So I decided to upgrade the FC board to the new ACC for better altitude hold -- which, I confess I tend to use for follow-shots, low passes over orchards (yeah, yeah orchards means my nemesis, trees). The new ACC is only recognized by the latest firmware (FC 0.90j, NC 0.30h, MK Tool 1.80b), so having modified the FC board and checked that all the sensors were working with MK Tool, I tried to upgrade the firmware via the Smart On Screen Display board USB connection. Uhuh. The NC board went OK, but the FC board kept failing -- I think there's something about going through the SOSD interface that will allow you to use the general functions of MK Tool but not the firmware upgrade. Solution: Unplug the short ribbon cable between the SOSD and NC board and connect up the MK USB adapter directly to the FC board. Then I deleted the SETTINGS.INI file from the microSD card, forcing it to be recreated on next start, and, with the props removed and the FC board (well, actually, the battery plate) level, I calibrated the ACC (left stick top right hand corner until it beeps), checked that the FC sensors were moving appropriately using the MK Tool Scope (they were) and 3D model. I then went through making sure all Channels were working correctly and that the mapping for the Altitude Hold and GPS hold were to the correct POTI. I had to set POTI one to RC channel 7 as that's what I use on my transmitter. Did a test flight this morning in my driveway. Copter took off but felt "odd." Hmm. Yaw? OK. Roll? OK. Nick. S*(t S67t S^&*t!!! I don't have any nick control! Copter now starts drifting towards me (about 15 feet away from me), so I landed "briskly." Nice bounce. Good reason for keeping the landing gear legs just a tad loose to allow them to rotate to absorb some shock. Now that was really, really scary. My body has now served myself a Vente Adrenalin! Who needs caffeine, eh? Connected up to MK Tool. Searching, searching, searching....check the Channels. They're OK. Check the...what....really.... Huh? Nick and Roll are both mapped to channel 2. Well THAT would certainly explain what was going on. I've never seen that before and I must have overlooked checking that (why would that ever, ever be a valid default?) Please feel free to learn from my mistake -- double check both the RC channels are moving correctly AND that the assignment to roll, nick, yaw and gas haven't somehow mysteriously got changed. Did the new ACC make AH work better? I don't really know yet -- I tested it and it did hold altitude -- although it drifted down just about a foot or two over about two minutes. I'll test it more thoroughly. Unfortunately, there doesn't seem to be any way of actually proving positively that the FC is using the data from the new sensor. Andy.
Andy what version did you start with? I just did a .g to .j upgrade without issue. Haven't flown yet due to high winds but .g was fine. I did notice that the MKTool also was upgraded during the .90 release. I do my updates FC first followed by the NC. When I went from the version before .g I loaded the FC just fine with the previous MK version but the NC wouldn't load until I updated MKTool. Also thought that the correct order is always FC then NC. Not sure if that is an absolute, best practice advice or urban myth. I try to take a screen shot of all of the setup screens (you can do the data file but I find screen shots are easier) before the upgrade, validate that nothing has changed after the update. Moving to .90 basically cleared everything out so had to make lots of data inputs to get back to my starting point. Also don't forget to make the additional settings needed to make the new AH ACC work properly.
Pretty sure I was at .g too. I don't think it matters whether you upgrade the NC or the FC board first -- completely separate subsystems connected by the I2C bus, so I think it might be urban myth. The reason I say that is that using the MKUSB adapter to power either the FC or NC board you can update when it's not even mounted on the copter and just staring at you from the table top. I did notice that the new ACC can be tested. The MK Tool scope should show a value of < 700 for AccZ. I'll check then when I get to the field -- just heading out now so it's all packed up. Also, for my test, I probably wasn't above 1.5 meters so it wasn't a valid test. I'm just heading out to do some proper tests -- but I'll also check that the "Hoover [sic] variation" described in the documentation is = 1. It was certainly set to Vario mode. Yeah -- I also take screen shots, but The Big Mistake was thinking that, while other settings would get changed (PID etc.) that the fundamental channel assignments could not possibly get changed too. Whooof. My palms are still sweaty. Andy.
