Copter-4.1.0-beta8 released for beta testing

ArduCopter Copter 4.1
,
3

Plan to test it soon on one of my frames today.

1 Like
5

I believe it was already suppported since 4.0?

EDIT: Looks like this is about MatekF765-SE

1 Like
6

@sergbokh,

Thanks for that correction, fixed above!

7

Flown this version on my micro copter with GSF enabled. There was cloudy so GPS suddenly lost good HDOP midflight. There were GPS Glitch and EKF Variance messages, and Loiter become slightly strange. So I switched to Stab. Then after a seconds GPS become good again and EKF recovered fast, and there were no issues then.

1 Like
8

I’m configuring the notch filter for my mulitirotor following the guide, I used the firmware server to make a pixhawk1-1m with dsp. I noticed that after I did the fft_enable=1 and reboot, when I did ins_hntch_enable=1 it gave me a parameter set error unless I also set this other parameter ins_notch_enable=1 which i don’t see mentioned. Just thought i’d mention this, haven’t flown yet

9

The two are unrelated. For FFT you need both HAL_WITH_DSP and HAL_GYROFFT_ENABLED set on your custom build.

1 Like
10

Just wanted to say I tested out Beta8 on the weekend with my trusty F450 running a Matek F405-std. It was remarkably uneventful. (Just how I like my flights) Considering how bad the vibrations are on that quad it flies really well. If you’re looking for data I can post a log.

1 Like
11

cool I going to test Beta as well now i have a good reference and everything is working well.

1 Like
12

Did some Autotune testing on a 7 Inch Iflight Chimera frame today, and found a few strange issue’s during this run.

Is position hold still the way to start-off the Autotune, as it looked like there was initially no Position Hold at all… Drone slowly drifted away, while the autotune did it’s thing, Just Pitch and Yaw only.
Only after a while it stabilized the position and the drifting stopped completely. (no change in wind for sure)… The strange thing was that loiter was perfect/firm without any drifting…
GPS started with 13 Sats, and improved to 18, Hdop around 0.78

After looking at the log files after the test flight a found a lot of GPS Glitching which may have been the reason initially…

However there looks like a repeating sequence of Glitches, which I have never seen before in my logs.
Flight Controller is a Matek F405 STD, GPS is pretty old Ublox NEO M8N
Which might not be fully supported any longer?

The BIN file: Microsoft OneDrive - Access files anywhere. Create docs with free Office Online.
Picture of Glitching: Microsoft OneDrive - Access files anywhere. Create docs with free Office Online.

I hope someone has an Idea related to the GPS Glitching?

13

Hi @FPVDroneEG,

Drifting while in autotune is normally because the vehicle was switch to AutoTune mode from AltHold… or the roll, pitch or throttle sticks are not centered… it is most often the stick positions that are the problem and sometimes it is because the RCx_DZ is set too small so any stick position appears to be non-zero.

14

But the drifting occurred while it was in PosHold, and that puzzled me a bit…., I would expected a kind of fixation you get in Loiter…, and that a little Wind would not impact the Tune….

15

A little wind, no matter how little, will impact the tune. The copter will drift. It is your responsability to bring it back.

16

Ok thanks, so Poshold is simply not as fixed as Loiter…

Good to know !

17

As per the Wiki:
AutoTune performs a weak position hold if invoked from Loiter or PosHold

“weak” being the key word there.

18

Thanks Dave, Randy

So Autotune itself will enable a weak PosHold ?
Years ago you had to switch to PosHold and then Invoke Autotune, on a separate switch…
That has indeed changed since the earlier versions of ArduCopter.
Good to know !

I am using Arducopter since 2014, and version 3,14, and have been testing quite some beta versions during the years…

19

Yes, if you switch into it from PosHold or Loiter. If it’s switched from AltHold you will have to re-position as it drifts around.

20

Thanks I’ll try the Loiter first option next time.

21

So just to be clear, the “PosHold” flight mode should hold position just as well as Loiter mode. AutoTune also holds position (if the mode was Loiter or PosHold just before the vehicle was changed to AutoTune mode) but it doesn’t not use the PosHold or Loiter flight mode to do this. Instead it does a weak position hold which will lead to it drifting around a lot.

22

Thanks Randy that explains it. I was expecting these flightmodes were the source of the position fix, but I guess that got changed over time…
Which is fine as long as you expect the behavior…

What about the GPS side of my question…, I guess I need to try to fix the baudrate to 115 with GPS_DRV_OPTIONS to 2?, I,e, to get rid of these strange spikes (which a rather consequent in the log, see picture)
I can’t believe the glitching is this regular and repeating….

In the mean time I have ordered a couple of new Matek Gnss M9N-5883 NEO-M9N versions, to replace the old ones…

Erik

23

@FpvDrone-EG,

Re the GPS, I don’t see the picture with the issue but -beta8 (and higher) include a fix for some 60ms delays we were seeing on Matek boards.

This is the line in the release notes for the fix.

b) ChibiOS scheduling slip workaround to avoid occasional 60ms delays found on MatekH7