MultiWii 2.0 bug report

This forum is dedicated to software development related to MultiWii.
It is not the right place to submit a setup problem.
Software download
Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

MultiWii 2.0 bug report

Post by Alexinparis »

Hi,

I will try to maintain a list of reported bugs.
This post can be used only to report new bugs and discuss about solutions.
One strength of multiwii if the important user number. So we should benefit from it ;)

For the moment:

- HEX problem on pro mini (reported at least 3 times)
viewtopic.php?f=8&t=1290&start=240#p11414
edit: should be corrected in 14 April dev version

- VBAT reading (reported 2 times)
viewtopic.php?f=8&t=1461

- potential bug with CRIUS LCD (reported at least 3 times, worked fine with 1.9)
viewtopic.php?f=8&t=1463#p11473

- in flight acc calibration, corrected in dev
memberlist.php?mode=viewprofile&u=471
edit: should be corrected in 14 April dev version

- potential bug with nunchuck:
viewtopic.php?f=8&t=1455&p=11561#p11561
Last edited by Alexinparis on Sat Apr 14, 2012 12:51 pm, edited 2 times in total.

airmamaf
Posts: 4
Joined: Sat Mar 31, 2012 6:47 am

Re: MultiWii 2.0 bug report

Post by airmamaf »

A problem with Nunchuck
viewtopic.php?f=8&t=1455

User avatar
dramida
Posts: 473
Joined: Mon Feb 28, 2011 12:58 pm
Location: Bucharest
Contact:

Re: MultiWii 2.0 bug report

Post by dramida »

Standard serial LCD config in multiwii 2.0 not working as it should. Strange characters and shifted words appears on lcd. To mention that on MWC1.9 the same LCD is working fine. See the clip i made this morning to show this behavior.

http://www.youtube.com/watch?v=L7TtpjYjvGo

Hardware:
Crius SE with atmega 328p and serial LCD
Software: MWC 2.0

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: MultiWii 2.0 bug report

Post by Hamburger »

What is 'standard LCD'? 3 wire sparkfun?

User avatar
UndCon
Posts: 293
Joined: Mon Feb 21, 2011 2:10 pm

Re: MultiWii 2.0 bug report

Post by UndCon »

My 3wire sparkfun works OK

the chinsese copy "crius LCD" is crap (shifted but in previous firmware as well)

JohnyGab
Posts: 144
Joined: Sat Oct 29, 2011 4:41 am

Re: MultiWii 2.0 bug report

Post by JohnyGab »

I haven investigated it, but no effect turning on/off the mag mode, on a Quad usign the HMC5883L sensor

User avatar
jevermeister
Posts: 708
Joined: Wed Jul 20, 2011 8:56 am
Contact:

Re: MultiWii 2.0 bug report

Post by jevermeister »

I found a bug in INFLIGHT_ACC_CALIBRATION:

when the calibration is done inflight it is commited by a beep with:

Code: Select all

         blinkLED(10,10,2);      //buzzer for indicating measurement finished
in line #302 of sensors.pde
this is a very bad idea inflight: blinkLED is delaying the code, hamburger warned me about that some time ago: Shame on me!!
Some users reported out of control copters, and I noticed a rise in cycle time until overflow while the speaker beeps using this function.
Perhaps we should delete the function blinkLED and replace it with a better one.

I replaced this bug in sensors.pde with:

Code: Select all

          toggleBeep = 2;      //buzzer for indicatiing the end of calibration


Nils
ps.:regarding mag mode: I use QuadX BMA020, WMP, HMC5883L,BMP085, maghold and everything working fine.

ronco
Posts: 317
Joined: Thu Aug 18, 2011 2:58 pm

Re: MultiWii 2.0 bug report

Post by ronco »

Alexinparis wrote:- HEX problem on pro mini (reported at least 3 times)
viewtopic.php?f=8&t=1290&start=240#p11414


Hi,

it may be the combination of the new soft pwm code and ppm RX!

