Beautiful day here in Rome. Air was crisp but not freezing with a slight wind. After having made the appropriate adjustments that my brethren suggest we took it out for a second attempt. Very nice and quite smooth. All involved were and big smiles to see the beast rise off the ground without incident. Didn't 't work on the gimbal portion because for some mysterious reason my BEC stopped working. I still have to sort through some of the video issues. Problem though. When we flipped on the GPS position lock the bird wanted to take off across the river. I mean it tore away quickly from it's position. Not good. Altitude hold worked well. Did not try Come Home because of the position hold issues. MK Tools was showing 8 satellites. Any ideas?
Hi Mark Did you have a GPS fix before you started the motors so that the "Home" position was set? With 8 sats you should have had a GPS fix. Also, if you have a GPX file recorded on the MicroSD card in the Navigation Control Board, could you upload it so we can analyze it, please? Andy.
Just got in from a night out on the town. I'm way too old for this shit. I'll upload the file in the morning. Thx
Any here the GPX files. What program do you use to read these oh great God of Forensics. Maybe we took of to quick after we set the gyro's? Side Note: Smartest thing Apple ever did was rewriting their OS in UNIX.
I am merely a mortal, but it seems I woke up before our resident deity, so I had a go at it. It's called "MKGPXTool" and you can read all about it here: http://www.mikrokopter.de/ucwiki/en/MKGPXTool I loaded a couple of these in, and here's what I saw... Are you running a 4S system? If so, you're running at a pretty low voltage. A fully-charged 4S LiPo should read 16.8V or so. Yours are reading way lower than that, and that could be dangerous. The GPS info in the files looks OK. You can see the route using only Google Earth by opening any of the GPX files and zooming in. Play with the visual settings, and you can see cool stuff like this: Otherwise, I don't see why it wanted to fly away. I ran a replay of your flight using the OSD function, and it did look like there was an escape attempt.
Actually this mortal was up early while it was still quiet and shooting video, so just taking a break now. MK_GPXTOOL is certainly the tool I prefer especially if you have FPV with OSD footage as it allows you to sync that with the GPX file data so you can see what you were doing to the copter. There's also MK_GPX (Mrs Google knows where to find it) and GPX Viewer -- which Holger recommends. <Geek mode> The GPX files are just plain ASCII text using a simple markup language, so if you feel adventurous you can also just read the raw data with a text editor</Geek mode> Mark: On the first GPX file GPS00004.GPX, you have a good GPS fix, <trkpt lat="+41.8590211" lon="+12.4678435"> but the battery is showing 13.9 volts: <Voltage>13.9</Voltage> I see Position Hold coming on quite late in the flight at Trackpoint 82 (the one second log file entries are called Trackpoints), so 82 seconds into the logfile/flight. The actual track point data is below the main body of this posting. When the PH comes on, the GPS log entry for 14:23:43 GMT, shows that the GPS does indeed have a home position set. (I suspect it is the position shown in the paragraph above. The voltage is now down to 13.5 volts. When PH comes on, the bird heads north towards the Tiber instead of west back towards you. What I cannot see is why it did that. Unfortunately, the actual home position is not logged but has to be inferred from the start of the flight log. EDIT: Not sure what I was thinking when I wrote the above as it's written as though Come Home was enabled, but I think I was pondering whether the bird was executing Come Home instead of PH. It's not -- it's Position Hold -- so the bird should have basically stayed put in the sky.. The compass data looks OK, so I presume you calibrated the compass for that location? Copy and paste this into Google Maps: +41.8590211,+12.4678435 (Looks like a nice part of Rome). I agree with you re: OS X. Darwin was derived from NeXTSTEP, BSD and some other FSF projects. Thank goodness. A Real OS. I'll have a quick look at the other GPX files and see if anything seems odd. Andy. TRACKPOINT 82 : File GPS00004 <trkpt lat="+41.8589992" lon="+12.4679083"> <ele>5.930</ele> <time>2013-01-05T14:23:43Z</time> <sat>10</sat> <extensions> <Altimeter>87,' '</Altimeter> <Variometer>2</Variometer> <Course>192</Course> <GroundSpeed>75</GroundSpeed> <VerticalSpeed>2</VerticalSpeed> <FlightTime>84</FlightTime> <Voltage>13.5</Voltage> <Current>42.3</Current> <Capacity>1106</Capacity> <RCQuality>195</RCQuality> <RCRSSI>0</RCRSSI> <Compass>352,355</Compass> <NickAngle>000</NickAngle> <RollAngle>000</RollAngle> <MagnetField>108</MagnetField> <MagnetInclination>61,03</MagnetInclination> <MotorCurrent>35,68,34,88,55,83,35,71,0,0,0,0</MotorCurrent> <BL_Temperature>35,51,46,65,52,58,39,45,0,0,0,0</BL_Temperature> <AvaiableMotorPower>255</AvaiableMotorPower> <FC_I2C_ErrorCounter>000</FC_I2C_ErrorCounter> <AnalogInputs>32,29,33,33</AnalogInputs> <NCFlag>0x91</NCFlag> <Servo>128,128,0</Servo> <WP>----,0,0,0</WP> <FCFlags2>0xc3,0x18</FCFlags2> <ErrorCode>000</ErrorCode> <TargetBearing>000</TargetBearing> <TargetDistance>0</TargetDistance> <RCSticks>3,2,-2,-29,0,100,126,0,1,1,1,1</RCSticks> <GPSSticks>-11,20,0,'D'</GPSSticks> </extensions> </trkpt>
MarK Flight #2 (GPS00005.GPX) looks normal. No error flags (not even low battery). I see that you tried PH twice, just for a few seconds at Trackpoint 33 and 43 respectively. In both cases, the bird headed north towards the Tiber. It really flew quite fast. At TP 32 it's showing a speed of 21 M/s, then PH comes on at TP 33, and by TP 34, the speed is 72 M/s. Sorry, but nothing leaps out to suggest what is going on here.... Andy.
