/dist/images/branding/favicon

MK GPS V3

Discussion in 'Electronics' started by Adam Paugh, Sep 4, 2014.

  1. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,981
    Likes Received:
    807
    Good point, Jose Luis, especially on Xoar props (which are quite soft).
     
  2. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,383
    Likes Received:
    1,164
    I think you may find that you're confusing *magnetic* interference with radio frequency interference. Normally magnetic interference (the "ferrous" nature of the location) should not interfere with GPS (but will mess with the compass) -- but high intensity radio frequencies (e.g. cell phone and other communications towers, WiFi, etc). can certainly "swamp" the GPS signals.



    Just to restate: magnetism as such does not interfere with the GPS. It will interfere with the compass.
    The GPS is prone to radio frequency interference. It's operating at around 1.5 Ghz and 1.2Ghz if I'm recalling the frequencies correctly.

    Andy.
     
  3. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Jose,

    I use em just for that reason. Forces me to inspect the props every time and make sure they are in good repair. I used to do lock nuts but sadly my drill walked away on a job. Once I get back home a drill and some Nylocks are on the list.

    Andy,
    The magnetic reflection of the area does cause some interference. Without the shield, I have no issue and maintain GPS even on takeoff.

    Steve,

    Reno is great=) Tons of work here and in Tahoe and our city council loves the UAV's.

    Dave,

    Got it solved. Shield was causing some issue with the unit. Research partner thinks it was "trapping" EMI under the unit
     
  4. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,383
    Likes Received:
    1,164
    Ryan: Wow....based on what you're saying you've got a serious problem with the GPS unit or the heavy current wiring on the copter is radiating way too much radio frequency interference up in the 1.2 - 1.5 Ghz end of the spectrum.

    Secondly, the GPS shield is supposed to enhance the ground plane around the antenna (and shield the antenna from RFI emanating from the MK boards). It sounds like the GPS shield is making matters worse -- so I suspect there is a fundamental issue with the "grounding" on your electronics and the GPS shield is neither acting as a shield, nor as a ground plane.

    See what MK has to say about the GPS Shield here: http://www.mikrokopter.de/ucwiki/en/GpsShield and look at the relative performance with and without the shield that they show. With the shield (as it states) you should see a 10dB improvement in Rx signal strength.

    Clearly you've got one sick GPS unit if the GPS shield is making things worse.

    Can you confirm that you have soldered all the seams for the four corners of the GPS shield, and all edges are soldered around the GPS board? It has to form a "solid" ground plane/shield.

    By the way, when I use quotes around "grounding" it's because the copter cannot have a true ground -- I just mean the negative rail of the LiPo and everything that connects back to it. There could be some resistance on the negative rail between the GPS unit and the LiPo that's causing the negative rail at the GPS board to float above LiPo negative. Alternatively you could have noise (electrical, or radio frequency noise) leaking into the GPS unit along the negative rail.

    Andy.
     
  5. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Andy,

    I did shorten the wires up on the lead into the board, Been meaning to do this for a while but with the amount we have been working and the issue only here in Reno, it was not high on the list. Reduced the field strength by about 20% around the wires.

    The solder points were solid and done correctly. I followed what the guys at QC showed me for building and installing the GPS. I am planning on tossing the shield on the 2nd gps I have for the Octo and seeing if it is in fact the board. W/o the shield I do lose a bit of signal strength, but the signal stays solid. When I re-solder it tonight I will post some pics.

    One thing I found was the 6pin SPI cable had got itself under one of the larger caps on the BL3.0. Through just flying and the minute vibrations it wiggled under the cap and was touching it, and the EM field on that cap is about 200 gauss during takeoff, and less then 100 in hover (EM on the wires with signal in them is 10 gauss). That has been rectified and the cables can no longer get near that cap. I will be testing with and without shield to see if that was part of the issue. I think this may be the main culprit so I am going back to test everything the way it was, minus this.

    The "grounding" looks good across the board. If there was an issue with that I would tend to think on any time I throttle up and do a rise I would lose GPS.

    The other thing suggested to me was the shield COULD act as a collector for EMI. In such that field gets "trapped" in the shield if it was not soldered correctly and then rapid dissipation of the field returns GPS. This was a bit farfetched but I was shown that it is possible for EM fields to get "Trapped" in or around antenna. I HIGHLY doubt this is the case but if it is...I owe that man a beer or two.

    I also just ordered a Mk3 GPS as I want to see how it performs vs what I have now. Beta testing ho..
     
  6. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,383
    Likes Received:
    1,164
    Can you put an ohmmeter on the GPS shield's soldering and the negative rail and see if there is measurable resistance -- assuming you've got a VOM that will measure low resistance (and use a low voltage from the internal battery)? I'm on vacation right now so don't have access to aircraft, test gear, etc.

    Andy.
     
  7. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,981
    Likes Received:
    807

    Sounds more like a scene from "Ghostbusters"...

    [Holds up the smoking ghost trap]
    Dr Ray Stantz: Sir, what we have here is what we call a non-repeating phantasm, or a class-5 free roaming vapor, real nasty one too.
     
  8. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Andy,

    I think we have a multimeter that will go down to 3.3v at the lab.

