so the telemetry locked for up those intervals as well. Looks like a new issue.
  • PublisonPublison Posts: 12,366
    edited 2016-08-21 22:18
    GS 1.0.2 seems to work the best.
    Downloaded fresh 1.0.2 from ELEV8 site. Downloaded ELEV8 firmware from C code via SimpleIDE.

    1. XBEE USB plugged into computer.
    2. Plug battery in ELEV8.
    3. Start GS 1.0.2
    4 GS reports Firmware 1.1.6
    5. No glitches for 30 minutes.

  • That's bizarre. Between 1.0.2 and 1.0.3 the only thing that changed in the way the GS talks to the FC is that it now checks a serial port flag to see if it's been disconnected, and if so, begins trying to connect again. That should make it *better* not worse (I had issues when I unplugged the FC and plugged it back in - it wouldn't attempt to reconnect, and this fixed it).
  • We should probably make this a different thread since it's not hardware related? Happy to start that or just make it a GitHub issue. Whatever is best. I can also screen cap the behavior, but I'm away at a conference until Friday.
  • geo_leeman wrote: »
    We should probably make this a different thread since it's not hardware related? Happy to start that or just make it a GitHub issue. Whatever is best. I can also screen cap the behavior, but I'm away at a conference until Friday.

    It would be good to keep this thread going. I don't think it's hardware related at this time. The hardware has not changed; the software did not change.

    @geo_leeman, and@JasonDorie,

    I am going to do a spreadsheet with results between WIN7 and WIN10; Firmware SPIN and C, and GS 1.0.1 and 1.0.2.

    Take me a couple of days.

    Where's Kyle when you need him. :)



  • It would be great to see if running the most recent everything it could be reproduced at Parallax. Any chance Miguel et al?
  • Chris SavageChris Savage Parallax Engineering Posts: 14,406
    I did mention this thread to Miguel.
  • Thanks Chris!
  • Hi John I will test that out today and let you know my findings.
  • Kyle M.Kyle M. Posts: 112
    edited 2016-08-24 19:32
    Publison wrote: »
    Where's Kyle when you need him. :)

    Right here!

    I'm sorry you all are having so much trouble with this stuff. If its any consolation, we're hoping to finally release firmware and software v2.0.0 in the next week or two; we're just trying to sort out a few last bugs.
    After much deliberation, we will be more closely adhering to Semantic Versioning for the software, and transitioning to that system for firmware, so as long as the major rev numbers of the firmware and software match, they are guaranteed to be compatible. By that standard, we should have bumped to 2.x.x once compatibility was broken with 1.0 (which happened with 1.0.2 or 1.1 I think), but since that didn't happen then, we're doing it now (especially since we haven't actually publicly pushed out any rev since 1.0). So we are effectively jumping form v1.2.0 to v2.0.0 (which is what you'll notice if you look in GitHub). I'm sincerely hoping that when we push v2.0.0 out, it will alleviate all of the compatibility issues we have been having.

    But back to the matter at hand. Running GS v1.1.1 and FW v1.1.0 on Windows 10, I have not encountered this particular issue I've done half a dozen flights with this same setup over the last week or two, and the only xBee issues I had were caused by a faulty USB cable. About a week ago, I did some quite extensive testing with the latest firmware/software and found (as I was hoping) that I achieved a successful wireless link regardless of the order in which I open GroundStation, plug xBee into computer, and power up the ELEV-8. However, I did find that if the PC-side Xbee is disconnected (then reconnected) from the computer while communicating with the other xbee, then ELEV-8 must be restarted to restore connection. Beyond that, I found no other "quirks" that I thought might lead to problems.
  • Kyle M. wrote: »
    Publison wrote: »
    Where's Kyle when you need him. :)

    Right here!

    I'm sorry you all are having so much trouble with this stuff. If its any consolation, we're hoping to finally release firmware and software v2.0.0 in the next week or two; we're just trying to sort out a few last bugs.
    After much deliberation, we will be more closely adhering to Semantic Versioning for the software, and transitioning to that system for firmware, so as long as the major rev numbers of the firmware and software match, they are guaranteed to be compatible. By that standard, we should have bumped to 2.x.x once compatibility was broken with 1.0 (which happened with 1.0.2 or 1.1 I think), but since that didn't happen then, we're doing it now (especially since we haven't actually publicly pushed out any rev since 1.0). So we are effectively jumping form v1.2.0 to v2.0.0 (which is what you'll notice if you look in GitHub). I'm sincerely hoping that when we push v2.0.0 out, it will alleviate all of the compatibility issues we have been having.

