Steve I have had this happen to me but the graupner lady always says "GPS error". I have flown in some locations where the RF spectrum was really crowded as far as cellular and other wireless signals go and it just switches to manual mode. Have you upgraded to the latest Nav firmware? Not that it is related but it does fix an important bug where it stops datalogging if it gets certain GPS or compass errors.
FWIW In the full scale world of aviation where live's are on board the aircraft we have multiple GPS receivers as well as RNAV receivers that are all talking to one another to verify that position is within acceptable parameters. These systems undergo a rigorous certification process and are subjected to RF shielding studies. Once installed on a jet, they are tested and re-certified by trained electronics technicians on a regular basis. If for instance the GPS receivers don't agree then we get a "GPS position disagree" message in the flight management computer display and it restricts our use of "GPS only" navigation. While extremely rare, it does happen and when it does happen it is typically in the approach environment or on the ground... (read ... near other sources of interference) Of note as Gary already pointed out... these are nothing more than off the shelf electronics that have had no (as far as I am aware) RFI shielding studies performed and are subject to virtually no accountabililty... "fly at your own risk" We all fly with different levels of proficiency and acceptable risk levels commensurate with our own personal experience or lack of Events like Justin's serve to illustrate how important it is to have a solid understanding of our flight control system, nav system and power systems... not an understanding of just how to operate them ... but more importantly "how they operate". The more systems we attach to our aircraft the more complex the equation becomes. I leave my XBEE off the bird unless I'm using it for a particular reason such as tweaking the CG. I can only imagine how you must have felt in those few moments Justin. I'm sure your situational awareness goggles narrowed to the size of a soda straw as you focused all your attention on being reactive and simply trying to gain some sort of control over your bird. I'm glad you were able to semi control things to avoid a potentially catastrophic event.
Scott, thank you for the info.. Very interesting.. Out of curiosity, is it true that a lot of airlines are using GPS, etc to land the planes with no pilot input?
Very end right as I landed and shut off the motors. Flew several flights today with all the same gear and no errors at all.
Justin, lots of airliners are incorporating GPS navigation technology...none that I am aware of are using GPS or any other technology (except the DOD) to land aircraft with "no pilot input". The autopilot has been capable of landing the aircraft for years (prior to GPS) technology. Industry "best procedures" have revolved around the idea that in very low visibility.. below 1/4 mile that it's probably safer to have the human monitor the performance of the auto pilot and the approach coupler to intervene (manual mode in MK speak) should things go south. The GPS has enabled us to fly approaches thru mountainous terrain where the final approach is a curved path rather than a straight in.
Scott is perhaps too modest to say that he is a senior airline captain and therefore knows whereof he speaks. Do a Google search for a pilot training video called "Children of the Magenta" (the magenta line is the line on the display that shows where a commercial aircraft is going to go in the sky) -- but it's a metaphor for "all the automation in the plane." The theme of the training is, do not rely on the automation. Learn to fly the copter "naked" -- no PH, no AH, just the bare minimum. As I mentioned to Justin in a Conversation, AeroSimRC now has a nice realistic Cinestar 8 in it, plus it has FPV and "free camera" so you can say, observer your ability to fly a figure eight from directly overhead. It's quite revealing if you *do* turn on altitude hold and then fly figure eights in an orientation other than nose out....and watch from overhead. Ugh. (That last ugh was a comment about the last time I tried nose-in last week. Not pretty. But I did have the wind and turbulence set to max! Andy.
Justin .. thats not a stupid question at all. The autopilot systems on modern airliners are a compilation of very sophisticated computer systems all talking to one another. These computers are fed information from various sensors and probes on the aircraft. Some are decades old and relatively simple technology like the pitot and static system providing ram air pressure and static barometric pressure. Others are fairly sophisticated such as the RNAV or GPS systems, laser gyros etc. Data from all of these sensors are fed into aid data computers and flight management and guidance computers that triangulate a position, speed, altitude etc. The autopilot system is then programmed by the pilot to fly a certain path, altitude, speed etc and communicates with the various flight controls and power plants to effect the commanded change... or to simply fly a commanded altitude, heading, course, glide path etc.
Are you sure its the end of the flight? Without the new firmware the datalogging stops when it sees GPS errors.
Justin: Here's what Holger has to say about the GPX file: it was the compass value.As you can see in the screenshot [which I've posted below AJL] , the value jumps about 50-60° up and down right from the beginning.The GyroCompass is the (kalman-) filtered value and the raw-value is usually inbetween an error-range of +-5°.Short disturbances (below 1-2seconds) are normal and perfectly filtered by the algorithms, but here the value was disturbed too much and too long.After some minutes flight (around 260seconds) both values didn't merge, because the error band became too big.When the absolute error of the GyroCompass is bigger than 30°, the MK starts to circle.The magnet strenght itself wasn't too bad.There seems to be an influence by an alternating magnetic field.Using an external compass would solve the problem probably. Hope this helps. Andy.
Fascinating...great find. I have been procrastinating about installing my MK external compass, as I have a 2.1 Nav board which requires the leads be soldered. But I guess this is a Very Good Reason to do it now.
Steve: I'm still puzzled. True, we know the compass (according to Holger) was the issue, but why, I wonder? From my reading of Holger's reply, it's as though there were issues with the compass all along, and it took a while for the filtering algorithm in the code to allow the compass errors to manifest. But why would those compass errors suddenly cause the copter to toilet bowl or roll off to one side? My understanding was the compass plays a role in reducing uncommanded yaw motion... Last time I looked, I don't think MK publishes the source code for the NaviControl board -- at least not all of it. Andy.
Andy, great info! I think when the flight started it already had the issue.. Watching the video you can see after takeoff it's doing small circles.. My guess is it just got worse as it went along.. So, let's say I turned GPS off, would that have solved it? Is it the compass telling the PH what to do?
Yes, the compass is not required for "manual" flight. As a matter of fact, you can disconnect the Navigation control board altogether and still fly it just fine manually.
Hey guys, just curious, I have a Lipo battery that has a dent in it, it has not expanded, all the cells charge perfectly. Is there any reason I should not fly with this battery?
Justin a photo would help. Generally if you see physical damage I would retire it for ground testing only.
While calibrating the Compass for the first time no one warned me that the velcro straps aint gonna hold it from the NICK North calibration! Not sure why they would not include another strap so that doesn't happen.. All my batteries now have VELCRO on the bottom that attaches to the battery plate, and a handyVELCRO tie that Jose gave me that keeps it from falling out in the NICK.