Flight Log Education

[Rob666]

Hi All,
Been reading a few threads about the Mavic's flight logs and I too think the more we know about it, the better it is. I have been uploading my logs to Airdata to keep track of mileage, battery temps, etc. I also downloaded their CSV files to keep a bit of a logbook myself. But, if you're curious and want to do some digging yourself, or don't want to depend on others' analysis of your data, CsvView/DatCon is your friend. I just got it all to work and the amount of data that can be viewed is great. Some of it I understand, some of it I don't. I have uploaded an example of a graph in a new album "Mavic Flight Logs", it has general flight state (atti, sport etc), barometer, battery%, cellV to begin with. I also included a screenprint of the same flight, summarised by Airdata. It was a normal, successful flight, no issues. At least I think it was.
I hope this example may start a bit of a training class in flight logs - members who have experience in reading these logs, please comment on what you think is normal/abnormal, share your logs for others to provide input to, etc. The media album is open for members to add new files if they want. Sharing experiences will be very helpful in understanding and operating our Mavics with more fun, less issues, and more safely.
Who wants to comment first?

 

[Rob666]

Budwalker gave me some helpful tips, like including the images in the post, and splitting trends into different graphs. So here it goes.
Plot 1. Battery temp from start to end of flight. The solid colors display the state of flight, e.g. the purple patch is the period flying in sports mode.


Plot 2. Barometer / height.


Plot 3. Individual Cell voltage.


Being a normal flight, I'm curious what the drop in cell volts means. It seemed to happen shortly after take off, so I have zoomed in on plots 1 and 3 below, and added motor speed and current draw.

Plot 4. Motor speed picks up on motor start, settles a bit and then ramps up when the Mavic actually takes off to hover height. The cell voltage drop coincides with an increased current draw, a short time after the motors sped up. Lag? Or is something else drawing power? I didn't find the camera status in the SigPlayer but checking in Airdata's CSV file didn't show the camera was active at that time. I'm probably splitting hairs, but looking around is helping me understand what going on and what can be considered normal/abnormal. Anyone care to comment or share their experience?

 

[BudWalker]

Budwalker gave me some helpful tips, like including the images in the post, and splitting trends into different graphs. So here it goes.
Plot 1. Battery temp from start to end of flight. The solid colors display the state of flight, e.g. the purple patch is the period flying in sports mode.
View attachment 17482

Plot 2. Barometer / height.
View attachment 17484

Plot 3. Individual Cell voltage.
View attachment 17481

Being a normal flight, I'm curious what the drop in cell volts means. It seemed to happen shortly after take off, so I have zoomed in on plots 1 and 3 below, and added motor speed and current draw.

Plot 4. Motor speed picks up on motor start, settles a bit and then ramps up when the Mavic actually takes off to hover height. The cell voltage drop coincides with an increased current draw, a short time after the motors sped up. Lag? Or is something else drawing power? I didn't find the camera status in the SigPlayer but checking in Airdata's CSV file didn't show the camera was active at that time. I'm probably splitting hairs, but looking around is helping me understand what going on and what can be considered normal/abnormal. Anyone care to comment or share their experience? View attachment 17486 View attachment 17487

The battery info seems to have a 1.5 sec delay. Here the rFront speed and Battery:total:volts is shown. In addition to the 1.5 sec delay Battery:total:volts is sampled at 1 Hz.


To get a better idea of the voltage you can look at the motor volts data which is sampled at 50 Hz.

This is really useful in situations where there is an abrupt loss of power because the motor volts data will actually show a drop in voltage of about 50% before the recording ends.

 

[Rob666]

Thanks BudWalker, mine are very similar:




I was looking at the cell voltages for the reason that the guys at Healthy Drones use them to monitor battery health. Still wondering why the cell voltages show a drop like that only at one particular instance during the flight. Maybe it's nothing ...

 

[BudWalker]

Thanks BudWalker, mine are very similar:
View attachment 17496

View attachment 17497

I was looking at the cell voltages for the reason that the guys at Healthy Drones use them to monitor battery health. Still wondering why the cell voltages show a drop like that only at one particular instance during the flight. Maybe it's nothing ...

The .txt produced by the Go App for the Mavic doesn't have individual cell voltages. The cell voltages you see with Healthy Drones are each the total volts divided by 3. However, the .DAT does contain the individual cell voltages. If you want to see the actual values in the .txt you can use TXTlogToCSVtool to convert the .txt to a .csv. TXTlogToCSVtool doesn't assume any values into existence. TXTlogToCSVtool can be obtained by going to
[TOOL][WIN] Offline TXT FlightRecord to CSV Converter
You can also use CsvView which relies internally on TXTlogToCSVtool to do the conversion.