Mark: For flight #3 (GPS00006.GPX), I can see that you tried PH at Trackpoint 23. What I am seeing though is that the GPS sticks values (which basically record what the Navigation Control board is doing as though it had control of the transmitter sticks), are showing that the bird is going into Position Hold, then Manually Controlled and back to Position Hold. Were you "fighting" the Position Hold? If you look at the attached image (a screen shot of the data), look at the right hand end....that's GPS Sticks data. The very last data fields is a single character in single quotes. P means Position Hold, M means Manual Control. Just to the left of that is the manual sticks by channel, showing Nick, Roll, Yaw, and Gas. By TP 29 it looks like you're feeding in some Nick and Roll -- not unreasonably, of course, giving that the bird is heading for the Tiber! Again, I don't see any anomalies that would explain why PH is not working. Anyone else got any ideas? Andy.
I've still got a niggle doubt about the compass in all of this. I seem to recall that if you enable Position Hold and the compass isn't calibrated correctly, then the bird will "toilet bowl" (fly in circles). Could this be related to what's going on here? If I were you, I'd head out to a wider open space (farm lands, say), where you can afford to turn on PH and wait a few seconds longer to see what happens. Flying where you were, I'd be really nervous about the "bird meets Tiber" scenario. You'd need to re-cal the compass, and, of course, re-cal the gyros before you take-off. The internal contradiction here is: 1. GPS is good. Looks like you have a fix. 2. Compass values don't show mis-calibration as far as I can tell. Yet: 3. Position Hold cause high speed departure from present position. Have you by chance, made any changes using MK Tool to any of the settings that affect Gyros, Compass, etc.? If you had made any accidental (or even deliberate) changes to these values, they might explain what's going on. If you could follow this procedure, you could post your MK Tool settings -- much easier than having to transcribe them into a posting: 1. Connect up MK Tool via USB cable/Wifi/BT to the copter. 2. Start up MK Tool, and click on Settings. 3. Click on the Save button at the lower right hand corner. 4. Give the file a name and navigate to where you want to save it. The file will be of type .mkp (it's actually just a straight text file so you can open it with any text editor/Notepad/Wordpad). 5. Rename the file (either change the .mkp to .txt or suffix the .mkp with .txt giving name.mkp.txt). You need to do this to upload the file. 6. Upload the file as part of your posting. Thanks Andy.
Thanks, Tomorrow is our first day back on set after Christmas hiatus so ill be slowing down a bit. I'm going to forward all of this to my partner/pilot. He's far better at this stuff than I am.
Mark, I would go with Andy's suggestion of going to an open field and then activating position hold. I rarely use PH but I do know it takes maybe 2 sec to activate and I'm fairly certain there should be no stick movements during that time. I think with older MK firmware versions the process was near immediate. It does seem odd that it would haul off that fast with those stick movements....
Can anyone tell me if there's a specific process to engage AH or PH? Does it matter which is engaged first? Should you try to refrain from using the sticks in these modes?
Anyone else notice on GPX0004 log file that the battery voltage went down as low as 10.6 volts with a max voltage of 14.6 only?
Another quick thought, what would happen if you calibrated the compass using a smart phone app that shows geographic north but not magnetic north? True north and magnetic north are about 590 miles apart.
You would be off by quite a few degrees. Not a good thing at all. In the Pacific NW, the Cascade line of volcanos cause a magnetic variation of about 16 degrees or more the closer you get to the mountains (the iron in the magma). Most such apps use the phone's flux gate compass so tend to give you magnetic north as the default. To correct, you need to have data for a given lat/long that shows you what the compass deviation and magnetic variation is at your current location. So getting true north is harder than mag. north. (Compass deviation: errors in the compass, magnetic variation: difference between magnetic and true north.) On the iPhone, I use an app called Spyglass -- it has the advantages of (a) using the camera and superimposing a compass ring on what the phone is looking at so you can determine a distant landmark at magnetic east (see below), and (b) you can get true north if you need it. Of course, it's blisteringly expensive at $3.99. Why magnetic east? It's easier to "sight" along booms #7 and #3 (Octocopter) and hold them alligned to magnetic east during calibration than to try and imagine where you have boom #1 and #5. You could, of course, sight along boom #3 and #7 and hold them to magnetic west if you're an Australian or New Zealander. <evil grin> Andy.
Thanks Andy. I did some investgation and found magnetic north to be negative 9 degrees west of geographic north. I do have a few questions for you if you don't mind answering If I take the C8 out to the farm land and calibrate the compass with a real true compass do I have to recalibrate the Gyro's? When I say recallibrate the gyros I mean make sure the flight control board is level and hit left stick up and to the right? Or can I just initiate the gyros like I do with left stick upper left corner for every flight? I wasn't sure what's needed after the compass. The way you do it in the video is to level the board, calibrate the compass, and then just initiate the gyro's with upper left corner. Question 2: Is there a proper procedure to apply AH and PH? Can they be applied seperately? Do they have to be applied in any order? Is there any rules about not using stick commands when they are applied? Question 3: How do you normally fly? Control the C8 through the eyes of the FPV or just by looking up at the orientation of the C8? I find it much harder to fly it without FPV but I know I lack experience. Can you give any advice for inexperienced flyers to avoid crashing their large investments?