zeb_
lvl.3
Germany
Offline
|
Hi,
After having played with a friend's MM for a few weeks, I finally bought my own. And having followed all that prop story, I can report some stats made with a brand new MM.
- I purchased the MM with the Fly more combo. The MM arrives already packed in the case. Props were maintained (very tightly) horizontally with paper bands and were stored in the case as pictured (albeit the fact the drawing is slightly incorrect, as reported before). I immediately took pictures of the props as shown below, before first flight.
One can see that the LBack props look slightly more flatten than the other ones (in particular RFront, which is the same CCW prop type). Contrary to the others it does not have this slight pitch up at its extremity. This is less dramatic as reported by sean-newbie in this thread: https://forum.dji.com/forum.php? ... 172&pid=2155310 but looks similar to Ralle's picture at https://forum.dji.com/forum.php? ... 705&pid=2150960 (notice that Ralle did not have the motor speed error). - I used DJI Assistant 2 on a W10 PC to update the firmware. Assistant 2 indicated that the MM arrived with FW version 0400, and I could upgrade to 0500, which I did. Everything was fine and Assistant 2 showed that current FW was indeed 0500.
- I installed DJI Fly (and checked server sync was indeed disabled), connected to the RC and linked the RC to the MM which I also registered. Interestingly, the DJI Fly app notified me that there was a FW update available. This was surprising and I went ahead anyway with the upgrade, which seemed to complete fine. After restart of the MM DJI Fly did not request to upgrade the MM anymore. I checked in Assistant 2 that the FW was 0500 and that there was no error. Must be a bug in DJI Assistant which fails to bump the version number read by DJI Flay maybe...
- Inserting the 2 other batteries led to their upgrade too and everything went fine.
- Maiden flight was done outside but with absolutely no wind. I auto took off and did the hover test: 2 times around 1 minute. No error shown. .DAT file was extracted from phone, used DatConv to get the CSV data and plotted data and calculated speed mean over 8000 rpm with an R script. Here is the figure:
Hover
RFront: 8717.768
LFront: 9585.989
LBack: 9335.398
RBack: 9018.147
One can notice that all speeds are under 10K rpm. RFront is significantly lower, but LBack, which is often the culprit when it comes to the speed error, is not the fastest with 9.3K. - The day after I did a first test at low altitude outside with almost no wind. Did not go higher than 10 meters and was very careful and gentle with the MM, flying slowly around trees, using Mode P and C. Here are the results:
Low altitude
RFront: 8906.579
LFront: 9519.387
LBack: 9355.405
RBack: 9240.972
Again very consistent results with the RFront and RBack slightly faster, which is expected when manoeuvering. No error. - In the afternoon I went in a countryside place clear of houses and with some woods. Low wind. This time I pushed it a bit more, going up to 30 meters, and piloting a bit farther and faster (Mode S at some point). Here are the results:
Higher altitude
RFront: 9141.595
LFront: 9713.071
LBack: 9741.867
RBack: 9531.982
As expected, motors go a bit faster on average, since the MM went faster and higher and in a more open space. But still no error and we never reach 10K on average. Highest rpm was around 13K with LBack and RBack at offsetTime 510, but this was very transitory and possibly while using the Mode S (I cannot ascertain this, is it shown in the logs?) - Conclusion: I am delighted with my MM. No motor speed error and it seems that even with a brand new MM it can be a miss and hit. But I was really interested in the rpm data and here I show that despite the LBack props looking (slightly) more flatten the MM flies extremely well and the motor never reach 10K on average. These props have therefore suitable lift power and there is no outlier. We know from published graphs that the motor speed error is consistently linked to a high mean rpm. This means that the difference between props leading to high mean rpm and 'normal' speed may be small, and that an apparent flatten shape may not be a problem (to a certain point I suppose). Maybe other factors (such as plastic quality, flexibility?) play a role.
Voilà! Hope this helps.
|
|