in the new code we use one timer channel for two motors.. soft pwm is general very "other interrupt sensitiv" so it may be that the ppm interrupts do more jitter to the soft pwm then a normal rx does.

hard to explain .. the old soft pwm method starts at 125 (1ms) and ends on 250(2ms) the new starts on 5(1ms) and ends on 250(2ms).

so if the "other interrupt jitter" pushes the new soft pwm method lower then 0 (-6 for example) it switches to a lower frequency where the duty time is arround 2ms(full throttle). this cant happen with the old one because it will not fall under 0 (jitter will not be -126).

so it may be a solution to add the old method for hexas without servos.

regards Felix

User avatar
fr3d
Posts: 97
Joined: Sun Feb 06, 2011 11:21 am
Location: Cappelle la grande near the ch'ti village
Contact:

Re: MultiWii 2.0 bug report

Post by fr3d »

dramida wrote:Standard serial LCD config in multiwii 2.0 not working as it should.

mine lcd from sparkfun work fine
tested on 328 et 2560 with mc 2.0

User avatar
fr3d
Posts: 97
Joined: Sun Feb 06, 2011 11:21 am
Location: Cappelle la grande near the ch'ti village
Contact:

Re: MultiWii 2.0 bug report

Post by fr3d »

JohnyGab wrote:I haven investigated it, but no effect turning on/off the mag mode, on a Quad usign the HMC5883L sensor

same with allinone + atmega2560 tested with orginal pid

User avatar
fr3d
Posts: 97
Joined: Sun Feb 06, 2011 11:21 am
Location: Cappelle la grande near the ch'ti village
Contact:

Re: MultiWii 2.0 bug report updated 03 April 2012

Post by fr3d »

put level PiD "D" term @ 99 or less,
all other pids @re originals

active stable mode
after 60/120 secondes the copter go down slowly ..... until it touch the grass
impossible to takeoff until replug batteries.

test and tell me (bma020+wmp+bmp085)
put level D @ 100 the pb doesn't appear.
plz test in the grass @ low altitude
I've changed bma020 address @ 0x82
I've pluged wmp+ and bmp085 on bma in order to be have a vcc and i2c @ 3volts in this way bma give me a good llc with arduino @ 5volts. No i2c error if connected with a battery. board is jussih v1.11
I've tested with mw 2.0pre1(ve r201) and multiwii 2.0 (ver 20)

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report updated 03 April 2012

Post by Alexinparis »

fr3d wrote:put level PiD "D" term @ 99 or less,
all other pids @re originals

active stable mode
after 60/120 secondes the copter go down slowly ..... until it touch the grass
impossible to takeoff until replug batteries.

test and tell me (bma020+wmp+bmp085)
put level D @ 100 the pb doesn't appear.
plz test in the grass @ low altitude
I've changed bma020 address @ 0x82
I've pluged wmp+ and bmp085 on bma in order to be have a vcc and i2c @ 3volts in this way bma give me a good llc with arduino @ 5volts. No i2c error if connected with a battery. board is jussih v1.11
I've tested with mw 2.0pre1(ve r201) and multiwii 2.0 (ver 20)


level D should only limit the ACC correction.
difficult to investigate...

User avatar
fr3d
Posts: 97
Joined: Sun Feb 06, 2011 11:21 am
Location: Cappelle la grande near the ch'ti village
Contact:

Re: MultiWii 2.0 bug report updated 03 April 2012

Post by fr3d »

Alexinparis wrote:
fr3d wrote:put level PiD "D" term @ 99 or less,
all other pids @re originals

active stable mode
after 60/120 secondes the copter go down slowly ..... until it touch the grass
impossible to takeoff until replug batteries.

test and tell me (bma020+wmp+bmp085)
put level D @ 100 the pb doesn't appear.
plz test in the grass @ low altitude
I've changed bma020 address @ 0x82
I've pluged wmp+ and bmp085 on bma in order to be have a vcc and i2c @ 3volts in this way bma give me a good llc with arduino @ 5volts. No i2c error if connected with a battery. board is jussih v1.11
I've tested with mw 2.0pre1(ve r201) and multiwii 2.0 (ver 20)


level D should only limit the ACC correction.

difficult to investigate...


it's like drift.... but not in roll/pitch but with power
the idea it's keeping copter stable....
an autolanding eastereggs ?