It's amazing the details you have to check before flight. I had a slightly similar experience where all of a sudden my joystick calibration on the MX20 for YAW changed and when the stick was in the center position it was actually what it would normally if I was half way to the right. It was spinning clockwise like crazy. Let us know how its working for you. I just ordered my ACC. BTW: Did you switch to the 55mm standoffs?
Very few "more uncomfortable" moments than when your flight controls don't respond the way your expecting. Glad this all turned out good in the end.
Thanks for posting this. Today I upgraded the FC firmware to 90j on both copters. Both my flights I got the motors up to speed just to the point where it would almost take off then pushed all the control sticks around to see how it responded. Once i knew they were responding correctly I took off. Josh
@DaveK: Yes, I did switch to 55mm "axial posts" -- in a snazzy shade of green: http://www.axialracing.com/products/axa1319 @Scott: You're right -- it's how I imagine it to be if you're driving your car and the steering wheel comes off in your hands! @Josh: I did that too, and it sorta kind felt weird, but I thought it was just a ground-effect problem....nooooooo...... Andy
Dave: As Gary suggests, it's a standard procedure to delete the SETTINGS.INI file from the microSD card when you do a firmware upgrade. It will be recreated by the Navi Control Board firmware next time you power it up -- but it's recommended by MK that you do this (the reason being that a new version of firmware may use different values in the file -- but the firmware is not smart enough to realize that current SETTINGS.INI is "out of date." (That could have been corrected by having the firmware write a version number as one of the parameters in the SETTINGS.INI file, but MK doesn't use that technique, I'm afraid.) Andy.
thanks guys thats really good info to know. What kind of things can happen if you don't update the ini file? Can it affect GPS funtions?
There is an GPSAUTOCONFIG entry in the file -- see http://mikrokopter.de/ucwiki/en/MKGPS for what that's all about (scroll down to Auto Configuration). Open up the SETTINGS.INI file using a text editor and you'll see the values -- it's just a straight text file. Here's what you'll find in a typical file: ######## MikroKopter configuration file #########------------------------------------------------------# KML logging interval in ms (0 = disabled) KMLLOGGING = 500# GPX logging interval in ms (0 = disabled) GPXLOGGING = 500# max. altitude in m MAX_FLYING_ALT = 0# max. range in m MAX_FLYING_RANGE = 0# Auto-descend range in m (0 = disabled) (only comm. License) DESCEND_RANGE = 0# GPS SBAS mode (0 = off, 1 = on) GPS_SBAS_DGPS_ON = 1# minimum time of the Waypoint-Event value (seconds) MIN_EVENT_TIME = 2# acceleration for flying waypoints in percent (0-100) WAYPOINT DYNAMIC = 100# GPS configmode (0 = off, 1 = on) GPSAUTOCONFIG = 1 If you want to know what the above means, check http://www.mikrokopter.de/ucwiki/en/NaviCtrl_2.0#en.2BAC8-NaviCtrl_1.1.SETTINGS.INI Hope this helps Andy.
thanks Andy. I deleted my settings file this afternoon and made a flight so it created a new config file. I think this is why I had the MK lady say "max distance" when I was out around 400 meters when I first went to 90H.
Andy who did you buy them off of? I tried calling axial racing direct and can't get a live person. <EDIT> I found them at Tower Hobbies
I just upgraded with very little hassle. I would also add that I just reviewed Andy's Rather Good Guide (beginners) to MKTool before I upgraded, and I have a bone to pick with him. It should clearly be named an Awesomely good Guide. His review of fundamentals as well as flow of the topics is fantastic for newcomers, and a great review for the old salts. I recommend it. Currently, I'm at 20h for the Navi is 30h and 90j for the FC. MKTools is at 80b. It's neat to see the FC at 2.2. For others who have done the ACC upgrade, could you share what you are using for the parameters for the ACC? The good folks at Quadrocopter who did the upgrade (Riley is awesome) set them at: P=20 Barometric-D=40 Z-ACC-Effect=50 HoverVariation=1 What have other folk tried? These are a bit off the defaults.
I'll ask Riley tomorrow. I just flew it for the first time. It was like it was on rails. Everything else checks out OK, HOWEVER - the aural announcement for "Calibrating" does not occur. It calibrates - several beeps and the telemetry resets to proper values. Anyone have any ideas where the aural announcement for calibration may have gone? Other aural announcements are fine. All opinions welcome. -m