    But back to the matter at hand. Running GS v1.1.1 and FW v1.1.0 on Windows 10, I have not encountered this particular issue I've done half a dozen flights with this same setup over the last week or two, and the only xBee issues I had were caused by a faulty USB cable. About a week ago, I did some quite extensive testing with the latest firmware/software and found (as I was hoping) that I achieved a successful wireless link regardless of the order in which I open GroundStation, plug xBee into computer, and power up the ELEV-8. However, I did find that if the PC-side Xbee is disconnected (then reconnected) from the computer while communicating with the other xbee, then ELEV-8 must be restarted to restore connection. Beyond that, I found no other "quirks" that I thought might lead to problems.

    Thanks for the update Kyle,

    I'll forgo hashing out all the old stuff. since we are going to have "new stuff".




  • Hi everyone sorry for the delay but I finally got a chance to test out the telemetry kit with the ground station software. I tested 1.0.2 and I was getting consistent data for approximate 5 minutes then I got a drop out in data. After 10 seconds or so GS recovered and data started updating again. Version 1.0.3 and 1.1.6 are only semi official releases at best and are considered experimental but it's also important for theses versions of GS to match the same version of FC firmware. One important topic to understand is that the Propeller on the flight controller is already doing so many things at once that the Xbee data slowly starts to lag behind. This compounded with the FTDI chip's buffer not being large enough will eventually cause a drop in communication between an Xbee and the GS. After speaking with a couple of people close to development of this project it's hard to guarantee that communication between the Xbees and GS will not drop out from time to time.
  • PublisonPublison Posts: 12,366
    edited 2016-08-25 18:31
    Thanks for the update Miguel!

    I can't see the XBEE data being used for mission critical events, be cause there has always been a lag, so a dropout from time to time may be acceptable.
  • I'll screen grab the behavior I see when I'm back at home this weekend, but I'd say I have max 50% data uptime.
  • Kyle M.Kyle M. Posts: 112
    edited 2016-10-13 20:57
    Publison, are you still having issues with the xBee when running firmware and software 2.0.#?
  • Kyle M. wrote: »
    Publison, are you still having issues with the xBee when running firmware and software 2.0.#?

    Kyle,
    I'll give you an update on Friday.



  • A comment I posted a short while ago on the GitHub Issue:

    I've been able to re-create (and then fix) the intermittent connectivity issue by changing the CE Routing/Messaging mode setting of base-station xBee using XCTU.

    To see if that's what's causing your problem, try this fix:

    1. If you haven't already, update your firmware to the latest github release.
    2. Plug you base-station xBee into your computer.
    3. Launch XCTU (the xBee configuration program) on your computer.
    4. Click on the "Add Devices' Icon
    5. Select the COM port corresponding to your xBee, and click "Finish". XCTU should automatically find and add the xBee to your radio modules list. (If it doesn't, give me a call at 916 625 3003 and I'll walk you through recovering the module)
    6. Click on the module listing below Radio Modules . Wait for XCTU to load all of the module's settings.
    7. Scroll down until you find the setting for BD Baud Rate, and double-check that it's set to 57600.
    8. Next, scroll back up to CE Routing/Messaging and select Non-Routing Module from the drop-down menu
    9. Click the pencil icon to the right of the Routing/Messaging setting or the Write icon in the top bar to save your settings.
    10. Once XCTU confirms it has successfully made the settings change, close out of the program, and try connecting to your ELEV-8 again. (note that you should only be changing the base-station xBee to a Non-Routing Module, leave the xBee on your ELEV-8 as a Standard Router.

    Let me know if that fixes it.
  • Getting back late, Internet, TV, and Phone were down today, what's up Comcast? :)

    Refresh everything up to Firmware 2.0.1 and Ground Station 2.0.1. Upgraded XCTU to 6.3.2. Incorporated to change you suggested to
    changing the base-station xBee to a Non-Routing Module
    .

    It has been running for an hour with no lock ups.

    Good job narrowing that down!
  • geo_leeman wrote: »
    I'll screen grab the behavior I see when I'm back at home this weekend, but I'd say I have max 50% data uptime.

    Did you get a chance to check out the latest (2.0.1)? Plus Kyle's post on reconfiguring the PC side XBEE.

    Seems to work.


  • @Publison Yep we've been conversing over at GitHub about it. I'm going to make the changes tomorrow morning and see how it goes. Hopefully that solves it and the online instructions can be updated.

    Kyle - Did this show up due to changes in the XCTU software?
  • Good news - I ran successfully on Windows and Mac. Very rarely the tablet would drop for a few seconds, but probably once every 5 minutes at most. Mac ran for quite awhile with no drops. I'd say that did the trick.
  • JasonDorieJasonDorie Posts: 1,930
    edited 2016-10-15 06:12
    For those like Publison who may not be following the GitHub thread, there are a few more settings you can change to improve latency and throughput mentioned here:
    https://github.com/parallaxinc/Flight-Controller/issues/40#issuecomment-253965154
  • I'm following it now! :)

    I implemented your upgrades, and they seemed to work fine. (Gives me a chance to advance in learning C). Twenty minutes with no dropouts.

    I do have another problem, not related, which I will start in another thread.
Sign In or Register to comment.