User avatar
dramida
Posts: 473
Joined: Mon Feb 28, 2011 12:58 pm
Location: Bucharest
Contact:

Re: MultiWii 2.0 bug report

Post by dramida »

The serial LCD from Crius was working fine in MWC 1.9 but in MWC 2.0 dose not work anymore. This is definetly a BUG, regardless of your opinion about LCD quality.
The demonstration of the two software MWC version and the BUG is shown here http://www.youtube.com/watch?v=L7TtpjYjvGo


UndCon wrote:My 3wire sparkfun works OK

the chinsese copy "crius LCD" is crap (shifted but in previous firmware as well)

klaudius666
Posts: 14
Joined: Thu Apr 05, 2012 5:59 pm
Contact:

Re: MultiWii 2.0 bug report

Post by klaudius666 »

Same problem on Crius LCD with multiwii 2.0.
Looked at difference between 1.9 and 2.0 lcd code, but there are really too many change.
Perhaps a bad serial transmission, or character set problem ?

If someone give me some suggestion i can recompile and test for some solutions

Joachim08
Posts: 31
Joined: Sat Mar 17, 2012 10:10 am

Re: MultiWii 2.0 bug report

Post by Joachim08 »

More and more people report about the toilet bow effect they experience with 2.0 and ACC.
Alex can you seriously look into the matter ? Changing PID simply doesnt help.
In 1.9 there is no such behaviour.

Thanks

Joachim

kalle123
Posts: 106
Joined: Sun Oct 09, 2011 10:07 am

Re: MultiWii 2.0 bug report

Post by kalle123 »

1. Failsave settings

#define FAILSAVE_OFF_DELAY - no motor STOP after xx sec.

2. When activating alternative motor stop by uncommenting #define MOTOR_STOP

Failsave settings NOT working.

BR KH

kalle123
Posts: 106
Joined: Sun Oct 09, 2011 10:07 am

Re: MultiWii 2.0 bug report

Post by kalle123 »

kalle123 wrote:1. Failsave settings
#define FAILSAVE_OFF_DELAY - no motor STOP after xx sec.
2. When activating alternative motor stop by uncommenting #define MOTOR_STOP
Failsave settings NOT working.
BR KH


Sorry, but I think I found the reason ....

Using JETI REX MDP RX.
Attachments
1.jpg

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

Joachim08 wrote:More and more people report about the toilet bow effect they experience with 2.0 and ACC.
Alex can you seriously look into the matter ? Changing PID simply doesnt help.
In 1.9 there is no such behaviour.

Thanks

Joachim


Really strange - it should only affect the differential between motors e.g between left/right not all at same time. I don't see how it can from the code. Really is a minor limit change.

I have flown many flights with it on start to finish. No such issues. Is Baro on at same time and causing some strange effect? I've had mine on by mistake and wondered why I couldn't fly it level....

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

dramida wrote:The serial LCD from Crius was working fine in MWC 1.9 but in MWC 2.0 dose not work anymore. This is definetly a BUG, regardless of your opinion about LCD quality.
The demonstration of the two software MWC version and the BUG is shown here http://www.youtube.com/watch?v=L7TtpjYjvGo


UndCon wrote:My 3wire sparkfun works OK

the chinsese copy "crius LCD" is crap (shifted but in previous firmware as well)


Don't leave lcd in when reprogramming..... Same happened to me twice. I had to reset LCD to default speed of 9600. I posted in LCD thread earlier

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

I think the two Sirius mag defines are incorrect. Could not get it to look good in GUI and gave me weird yaw issues (as you might expect :roll: )
I think the following is correct. Not flown, but it looks good in GUI - needs a recal after change....

#define MAG_ORIENTATION(X, Y, Z) {magADC[ROLL] = X; magADC[PITCH] = Y; magADC[YAW] = -Z;}

User avatar
jevermeister
Posts: 708
Joined: Wed Jul 20, 2011 8:56 am
Contact:

Re: MultiWii 2.0 bug report

Post by jevermeister »

You should use the method descriped in the faq to test if the sensour orientation is correct.I had a very hard time finding the correct orientation via trial and error until someone shoeed me the method in the faq :-).
Sometimes all goes back to RTFM. I do these mistakes again and again *sigh*
Nils

ApoC
Posts: 31
Joined: Fri Feb 10, 2012 2:16 pm

Re: MultiWii 2.0 bug report

Post by ApoC »

viewtopic.php?f=8&t=1480

This is the Problem, what i have with 2.0. And its only in 2.0 so, that i have the Errors. But i cant get this "2.0 flashed Arduino" back to work with an older version, like 1.9 or 1.8.

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

You are right - I should have ! :)
But why make it easy? Always read the manual last right !! I didn't know that was written down, but it is a good idea. Sort of thing real easy to forget over time when a new board comes along.

Actaully I don't usually use the mag - first time I really looked at the code even. The other sensors are dead easy to figure out, but this one did take a little longer.

Anyway - tested and it flew great. Vid to follow :D

jevermeister wrote:You should use the method descriped in the faq to test if the sensour orientation is correct.I had a very hard time finding the correct orientation via trial and error until someone shoeed me the method in the faq :-).
Sometimes all goes back to RTFM. I do these mistakes again and again *sigh*
Nils

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

Start with basics.
2.0 should be fine

first try with just WMP
to do that use I2C speed of 100hz. All specific boards and independant sensors disabled.

If good:
Now add specific sensor bma020
//#define BMA020

If no good, try alternative address for BMA...
#define BMA180_ADDRESS 0x80




ApoC wrote:http://www.multiwii.com/forum/viewtopic.php?f=8&t=1480

This is the Problem, what i have with 2.0. And its only in 2.0 so, that i have the Errors. But i cant get this "2.0 flashed Arduino" back to work with an older version, like 1.9 or 1.8.

ApoC
Posts: 31
Joined: Fri Feb 10, 2012 2:16 pm

Re: MultiWii 2.0 bug report

Post by ApoC »

Hey

Im not new in MWii.I know how to setup the conf. That is not the Problem.

User avatar
shikra
Posts: 783
Joined: Wed Mar 30, 2011 7:58 pm

Re: MultiWii 2.0 bug report

Post by shikra »

ah - it's hard to tell knowledge level.
Is it possible to separate the wmp and BMA and recompile to test individually? Maybe one of devices has broken and is affecting the other?

Magnetron
Posts: 124
Joined: Tue Jul 05, 2011 4:32 pm

Re: MultiWii 2.0 bug report

Post by Magnetron »

I Alex,
some user report on italian forum that with MultiWii v2.0 on Nano with hexa configuration seems that motor 5 and 6 have differents speed. With v1.9 seems that this problem was not present...

ronco
Posts: 317
Joined: Thu Aug 18, 2011 2:58 pm

Re: MultiWii 2.0 bug report

Post by ronco »

Magnetron wrote:I Alex,
some user report on italian forum that with MultiWii v2.0 on Nano with hexa configuration seems that motor 5 and 6 have differents speed. With v1.9 seems that this problem was not present...



hi Magnetron,

im not Alex but i did the changes to the promini with hexa..
thay dont have a different speet but the signal range is different now .. it starts end ends higher then it was in privius MWC versions .. but now thay are similar to the motors 1-4 means 1000 -1850 bevore it was ~950- 1800 ms.

