Glad you figured that out. Can you check the Recoms on the bad FC if they are coming loose? One of my FC 2.5 boards started to act weird once I realised that it was due to the broken leg of one Recom. Even though I was not powering anything from the board, it acted weird.
This is truly a incredible machine! Here is a feedback from the real world. The shooting day came up and we should all meet up in a marina to be shipped out to a remote island in small boats. I´m glad I decided to pack everything up in flight-cases as the waves was spraying salty water all over the place. The wind early in the morning was around 15-20 knots and increasing. On the island they start shooting with a steady cam. 2 guys had to cover the camera operator with a wind shield to make decent shoots. At that point I was seriously thinking to call of the flight. We did a test flight in 25 knots with the RED and it was sort of OK. Far from perfect but safe enough. One hour later we got the 2 actors for 30 minutes. The wind was now around 30 knots gusting 35..... We let the actors run downwind to get a stable shoot and to my surprise it actuality looked good. Close to the island there is a manned lighthouse. They sends out wind info every 30 minutes. And when we got back to the mainland I checked the weather info from the lighthouse and it turned out that we have bin flying in 35 knots - middle-wind.... The captain´s also refused to ship us back to the mainland and we had to wait some hours for the wind to ease of. Here comes the best part. The feedback from the director was that he was very surprised that we actually have made more stable shots than the steady-cam man on the ground... Have a nice day
Hey guys. I have been flying MK since about 4 years ago and before they switched to 6S I did a couple Cinestars with WKM. Now I finally got around to building up a Herk III / MK 2.5 / GPS 3 rig as an X8 with KDE 4215s. It really files great...no troubles at all EXCEPT for the AH. I have read in a couple other threads that people have struggled to get AH to hold during forward flight of any speed. I am getting the same thing. I have been keeping a change log of the things I have tried adjusting to get it to hold better and as of now I have the following Altitiude settings: Altitude P has been moved from the stock value of 20 to 25. Altitude Z-ACC has been moved from the stock value of 30 to 40 Barometric D has been moved from the stock value of 40 to 50. I understand Barometric D should make it more precise but prone to fluctuation in the wind - but I thought my increasing the Z-ACC value the acc sensor might catch the fluctuation before it gets to much. So far nothing has kept it from losing altitude in forward flight at any speed. Slow flight seems fine. Has anyone cracked this issue yet? Thanks!
Michael, I designed and printed a cover that surrounds the FC board. It's basically an open topped cube, with a passthrough hole for the 4 standoffs. The height of the cube's walls is level with the header pins. Then I stuck down some foam strips on the top rim of the open face, to fill in the remaining area between the top of the print and the FC board. Combining that, with the MK Tilt Compensation, has given me the best results so far. I think they are decent, but still far from perfect. I think my Tilt Compensation is about 104% and that goes pretty good, but it does start to climb still after a time
I'm leaving default tilt compenstation value at %110. There is cover around center plate, if I move fast, the altitude loose is still there. After this problem, I found myself always compensating the altitude with gas stick. Cam, can you post a picture of your protective case?
My question is why is there only altitude drop issues in X8 and not with flat 8? Last summer I converted a flat 8 to X8 where the flat 8 was running perfect,converted it to flat 8 and had AH issues. So after I tried changing AH settings unsuccessfully I converted it back to flat 8 and the problem completely went away. IMO the MK mixer algorithms directly affect AH when moving in a forward direction and none of the AH parameters (at least available in the software) are not addressing the proper dynamics of the X8. IMO the bigger the motor the worst the issue. When the copter in question was still in X8 configuration with the AH issues, Holger told me to put a cover on it and then change out the FCB. I did add the cover and it made no improvement. Changing FCB's improved it slightly but still had drop off in a forward or backwards direction. At that time he admitted to me that he does not like X8 configurations and believed they are very problematic. So when Holger develops new products such a FCB he is not testing it in both configurations unlike DJI which has no issues with this in a X8 configuration. The longer I fly both types, the more I really like DJI's flying characterics even in flat 8 configuration. I have about 100 flights on the Wookong and I find the overall flight smoother, and tracks better. I find that in non GPS mode the Wookong tracks much better and doesn't want to move around like a rocket ship that the MK ship does. I was experiencing problems with the Wookong in the wind but I am making good progress with that and its no worst than the MK ship now. I am starting to migrate away from MK and only use it for mapping and surveying applications. I know everything has its own advantages and disadvantages but for over flight performance DJI is better. Just my 2 cents.
Thanks for the replies everyone. Cam: If you would not mind either sending a photo or sharing the file to print one I would sure like to give anything a try that might help. Does anyone know how much control or help the ACC sensor for the altitude hold can provide? I have not really cranked up that value too high yet - because the first adjustment I made seemed to have very little effect on anything. If that sensor is helping detect vertical movement and compensate for height changes (in addition to the baro or just a quicker response to minor altitude changes) you wool think that it would not allow the copter to drop. But maybe there is a relationship to some other settings that need to also be adjusted to make it work better/quicker. Anyone spend any time trying to get this sensor to play a bigger role in height control?
Here's what I did. This image doesn't show the layer of foam that I applied, however you can see where it will go. There is a 3mm hole in all four corners for a screw for the standoffs. I took inspiration from APM/Pixhawk and how they evolved to include a housing for the barom. What I think might work really well however would be to put the Barom on a GPS-like mast, far from any possible pressure fluctuations and see how that performs. It's only a thought at the moment, I'm not sure if it's even possible. As I said, it's an improvement, but it's not there yet. I also don't believe that this problem is restricted to MK co-ax birds. I've spoken to a number of X8 operators on A2 systems and they experience a similar problem, however weren't aware it was common across the board. There's certainly something going on with 4, uneven disks of air (at double power), versus 8 disks of evenly distributed air.
If anyone would like to print the plate and test it for themselves, I've uploaded the .STL, .gcode and Solidworks Part to play with. It's hosted on this wetransfer link, which will expire in 14 days. I'm fairly certain this is the .STL I ended up using, however my solidworks machine died a painful death, and I'm on a borrowed computer for another week or two until I am up and running, user beware! https://www.wetransfer.com/download...eadedb19accd6ecdc35fd7bb20150609060412/8a99fb
This problem also exists on DJI A2. Some of you already know the user R_Lefebre; he thinks this way when he looked at DJI A2 logs: