search

GrandK / megapirateng-add-ons

Automatically exported from code.google.com/p/megapirateng-add-ons
0 stars 0 forks source link

Crius aiop v2/mpng2.9.1b R8 - Motors twitching/rev up for split second #2

Open GoogleCodeExporter opened 8 years ago

GoogleCodeExporter commented 8 years ago 
Quad X/1300kv/F-30A Simonized/Ublox/Bluetooth on RCtimer Crius aiop v2.
Erase eeprom in mission planner. Deleted Arduino 1.0.5/installed 1.0.3. Deleted 
old megapirate/downloaded new 2.9.1b R8 add on. No errors while uploading to 
board.

Using mission planner 1.2.55 - Acc/TX/Compass calibration are good. ESC 
calibration successful.

PROBLEM: When I activate with right rudder, give throttle, motors start to 
increase speed and then all of a sudden 1 or 2 motors rev up 100% for a split 
second - like a major motor twitch. I tried this from 0-50% throttle with quad 
in my hand and every few seconds this will happen.

What I tried - Recalibrate tx/esc`s a few times, deleted megapirate/arduino 
from my pc, erased eeprom and tried this for 2 more times. Checked odd 
tx/throttle behaviour in mission planner, but none.

Original issue reported on code.google.com by JacoBkin...@gmail.com on 26 Jul 2013 at 11:45

GoogleCodeExporter commented 8 years ago 
Oh sorry, when I go back to Megapirate 2.8 R3 there are no problems

Original comment by JacoBkin...@gmail.com on 26 Jul 2013 at 11:47

GoogleCodeExporter commented 8 years ago 
Have just tried 2.9.1b R8 on my tricopter with completely different 
esc`s/motors and of course the frame - Same issue. On 2.8 R3 no problem
Used the same tx/rx, maybe that can be the issue

Original comment by JacoBkin...@gmail.com on 27 Jul 2013 at 5:08

GoogleCodeExporter commented 8 years ago 
First let me know the Rx model and config /CPPM or std connection, I assume you 
use A8 for CPPM/

If you are using CPPM on A8 you may try to adjust frame timing:

#define MIN_PULSEWIDTH 900
#define MAX_PULSEWIDTH 2100
// PAKU tune MIN_PPM_SYNCHWIDTH if you fighting known FrSky to short PPM frame 
length problem.
#define MIN_PPM_SYNCHWIDTH  2500

http://code.google.com/p/megapirateng-add-ons/source/browse/trunk/libraries/APM_
RC/APM_RC_PIRATES.h

Mission Planner channels bars (green) are very inaccurate on fast changes, you 
would better try testing Rx under terminal window (best using Putty emulator)

Original comment by pa...@kulma.pl on 28 Jul 2013 at 8:36

GoogleCodeExporter commented 8 years ago 
Thanks for the reply. 
It`s a JR Propo DSX7 tx on a 6ch RD631 DSMJ rx. Think you know this, but the 
dsmj tx/rx came out in europa/japan regions with frequency hopping. RX are 
connected the standard way (+-8 for throttle, 9/10/11 for AIL/ELE/RUD and so 
on). SERIAL_PPM_DISABLED, Tx_mwi, CH7_DO_NOTHING, OSD_PROTOCOL_NONE and then 
quad in x mode and tri in x mode.
If my throttle are jumping, would all the motors not make a twitch? I placed it 
level/hold it in my hand and it`s always the front right motor on the quad and 
sometimes the right rear. On the tri, seems only like the right motor.
Thanks

Original comment by JacoBkin...@gmail.com on 29 Jul 2013 at 6:27

GoogleCodeExporter commented 8 years ago 
I think it's something outside the rx decoder. Try to change ESC connection 
order /just for test/ look if motor problem will still occur on the same leg.

You may try to connect other but 8ch receiver, but as I wrote I think it's 
something outside Rx decoder. 

Connect a ground signal to the other 2 channels pins while using 6ch Rx ...

Have no other reports like that so it could be your setup related /ESC software 
???/.

Original comment by pa...@kulma.pl on 3 Aug 2013 at 10:25

GoogleCodeExporter commented 8 years ago 
Quad X/RC Timer AIOP V2.0/930kv/ESC Q Brain 4 x 20A/Ublox
Spektrum DX7i/ OrangeRX R710
The PPM_Disabled

I have same experience on 2.9.1b. Now I'm on 2.8.R3 without problems.
The full throttle pulses occures after turn to ARMED state and it's not 
necesary to push throttle up.
Later I will send log files for investigation.

Original comment by Petr.Fru...@gmail.com on 6 Aug 2013 at 9:11

GoogleCodeExporter commented 8 years ago 
Tri esc`s are pre-flashed simonk from rctimer / Quad esc`s are hobbyking f-30a 
that I flashed with the latest simonk firmware - Think the esc`s are out of the 
question. Motors on quad and tri are from different company`s/different make.
Removed my 6ch and tried my 7ch on quad/tri - still the same problem.
Dont understand what you mean by: Connect a ground signal to the other 2 
channels pins while using 6ch Rx ...?

Original comment by JacoBkin...@gmail.com on 7 Aug 2013 at 6:07

GoogleCodeExporter commented 8 years ago 
on pins: just to be sure CPU is not getting any strange signals on not used 7th 
and 8th pins.

You may try CPPM Rx if you have one. I think the problem is that you are not 
using full 8 channels on std Rx.

Original comment by pa...@kulma.pl on 7 Aug 2013 at 8:14

GoogleCodeExporter commented 8 years ago 
Same problem here as original post. Fast motor rev ups at random times (about 
every 5-30 seconds). Not using CPPM. Nice hovering and sometimes fast rev up 
but very short time. Quad will be stabilized after rev ups. 2.9R7 worked 
without this issue, but R7 dropped whole quad from sky. Disarmed in mid air :(

450 size quad frame
30a turnigy plush escs
aiop v2 board + gps
turnigy 9x with frsky module and 6ch receiver all channels connected to aiop v2 
board with 6 wires.

Can someone tell witch release is stable enough to find and learn PID settings 
and fly modes without fearing these two issues :)

If you look this video at 0:20 and 0:42, you can hear that short revup, sorry 
but hard to see that in video, normally rev up is easy to see cause copter tilt 
some angle and slowly stabilizes that.

http://www.youtube.com/watch?v=0ZY-H5-6m-4

Original comment by jarmo.ha...@gmail.com on 6 Sep 2013 at 7:19

GoogleCodeExporter commented 8 years ago 
The Std Rx code is the same and come from latest Alex code. I know the code is 
running in other software versions so do not think that's the problem. 
But the problem could be, as I wrote, in the not connected ch7 and 8 pins. The 
signal could fluctuate there ...

You may try to GND connect signal input for ch7 and 8 or change your config to 
CPPM.

Could be it's not a fault but intentional APM code work on something. We need 
more testing /like disabling GPS etc/ to find out. For me it could be a data 
from accelerometers or gyros could have picks ... need more test data...

But anyway, do not worry about the disarms, have no reports for long long time 
and hundreds downloads :)

Original comment by pa...@kulma.pl on 6 Sep 2013 at 7:36

GoogleCodeExporter commented 8 years ago 
Only just found this release as a newbie to MPNG (lost my KK2 and went to AIOP 
and MPNG, just finishing build and read about the disarms.. Sir Alex maybe not 
working on it anymore.. started to panic and think about multiwii) and feeling 
VERY grateful thanks Paku! 
Sorry I'm rambling.. I had this issue when testing yesterday and that was on 
2.8R3, on a Crius AIOP v2 Tricopter with GPS, 30A SimonK ESCs.  Only using 4 
channels on Rxr and AIOP whilst testing on the bench. Had 'glitching' from 
Right motor at low/medium revs, and more worrying one time when it seemed like 
all 3 motors briefly ran up with zero throttle.

Will test again tomorrow with and without GPS.  Haven't experimented with logs 
yet so will need to read up on that.

thanks for all your hard work!

Original comment by nrlchamb...@gmail.com on 6 Sep 2013 at 6:34

GoogleCodeExporter commented 8 years ago 
Hi again. I tested with ch7 and 8 connected to ground. Still had this rev up 
issue. Disabled GPS options from mission planner, same issue still. I didnt 
disconnect whole gps unit to test this.

Original comment by jarmo.ha...@gmail.com on 6 Sep 2013 at 8:16

GoogleCodeExporter commented 8 years ago 
Is that normal to see 4.5 voltages in V+ pins going out from aiop 2.0 board. I 
measured 5.0 volts from HobbyKing 5V3A UBEC and in voltage input pins in AIOP 
board. I can measure 4.5V from TX  + - pins, and gps red/black wires, should 
there be 5V or is this 4,5V normal? 5.0V only in input pins, 4,5V everywhere 
else. I use Turnigy plush 30A escs with disconnected red (+) wire.

Somewhy I also want clear these out, I use:
#define TX_CHANNEL_SET TX_mwi
#define SERIAL2_BAUD             38400// GPS port
#define GPS_PROTOCOL GPS_PROTOCOL_UBLOX
#define CONFIG_BARO AP_BARO_MS5611_I2C

Original comment by jarmo.ha...@gmail.com on 6 Sep 2013 at 9:58

GoogleCodeExporter commented 8 years ago 
Should be 5V I think,  but board have it's own regulator after ESC connected- 
could be it's "making" 4.5V. Check/ask this particular clone users/manufacturer.

To fully disable GPS you have to do it in the config file and rebuild/reload 
firmware.
MP is made for ArduCopter I am not sure if disabling by it's menu will do the 
same as disabling in the config file.

Guys, I do not have this issue as most of people - you have to make tests and 
isolate when it happen /or when not/then I can go with the code.

But anyway it seems Alex will be ready with 3.0 version shortly ....

Original comment by pa...@kulma.pl on 6 Sep 2013 at 10:30

GoogleCodeExporter commented 8 years ago 
More weird things. Motors rev up in ground when armed and in stabilize mode, 
random revups without touching my transmitter! Time to reinstall all systems 
and codes again.

Where we can see Alex 3.0 status? Does it take fev days, weeks, months?

Thank you for fast support here.

Original comment by jarmo.ha...@gmail.com on 7 Sep 2013 at 10:15

GoogleCodeExporter commented 8 years ago 
I've had those random rev ups when sitting on the ground armed with transmitter 
throttle at zero (probably less than 0.25s of motors energised) - but again 
I've had them yesterday on 2.8 r3 and today on 2.9 1.b.  I have just used end 
point adjust to reduce my throttle minimum and haven't yet seen any more but 
need more testing.  Have the rev ups in flight with no gyro fitted but have yet 
to disable in config.

Original comment by nrlchamb...@gmail.com on 7 Sep 2013 at 12:50

GoogleCodeExporter commented 8 years ago 
Just finished to re-installing all, different laptop, fresh arduino 1.0.3, 
eeprom clearing, latest megapirate-add-ons. Same issue still here, random 
revups this time indoors when holding in copter on my hand. I watched tuning 
window 5 minutes when copter was sitting still on table connected to computer. 
I think there was one random acc peak, I missed to save screenshot. 

At hand testing, both front motors do that rev up. Same as my earlier setup. If 
I move copter little then revups can be seen about every 5-10 seconds!

I want to figure out this somehow, is next step disconnect GPS and compile new 
code with GPS_NONE define? I try this next.

Original comment by jarmo.ha...@gmail.com on 8 Sep 2013 at 12:48

GoogleCodeExporter commented 8 years ago 
I removed GPS unit, and defined GPS_NONE but same rev ups are still here :(

Original comment by jarmo.ha...@gmail.com on 8 Sep 2013 at 1:06

GoogleCodeExporter commented 8 years ago 
Pls attach a t-log file from Mission Planner here, I will look into it.
I am have a poor net access for few days so pls google for how to collect that 
data :)

Original comment by pa...@kulma.pl on 8 Sep 2013 at 11:48

GoogleCodeExporter commented 8 years ago 
I have the same issues with MPNG 2.9.1b on HK MultiWiiPro board in QuadV 
configuration. Not seen on MPNG 2.8R3 before.
I did reproduce the issue by simply connecting the board to USB at my desk. No 
battery connected. Arm with RC and wait. After several seconds I see CH1-CH4 go 
up simultaniously for appr. 1s in plot on Tuning Window. That happens 0-3 times 
before self disarmed on several tests. I did look at all available variables 
from "Graph This" couldn't find any correlation. Especially not with CHxIn. Are 
there more variables in the tlog file compared to what one could select in 
"Graph This"?
Anyway I attach a short tlog file from one event shortly after arming (@5865s).

Original comment by peter.r....@googlemail.com on 10 Sep 2013 at 8:56

Attachments:

GoogleCodeExporter commented 8 years ago 
Will look closer in few day as I am a little off the civilization right now ....
Self disarm here should happen here after /as I remember/ about 20sec as the AC 
safety feature.

Guys I assume all of you are using Std Rx configuration not CPPM ???

Original comment by pa...@kulma.pl on 11 Sep 2013 at 9:33

GoogleCodeExporter commented 8 years ago 
Reading through the post I assume so. At least I use Std Rx. Looking into the 
source I expect problems with throttle_mode = THROTTLE_MANUAL_TILT_COMPENSATED. 
Not finally sure so... But in get_pilot_desired_throttle() I don't like that 
THROTTLE_MIDDLE is defined as 500 and 500 not THROTTLE_MIDDLE is used in the 
function, besides that trottle middle  can be set in Mission planner as a 
variable.... I try to set STABILZE_THROTTLE back to THROTTLE_MANUAL as in MPNG 
2.8 and see if it helps.

Original comment by peter.r....@googlemail.com on 11 Sep 2013 at 10:22

GoogleCodeExporter commented 8 years ago 
Haven't played with my quad in a while but just did some updates to 2.9.1b R8 
and I am having the same issues here. I was balancing my motors today so had 
motors hooked up one at a time and all of them had the twitching while armed. 
Even running at full throttle it seems to have a periodic fluctuation. If you 
would like another tlog I can post later tonight.

Crius AIOP v1.1
2.9.1b R8
Turnigy 9x with ER9X w/ standard Rx config
RCTimer 30A Simonk ESC

Original comment by TClark8...@gmail.com on 11 Sep 2013 at 11:05

GoogleCodeExporter commented 8 years ago 
Here is a video of what mine is doing. All I am doing is arming (white LEDs 
solid) and then letting it sit without throttle input. The motors twitch at 
0:20 and 0:28.

http://www.youtube.com/watch?v=D0BT1K8UhGk

Original comment by TClark8...@gmail.com on 12 Sep 2013 at 1:50

GoogleCodeExporter commented 8 years ago 
Hm...i had this problem but with 2.9.1 r6 /r7

try this:
In MP:
Connect with USB
Save data as a Backup

In MP "Terminal":
type "setup" - wait
type "erase" - wait

Reboot AIOP:
Disconnect USB cable
Connect USB cable

(EPROM is now erased)

Arduino:
Upload data

MP: (USB still connected)
Restore backup
Check Radio calibration
Calibrate Arducopter Level 

Done

 for me this was the solution :)
Soon i will try Paku version of 2.9.1 r8(160)
Crius AIO pro V1

Original comment by DejanMat...@gmail.com on 12 Sep 2013 at 4:22

GoogleCodeExporter commented 8 years ago 
Hi8 guys,

I am back for a few days and will try to find the solution. The only problem is 
that I am not able to easily build a Std Rx configuration ... so would like to 
ask you to check if the same twitching could be seen after arming using the MP 
ARM button.

I just have to reproduce the failure first ...

Answer would speed up my work :)

BTW Do you think it's worth to work on this build as Alex is working on 3.0 now 
???

Original comment by pa...@kulma.pl on 13 Sep 2013 at 10:30

GoogleCodeExporter commented 8 years ago 
Some progress:
By adding 
#define STABILZE_THROTTLE THROTTLE_MANUAL 
into APMConfig.h I was able to reduce my big (1s high throttle) pulses to a 
very low level of twitches. Actual I can fly without danger :-)
Also in Tuning window I observed that ch3percent is always rising prior to my 
motor outputs rising..... Not sure what ch3percent is ...

I couldn't find the MP ARM button, where is it in MP?

I think it's worth working on this issue because:
- you don't know which other issues will show up in 3.0 :-)
- release date of 3.0 is not known and releases not public available without 
paying Alex (or wait...)

Original comment by peter.r....@googlemail.com on 14 Sep 2013 at 10:48

GoogleCodeExporter commented 8 years ago 
Under the HUD you should have the Actions tab and Arm/Disarm button on it.

Original comment by pa...@kulma.pl on 14 Sep 2013 at 11:36

GoogleCodeExporter commented 8 years ago 
Hi guys,

I am looking for a little help testing my new project.
If you have few minutes free pls follow this post

http://www.rcgroups.com/forums/showthread.php?t=2083451&pp=40#post27285744

Paku

Original comment by pa...@kulma.pl on 26 Jan 2014 at 9:55

GoogleCodeExporter commented 8 years ago 
Dear friends, I am having the same problem using AIO Pro V2, running Multiwii. 
Engine is oscillating, the two motors accelerate and decelerate the other two 
vagarozamente. Managed to find a solution? Use radio Turnigy 9XR Pro receiver 
and X8R Frsky. Motor GT2215 / 12 EMAX, EMAX ESC 20A, 3S Battery 5000ma.

Original comment by cidad...@gmail.com on 23 Aug 2015 at 6:21