so relearning the throttle range to the ESCs of motor 5 & 6 should fix this

regards

Felix

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report

Post by Alexinparis »

ronco wrote:
Magnetron wrote:I Alex,
some user report on italian forum that with MultiWii v2.0 on Nano with hexa configuration seems that motor 5 and 6 have differents speed. With v1.9 seems that this problem was not present...



hi Magnetron,

im not Alex but i did the changes to the promini with hexa..
thay dont have a different speet but the signal range is different now .. it starts end ends higher then it was in privius MWC versions .. but now thay are similar to the motors 1-4 means 1000 -1850 bevore it was ~950- 1800 ms.

so relearning the throttle range to the ESCs of motor 5 & 6 should fix this

regards

Felix


Hi Felix
For some reasons, some ESCs need to have a lower min value even after a good calibration.
This is the purpose of MINCOMMAND.
with the current code for 5,6,7,8 motors on a promini, it's not possible.
I think we should rethink the code to make this possible.

ronco
Posts: 317
Joined: Thu Aug 18, 2011 2:58 pm

Re: MultiWii 2.0 bug report

Post by ronco »

Alexinparis wrote:Hi Felix
For some reasons, some ESCs need to have a lower min value even after a good calibration.
This is the purpose of MINCOMMAND.
with the current code for 5,6,7,8 motors on a promini, it's not possible.
I think we should rethink the code to make this possible.


Hi Alex,
should be possible now for hexas without servos -> viewtopic.php?f=8&t=1246&p=12066#p12066

regards felix

tru168
Posts: 3
Joined: Fri Apr 13, 2012 5:00 pm

Re: MultiWii 2.0 bug report

Post by tru168 »

Hi,

I'm using BMA020 and WMP, works all version but when loaded with version 2.0 , I got I2C error. unselect the BMP020 in sketch and WMP works perfectly.

One weird thing happen if i use v2.0, when I check on my GUI, the I2C error keep counting fast , but when I turn my copter upside down, the I2C error stop counting and everything works again, turn my copter slowly, when it pass 90 degree and it start counting I2C error again.

IT IS A BUG for sure , I have a few boards fitted with WMP and BMA020, some are homebrew, and some are PARIS board. All of them behave the same as mentioned above.

Now I can't fly all my copters , unless I downgrade to lower version.

Joachim08
Posts: 31
Joined: Sat Mar 17, 2012 10:10 am

Re: MultiWii 2.0 bug report

Post by Joachim08 »

shikra wrote:
Joachim08 wrote:More and more people report about the toilet bow effect they experience with 2.0 and ACC.
Alex can you seriously look into the matter ? Changing PID simply doesnt help.
In 1.9 there is no such behaviour.

Thanks

Joachim


Really strange - it should only affect the differential between motors e.g between left/right not all at same time. I don't see how it can from the code. Really is a minor limit change.

I have flown many flights with it on start to finish. No such issues. Is Baro on at same time and causing some strange effect? I've had mine on by mistake and wondered why I couldn't fly it level....


Unfortunately i am not a coding specialist... On the German fpv forum you can find a 5 page thread about the tbe effect. Lots of pilots have this problem.
It is not related to the sensors or IMU, even with WMP and BMA020 it shows up... So Baro is off or not existing.
It is really strange as we are forced to use 1.9 where this issue is non existend. Some of us even cannot use 1.9 anymore as we have the MPU6050 sensor which is only supported in 2.0.

So our quads are grounded at the moment.

ronco
Posts: 317
Joined: Thu Aug 18, 2011 2:58 pm

Re: MultiWii 2.0 bug report

Post by ronco »

tru168 wrote:Hi,

I'm using BMA020 and WMP, works all version but when loaded with version 2.0 , I got I2C error. unselect the BMP020 in sketch and WMP works perfectly.

One weird thing happen if i use v2.0, when I check on my GUI, the I2C error keep counting fast , but when I turn my copter upside down, the I2C error stop counting and everything works again, turn my copter slowly, when it pass 90 degree and it start counting I2C error again.

IT IS A BUG for sure , I have a few boards fitted with WMP and BMA020, some are homebrew, and some are PARIS board. All of them behave the same as mentioned above.