Volts will show a drop whenever there is an increase in current. I think your example flight just had that one current increase that came from the initial climb. Here is a flight with more than one current increases.

 

[Rob666]

Thanks BudWalker, great info and tools! If the txt logs which are the ones we upload to Airdata all the time do not have cell voltages, how do they determine the deviations between the cells? If the cell voltages were computed as a third of the total volts, the deviation would be zero? I've also done my own calculation of the deviations from the CSV file I downloaded from Airdata - something i want to monitor to keep an eye on the batteries' health.

 

[BudWalker]

Thanks BudWalker, great info and tools! If the txt logs which are the ones we upload to Airdata all the time do not have cell voltages, how do they determine the deviations between the cells? If the cell voltages were computed as a third of the total volts, the deviation would be zero? I've also done my own calculation of the deviations from the CSV file I downloaded from Airdata - something i want to monitor to keep an eye on the batteries' health.
View attachment 17617

Well that's changed. Used to be that the Mavic .txt didn't have any battery info. It would seem that a recent RC and/or Mavic firmware update now records that data in the .txt.

 

[msinger]

Used to be that the Mavic .txt didn't have any battery info.

It's still not in the usual location (just checked). They must be storing it elsewhere.

 

[Rob666]

I'm still on the original FW, 01.03.0100 for the AC and 01.03.000 for the RC.

 

[BudWalker]

I'm still on the original FW, 01.03.0100 for the AC and 01.03.000 for the RC.

That's even more interesting. Please provide a .txt generated from that firmware combination. Since it's a .txt it can be attached directly to a post. I.e. there is no need to zip it or Dropbox it.

 

[BudWalker]

@ferraript has updated the TXTlogToCSVtool so that it produces the Mavic battery info that was missing. To obtain a copy go to
[TOOL][WIN] Offline TXT FlightRecord to CSV Converter

In a few days I will distribute a version of CsvView that uses the new TXTlogToCSVtool version.

 

[Rob666]

Here is the .txt log. Interesting, uploading this log to phantomhelp shows cell voltages 1-3 are all the same, so no deviation, but when I uploaded this to Airdata before, I got different (meaningful) cell voltages and their deviations.

 

[BudWalker]

Here is the .txt log. Interesting, uploading this log to phantomhelp shows cell voltages 1-3 are all the same, so no deviation, but when I uploaded this to Airdata before, I got different (meaningful) cell voltages and their deviations.

At one time the individual cell voltages were not recorded in the .txt. Neither PhantomHelp, AirData, TXTtoCSVtool, or CsvView (which uses TXTtoCSVtool) could show the individual cell voltages because they didn't exist. Then, with a certain firmware update (version unknown to me) the cell voltages started being recorded in the .txt but in a different location. AirData figured this out and then you observed that AirData was reporting those cell voltages. Knowing this, this mornin, @ferraript then determined where the cell voltages are and changed TXTtoCSVtool. Presently, we are waiting on me to release a CsvVersion that uses the new TXTtoCSVtool. I would also bet that PhantomHelp will be changed to accommodate the change.

What's odd is that your claim that the .txt you submitted is from firmware FW, 01.03.0100. Are you sure? If so then none of the converters should work.

 

[Rob666]

Double checked on the iPad and in Assistant just now, here are screenshots so you can see for yourself:

 

[msinger]

I would also bet that PhantomHelp will be changed to accommodate the change.

I'm working on it

 

[Scottd9000]

I have the DAT files from the Mavic and would like to convert them to .txt or .csv for upload. Is there a free tool for this?

 

[Rob666]

I have the DAT files from the Mavic and would like to convert them to .txt or .csv for upload. Is there a free tool for this?

If you mean upload to Airdata (Healthy Drones) you don't need to convert first, just get their app to upload to their site. If you want to look at the DAT files yourself you can do that using the tools mentioned in this thread.

 

[Scottd9000]

If you mean upload to Airdata (Healthy Drones) you don't need to convert first, just get their app to upload to their site. If you want to look at the DAT files yourself you can do that using the tools mentioned in this thread.

I wasn't able to drag and drop the DAT file into the Airdata website. Is there a trick?

 

[Rob666]

I wasn't able to drag and drop the DAT file into the Airdata website. Is there a trick?

No tricks. A DAT file produced from a 20 min flight is easily a couple of hundred MBs, so not practical for uploading. Your app, like DJI Go, writes its own log file in .txt format to your tablet or phone that controls your aircraft. These are the files that you can sync / upload to Airdata's website. Easiest way is to use their HD Sync app, which does the work for you.

 

[Scottd9000]

Ok, here is the deal. I purchased my Mavic used but the original owner didn't clear his DAT files off the Mavic so I am curious about his flying. Not a huge deal but I was looking for a way to convert. I tried to load them on to me iPad but the App didn't see them.