    I tested the octo GPS with the shield. No problems. In fact, I got 16 sats at one point. I then swaped to the hex GPS, with and without shield. Nothing. Swapped Navi boards. nothing. Would have done FC's but I forgot my laptop. So I put everything back to normal and shield on the hex, no issue. I think it was related to that SPI bus cable getting under and touching the capacitor.
     
  9. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Better then what I was thinking!
     
  10. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,383
    Likes Received:
    1,164

    In which case, the AC voltages on the cap leads would have been inducing interference into the SPI bus and that might have been the cause -- EMI causing RFI.

    Andy.
     
  11. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    That was exactly what I was thinking. The thing is I can't figure out how they got past the 25mm plastic standoff that was in between the Cap and where the 10pin/6pin run.
     
  12. Janne Hoglund

    Janne Hoglund Member

    Joined:
    Oct 4, 2012
    Messages:
    364
    Likes Received:
    23
    Hi Adam, any interesting news about the GPS v3 yet?
     
  13. Dave King

    Dave King Well-Known Member

    Joined:
    Dec 24, 2012
    Messages:
    2,712
    Likes Received:
    311
    Not Adam but Quadrocopter posted they are seeing 16 satallites with this new board.
    https://www.facebook.com/#!/quadrocopter
     
  14. Jose Luis Ocejo

    Jose Luis Ocejo Active Member

    Joined:
    Sep 20, 2012
    Messages:
    581
    Likes Received:
    44
    Hi Adam any ETA on the GPS V3
     
  15. Ryan McCrae

    Ryan McCrae Member

    Joined:
    Oct 9, 2012
    Messages:
    120
    Likes Received:
    4
    16 sats would have a noticeable impact on how accurate gps hold is wouldn't it? Im normally seeing 8-10 here in NZ.
     
  16. Steve Maller

    Steve Maller UAV Grief Counselor

    Joined:
    Oct 30, 2012
    Messages:
    3,981
    Likes Received:
    807
    I'm not sure that's the case. What is true is that it'd make GPS Position Hold less prone to dropouts due to interference or obstructions.
     
  17. Mark Harris

    Mark Harris Member

    Joined:
    Sep 29, 2014
    Messages:
    30
    Likes Received:
    3
    RF shields do not "trap" anything, its not even remotely possible. The rf shield will return any energy it absorbs to ground, if there is a voltage difference between your rf shield and your batteries ground.. you have a major grounding issue in your copter and that needs to be solved ASAP before it takes all your sensors and gives you wild readings everywhere.

    Do not remove the RF shield on any RF device, it is there for a very very good reason. It is likely to be illegal to remove the RF shield on any FCC certified device as its part of the certification and likely to cause the device to emit interference beyond the allowable limits.

    GPS runs on 1575mhz (L1 C/A.. what we get to use as civilians), a 1.2ghz video transmitter is more than capable of taking out the GPS even if it is not near by - if it's bleeding RF back into the power supply (which it will be, its just a matter of how powerfully) this can take out your GPS signal faster than pretty much anything else. A ferrite ring with the GPS cable wrapped 4-6 times through as close to the GPS as possible it will remove the majority of the AC (ie: RF) on the lines. Remember, all cables are antennas - they can pick up interference and/or radiate it.. filters are great things to have on everything.

    You may also have a really unclean power supply going to the GPS. GPS units are *extremely* sensitive rf devices, as are the microprocessors on them computing satellite positions. They do not like dirty power in any way shape or form. Stick a 220-470uF 10V+ rated electrolytic capacitor on the cable as close to the GPS as possible. The short leg on the same side of the body as the black stripe (or white stripe if the body is black) gets soldered to your ground wire (black), the long leg to the voltage line (red). Make sure you cover them with heatshrink or hot glue, if they touch together you'll have a bad day. This can also help take care of any RF getting into your gps too... but an inductor (as above with the ferrite ring) is much better.
     
  18. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    The problem was an SPI cable getting trapped under a capacitor that had via vibration worked its way under the cap. The result was as Andy had predicted, RFI via EMI (and tested) to a point were this could take a kopter down or cause a "fly away" like scenario.

    Just broke all records on the hex too flying some live stuff with the V3 GPS. 26:25 in one location GPS hold with movi M5, Canon XA 25 and Teradek Bolt 2000.
     
  19. Andy Johnson-Laird

    Andy Johnson-Laird Administrator
    Staff Member

    Joined:
    Jul 31, 2012
    Messages:
    10,383
    Likes Received:
    1,164
    26:25? Nice.
    Gives you time to break out a sandwich and a cup of coffee for in-flight catering.

    Andy.
     
  20. Ryan McMaster

    Ryan McMaster Active Member

    Joined:
    Mar 8, 2014
    Messages:
    289
    Likes Received:
    40
    Moving around we are seeing about 23:00 with that setup, the A7s M5 and Teradek clip are going out for a hold test in 15 minutes (this setup does 24:30 moving), I expect it to break 30 minutes easy in hold with the lighter weight of the a7s and clip (Bolt 2000 is about 1lb). On that job at on point I did let off the controls to drink some water and watch my camera man do his thing. I was even able to answer questions and hand out a few stickers/tshirts while our spotter went to lunch. We did 8 of these flights, never saw a motor over 45c and all the gear functioned flawlessly. Average amp draw per motor was around 6.7-7.1 with 15x5 xoars and all the flights were 20% battery remaining (running paired 10k QC batts).
     

Share This Page