There is also the curious phenomenon that the Compass Not Calibrated error disappears every 20 seconds....eg. rows 11, 12, 31, 32, and 51, 52 (there are two rows of data being recorded for each elapsed second). Andy.
I forgot that I actually had my wife film the first flights with my iPhone just incase anything went wrong, I could look back to see what happened. The video isn't all that exciting, but you can see and hear exactly what happened and when the throttle jumped to full speed. Looking back, I'm amazed at how fast it all happened. One other thing that concerns me is the speed of the auto-landing... I am going to have to slow that way down... password: cs8hl
Andy, look in the "Altitude (m)" column. In the first 21 rows, there's only one row where there's a '+'. All the rest are '-'. I believe that's analogous to the indication on the telemetry screen on the Graupner, where the altitude is suffixed with a +/=/- to indicate the throttle. But maybe that's more of an indication of where the actual throttle is, not the virtual throttle. Do you know?
I also see that the "Available Motor Power" field is horribly wrong. In Ingo's GPX viewer, that field is normalized to a percentage, and it's 5%!! I wonder if the copter used 20X the thrust necessary? In any case, this is considerable cause for concern. I'm also curious about the thrust/weight ratio, as it appears there was no dummy payload, and it's equally impressive that the copter managed to pull the kind of current it did without having any payload. I'd expect, even during a full throttle climb out, a small fraction of that power consumption with that rig.
The +/- entries in the Altitude are curious, but the actual calculated differences between successive altitudes are used to calculate the variometer reading. When I last looked at the code the formula was: CurrentVariometer = (PreviousVariometer + (2 * (CurrentAltitude – Previous Altitude)) / 2 (This was in line 613 of spi_slave.c for the NC board -- but doubtless that line number is wrong now because of updates.) So it appears to me that there is a curious contradiction between the +/- in the Altitude and the Variometer readings. The actual stick throttle is shown in the RC Sticks column. The GPS sticks show the autopilot input. Here's the verbatim script from the RGG006B DVD I did on GPX files: The RC Sticks values refer to the numbers being transmitted on the various radio control channels — these are the control signals that the receiver is getting from the transmitter so they can be invaluable in terms of trouble shooting if you see bizarre values — or sudden changes from large positive values to negative values. The order of the RC sticks is : Nick, Roll, Yaw, Gas, POT1, POTI2…., POTI8.You will need to refer back to how you have currently set up the Transmitter for the various POTIs to relate them back to channels and from there back to specific switches and controls on the transmitter.The next values are GPS “sticks” — they’re not really sticks, but they represent the values that the GPS system is feeding into the Navigation and Flight Control boards when the GPS system is being used to control the copter. The three numeric values are the Nick, Roll and Yaw values. These are followed by a single character that indicates the status of the GPS control system. '-' = no GPS fix '/' = off '?' = Coming home, but home Position unknown (goes to PH then) 'H' = Coming home 'W' = Flying Waypoints 'D' = Dynamic Position Hold 'P' = Position Hold 'm' = Manual controlledHere's a custom view showing the Altitude, Variometer and sticks values: So you can see Philip's control inputs on the throttle as the fourth value of RCSticks (-127, 16, -69, -127, -127....where -127 is throttle stick all the way back). You can also see that the Altitude +/- doesn't seem to be corresponding to the RCSticks throttle value -- well, not exactly. For example, the throttle stick went up to 16 (where 0 is 50% throttle) in row 2, but the "-" only changes to a "+" in row 3. It could be that the +/- is more intended to be a reflection of the delta Variometer (which is itself based on the delta Altitude)...but the data looks a bit contradictory. Let's see what Holger notices... I think Philip still holds the "Whoa! Maximum Wattage I have ever seen" award.... Andy.
By the way, have a look at the second GPX file. 1. No motor "NOK" (not ok errors). 2. No compass not calibrated errors. 3. Altitude +/-/= seems to be correlating with the Variometer values. I don't see anything apparently out of the ordinary -- although I only skimmed through the data. Andy.
The copter is a CS8HL, 500mm booms with Tiger 4012 motors and 15x5 t-motor props, so there is a lot of power/thrust on tap! There was no payload at all, so there was definitely no need for as much power as it was putting out.
I just flew it again this morning and had zero problems, no errors, no NOK and no uncommanded ascensions. It flew perfectly and handled extremely well, the only problem I had was a low voltage warning towards the end of my flight(landed at 45% capacity) when I gave it nearly full throttle to see if I could replicate the vertical speed I had seen during the first flight. It was fully in manual mode and seemed to ascend at about the same speed. I immediately lowered the throttle when I heard the voltage warning though, so I only climbed maybe 10m. I think I may of damaged the battery with the first flight, as it was it's first use and drawing that much current off a new battery can't be good for it. With my other setups I usually fly new batteries at least 5 times down no lower than 50% before I considering them "broken in" and really push them for any kind of performance. I seem to have one cell that is about 5-10% under the others when I put it on the charger.
I wouldn't worry about a temporary low voltage alarm, Philip....that's just Ohm's Law at work when you do a high-powered climb. You'll know the LiPo's bad if it gives you a low voltage alarm just after take-off or after only a very short flight (that is, it will not hold its charge). However, a cell that's more than 0.1 volt out of balance with the others is of concern. Can the charger not balance the low cell? Andy.
Holger emailed this morning: Hello Andy, I went through the code and made some changes to avoid a situation like this. http://mikrocontroller.com/files/PublicBeta 2.06f (21.7.2014) - Secure function: Avoid that the Altitude setpoint can be much higher than the actual altitude - Use MinGas setting before while "Waiting for takeoff" instead of "AltitudeMinGas". Regards, HolgerLooks like you need to upgrade the FC board ASAP to avoid this problem. Andy
By the way, Holger points out that there is a bug in MK_GPXTOOL -- the error that is showing is "compass not calibrated" is in fact a motor error. He recommends that we use http://www.mikrokopter.de/ucwiki/en/SimpleGpxViewer as a way of cross-checking GPX files -- this is software developed by MK themselves, I believe, and therefore will be better in sync with the meanings of bit flags in the GPX data. Andy.
Oh....and I've emailed Franco Scapin, the author of MK_GPXTOOL, and let him know about the bug. Andy.
Thanks for posting this! I have a couple questions... I noticed there are two hex files listed in the link provided: Flight-Ctrl_MEGA644_V2_06f_SVN871.hex Flight-Ctrl_MEGA1284p_V2_06f_SVN871.hex Which file should I use? I'm not familiar with the difference between "mega644 vs mega1284p". My second question comes after I do the update, will I have to re-input all of the settings into MK tools settings? I've read online that you should manually input all of the data back in, rather than use an existing config file, is that really necessary? Once I have that configuration setup again, will I have to do another compass calibration and ACC calibration? Thanks for all your help! -Philip
The FC boards version 1.0 and 2.0 (red printed circuit board) use the MEG644 central processor chip. The FC boards version 2.1 and 2.5 (black PCB) use the MEGA1284. I'd suggest that you do: 1. A Save from within MK Tool and write the current settings to an .mkp file on your computer 2. Upgrade to the new firmware. 3. A then use the Load button to restore the settings from the file you created in step 1. Andy.
Thank you! Would you recommend another compass calibration and ACC calibration after the update as well? -Philip
Philip: I think that would be a good precautionary idea -- it's maybe not necessary but there is no downside to doing a redundant calibration of the compass and ACC. Andy.