Now I can't fly all my copters , unless I downgrade to lower version.


Hi,

did you fly this boards with 1.9?

and do you have external pullup resistors ?

regards Felix

User avatar
fr3d
Posts: 97
Joined: Sun Feb 06, 2011 11:21 am
Location: Cappelle la grande near the ch'ti village
Contact:

Re: MultiWii 2.0 bug report

Post by fr3d »

playing with 2.0 wmp+ & bma020 & bmp085 without pb in acro. some trouble in stable mode...after flying a few minutes, quad drift, Quad can't find is ealier plane position,and motors slowdown if you change level d term pid.
I've modified bma like this
add 2.2 k resistor on i2c lines with 5v,
plug wmp+ & bmp085 with 3.3v,
scl & sda solder inside bma020.
config.h external pullup, wp @ 100khz, and bma_180_address @ 0x82
Attachments
fr3d_using_bma_llc.jpg

crashlander
Posts: 506
Joined: Thu May 05, 2011 8:13 am
Location: Slovenia

Re: MultiWii 2.0 bug report

Post by crashlander »

Hello,
I'm starting to implement MultiWii on flying-wing (Arduino Mega, ATAVRSBIN) but my right servo midpoint starts at 2000 (pic1)
pic1.png
PIC1
(12.05 KiB) Not downloaded yet
(in GUI and in real life), and only moves down to 1500 when I move elevator in maximum position (pic2)
pic2.png
PIC2
(9.64 KiB) Not downloaded yet
.
The behavior is same in pass-throu and in giro assisted mode.
My ugly fix is to change line 747 in Output from:

Code: Select all

servo[1]  = constrain(servo[1] + wing_right_mid, WING_RIGHT_MIN, WING_RIGHT_MAX );

to

Code: Select all

servo[1]  = constrain(servo[1] + wing_right_mid-500, WING_RIGHT_MIN, WING_RIGHT_MAX );

But that is probably not how it was meant originally.

Regards
Andrej

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: MultiWii 2.0 bug report

Post by Hamburger »

Andrej,
you must reset the eeprom. The code is interpreting some garbage from eeprom. This makes the midpoint go crazy.
In Arduino IDE in the examples is a small sketch for resetting the eeprom.
Or you could increment the eeprom version number in eeprom.pde
Hamburger

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report

Post by Alexinparis »

ronco wrote:
Alexinparis wrote:Hi Felix
For some reasons, some ESCs need to have a lower min value even after a good calibration.
This is the purpose of MINCOMMAND.
with the current code for 5,6,7,8 motors on a promini, it's not possible.
I think we should rethink the code to make this possible.


Hi Alex,
should be possible now for hexas without servos -> viewtopic.php?f=8&t=1246&p=12066#p12066

regards felix


ok, this bug should now not appear anymore with the dev release 14 April

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report

Post by Alexinparis »

Joachim08 wrote:More and more people report about the toilet bow effect they experience with 2.0 and ACC.
Alex can you seriously look into the matter ? Changing PID simply doesnt help.
In 1.9 there is no such behaviour.

Thanks

Joachim

As long as I can't reproduce this problem, I can't help
Main difference between 1.9 and 2.0 is the sensor orientation definition, and LEVEL D which should be left to 100 to have the 1.9 behavior.

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report

Post by Alexinparis »

tru168 wrote:Hi,

I'm using BMA020 and WMP, works all version but when loaded with version 2.0 , I got I2C error. unselect the BMP020 in sketch and WMP works perfectly.

One weird thing happen if i use v2.0, when I check on my GUI, the I2C error keep counting fast , but when I turn my copter upside down, the I2C error stop counting and everything works again, turn my copter slowly, when it pass 90 degree and it start counting I2C error again.

IT IS A BUG for sure , I have a few boards fitted with WMP and BMA020, some are homebrew, and some are PARIS board. All of them behave the same as mentioned above.

Now I can't fly all my copters , unless I downgrade to lower version.


I have a quad with a WMP clone + BMA020.
It works perfectly for me with 2.0.
I2C errors should have nothing to do with the multi orientation.

User avatar
jevermeister
Posts: 708
Joined: Wed Jul 20, 2011 8:56 am
Contact:

Re: MultiWii 2.0 bug report

Post by jevermeister »

Okay, I don't know if it is a bug with 2.0 because I flew some lipos without this problem:

My copter stutters and drops a few centimeters and banks to left to right and front and back randomly and this lead to a crash two times.

I was lucky that I was fas flying low at the first crash in FVP.

Here is a video ( do not laugh because of the missing front legs, they got killed in the first crash :-/)

All these small glitches are not me...
http://youtu.be/6nFd-ZBJeNE

Is this a Software problem or do I have some loose wires? I swapped the batteries of my TX but this did not help, even a new LIPO does not help...

Nils

ps.: I will post this into RCgroups too, so please do not object.

User avatar
Hamburger
Posts: 2578
Joined: Tue Mar 01, 2011 2:14 pm
Location: air
Contact:

Re: MultiWii 2.0 bug report

Post by Hamburger »

Nils
2.0 is not very precise or do yoh really use the official release?
Better name the revision from svn you use.

Quadro-Fan2011
Posts: 3
Joined: Tue Oct 11, 2011 11:39 am

Re: MultiWii 2.0 bug report - OCTOFLATP roll pitch error

Post by Quadro-Fan2011 »

Hi All,

this is my first of hopefully many comments and discussions. I have no education in prgramming so many many thanks to all of you who are able to know what "what if else means" ;)

Here is my problem. I built up an octocopter with a multiwii board (MWC MEGA beta). I uploaded the 2.0 Version and selected Octoflat p because it is a plus octo with one arm in front. After calibration I tested roll and pitch and it was wrong/switched. The arrow on the board for flight direction shows correctly to the front arm but the pitch/roll is wrong.

Then I tried the Version 1.9 and with this it works. Pitch and Roll is correct now. Foirst flight test was great. So I think this is a mistake in the 2.0

All Versions I downloaded from here
http://code.google.com/p/multiwii/downloads/list

By the way in the config tool the octo is named octox. The same in Version 2.0.

Best wishes

Michael

User avatar
jevermeister
Posts: 708
Joined: Wed Jul 20, 2011 8:56 am
Contact:

Re: MultiWii 2.0 bug report

Post by jevermeister »

Hamburger wrote:Nils
2.0 is not very precise or do yoh really use the official release?
Better name the revision from svn you use.


Hi hamburger.
Thank you for your answer.
I use the official MW 2.0 release.I downloaded it via the link.It is no svn revision.
I checked my connecftions but I could not find a loose connections by having a look at the gui while manopulating the connections.

Nils

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report - OCTOFLATP roll pitch error

Post by Alexinparis »

Quadro-Fan2011 wrote:Hi All,

this is my first of hopefully many comments and discussions. I have no education in prgramming so many many thanks to all of you who are able to know what "what if else means" ;)

Here is my problem. I built up an octocopter with a multiwii board (MWC MEGA beta). I uploaded the 2.0 Version and selected Octoflat p because it is a plus octo with one arm in front. After calibration I tested roll and pitch and it was wrong/switched. The arrow on the board for flight direction shows correctly to the front arm but the pitch/roll is wrong.

Then I tried the Version 1.9 and with this it works. Pitch and Roll is correct now. Foirst flight test was great. So I think this is a mistake in the 2.0

All Versions I downloaded from here
http://code.google.com/p/multiwii/downloads/list

By the way in the config tool the octo is named octox. The same in Version 2.0.

Best wishes

Michael


Hi,
I don't know the board you use and its sensors, but it seems that it is not a predefined one.
The sensor orientations changed in 2.0, it might explain your problem.
The GUI shows the same indication octo X for all octo configs. It's normal because the GUI is not fully implemented for 8 motor configs.

Alexinparis
Posts: 1630
Joined: Wed Jan 19, 2011 9:07 pm

Re: MultiWii 2.0 bug report

Post by Alexinparis »

jevermeister wrote:Okay, I don't know if it is a bug with 2.0 because I flew some lipos without this problem:

My copter stutters and drops a few centimeters and banks to left to right and front and back randomly and this lead to a crash two times.

I was lucky that I was fas flying low at the first crash in FVP.

Here is a video ( do not laugh because of the missing front legs, they got killed in the first crash :-/)

All these small glitches are not me...
http://youtu.be/6nFd-ZBJeNE

Is this a Software problem or do I have some loose wires? I swapped the batteries of my TX but this did not help, even a new LIPO does not help...

Nils

ps.: I will post this into RCgroups too, so please do not object.


It looks like a bad ESC+motor combination.
could you test the motor stop stress test ? : take the multi firmly with one hand, start the motor to mid-gaz and tilt the multi violently to test the motor response.
If one motor stops, it's not good.

Quadro-Fan2011
Posts: 3
Joined: Tue Oct 11, 2011 11:39 am

Re: MultiWii 2.0 bug report

Post by Quadro-Fan2011 »

Hi Alex thank you for that.

The board has no name but it has all sensors described in the sketch. So I can use them. If you say that the sensor directions has changed is that for gyro and acc? and why did you do that? Will it work if I switch the board?

I have an old copter with bma020 - the software is 1.7. That means that I have to change the sensor direction first if I want to use the 2.0?

Greetings

Michael

User avatar
jevermeister
Posts: 708
Joined: Wed Jul 20, 2011 8:56 am
Contact:

Re: MultiWii 2.0 bug report

Post by jevermeister »


It looks like a bad ESC+motor combination.
could you test the motor stop stress test ? : take the multi firmly with one hand, start the motor to mid-gaz and tilt the multi violently to test the motor response.
If one motor stops, it's not good.



I had this with my old HK SS 18A. One of them stops after this maneuver.
I switched to hobbywing pentium 18A.
Just djd the stress test wirh them. THey do not stop.
Could it be a loose I2C or rx connector.You can see in the video,that the glitch appears on all 4 motors.The copter drops a few cm.

I start to loosr faith here. I cannot trust my copter anymore. What if he drops from 100m with gopro onboard...
Nils

LenzGr
Posts: 166
Joined: Wed Nov 23, 2011 10:50 am
Location: Hamburg, Germany
Contact:

Re: MultiWii 2.0 bug report

Post by LenzGr »

jevermeister wrote:Could it be a loose I2C or rx connector.You can see in the video,that the glitch appears on all 4 motors.The copter drops a few cm.

I start to loosr faith here. I cannot trust my copter anymore. What if he drops from 100m with gopro onboard...


It could be several things. Maybe a bad solder joint on one of the ESCs or the connecting power or servo cables? Did you try re-calibrating the ESCs?
Or maybe the timing is too high? IIRC, the 2.0 firmware had some optimizations related to the PWM output that drives the ESCs. Could be they are now getting saturated (using an alternative firmware might help).

Alternatively, check the motor bearings - do they all run smoothly without any resistance?

Joachim08
Posts: 31
Joined: Sat Mar 17, 2012 10:10 am

Re: MultiWii 2.0 bug report

Post by Joachim08 »

Alexinparis wrote:
Joachim08 wrote:More and more people report about the toilet bow effect they experience with 2.0 and ACC.
Alex can you seriously look into the matter ? Changing PID simply doesnt help.
In 1.9 there is no such behaviour.

Thanks

Joachim

As long as I can't reproduce this problem, I can't help
Main difference between 1.9 and 2.0 is the sensor orientation definition, and LEVEL D which should be left to 100 to have the 1.9 behavior.


Level D is set to 100 as standard in 2.0 and the behavior is totally different from 1.9 .... I have tried to change values up and down without luck.

Post Reply