paparazzi-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Paparazzi-devel] motor command error on hexa------work arround prog


From: Eduardo lavratti
Subject: Re: [Paparazzi-devel] motor command error on hexa------work arround progress
Date: Mon, 25 Mar 2013 11:31:37 -0300

Yeap ... you are right Felix ... i found that need to be configure

With 60 the bus work @ 120Khz and have time to handle all transactions but the problem continue when i use 8 motors.

can you point me where i find how to downlink variables ?





Date: Mon, 25 Mar 2013 14:55:18 +0100
From: address@hidden
To: address@hidden
Subject: Re: [Paparazzi-devel] motor command error on hexa------work arround progress

Also for the LPC it is
<configure name="MKK_I2C_SCL_TIME" value="50"/>
and not define as documented here: http://paparazzi.enac.fr/wiki/Subsystem/actuators#MKK


On Mon, Mar 25, 2013 at 2:53 PM, Felix Ruess <address@hidden> wrote:
Hi,

MKK_I2C_SCL_TIME is only valid on LPC21xx.
For the STM32, use the more generic <define name="I2C1_CLOCK_SPEED" value="400000"/> as I wrote in the previous email.

Also, the "fix" from Sergey certainly helps to not accidentally drop transactions because the queue is full.
But if you can't send all transactions (for all 6 or 8 motors) within one 512Hz cycle, it is not good to rely on this as you will send transactions with old commands.
You can check that by downlinking the value actuators_mkk.submit_err_cnt variable, which counts the number of failed I2C transaction submits.
It would be much better to really explicitly only send e.g. 4 transactions for one (symmetric) set of motors in once cycle and the other 4 in the next.

Cheers, Felix


On Mon, Mar 25, 2013 at 2:45 PM, Eduardo lavratti <address@hidden> wrote:
Sergey,
changing the scl time i change the i2c clock frequency right ?

Well, i changed the scl time from 20 to 50 and 150 and the i2c clock continue away the same 48khz.

i use this config to change scl time
<define name="MKK_I2C_SCL_TIME" value="50"/>


i to some mistake ?



From: address@hidden
To: address@hidden
Date: Mon, 25 Mar 2013 09:33:17 -0300

Subject: Re: [Paparazzi-devel] motor command error on hexa------work arround progress

I updated the my local copy yesterday. ( i am using master)
Thank for you considerations.

> To: address@hidden
> Date: Mon, 25 Mar 2013 09:14:34 +0100
> From: address@hidden
> Subject: Re: [Paparazzi-devel] motor command error on hexa------work arround progress
>
> Hi Eduardo!
> If the problem is still actuall, try to checkout the latest
> actuators_mkk.c file from the master branch.
> I've added the i2c submission control, so the actual queue length could be
> leaved unchanged.
> I've also experienced problems with 8 mkk controllers, they a very
> sensitive to the bus speed. With the 400kHz there going to be the bus
> errors anyway. I've had problems with 8 mkk sometimes even with the bus
> speed about 60kHz. So try to let your bus speed down. But when you let
> your bus speed down, and it is not sufficient to control all the motors
> once per periodic time, the queue length increase is not going to help. So
> then you have to use the fixed actuators_mkk.c file.
> But anyway, I've experienced a crash with the stock MK autopilot because
> of i2c errors problem, so imho the solution with more than 6 mkk
> controllers is not reliable enough. I wouldn't recommend that.
> Regards,
> Sergey
>
> _______________________________________________
> Paparazzi-devel mailing list
> address@hidden
> https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________ Paparazzi-devel mailing list address@hidden https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

_______________________________________________
Paparazzi-devel mailing list
address@hidden
https://lists.nongnu.org/mailman/listinfo/paparazzi-devel




_______________________________________________ Paparazzi-devel mailing list address@hidden https://lists.nongnu.org/mailman/listinfo/paparazzi-devel

reply via email to

[Prev in Thread] Current Thread [Next in Thread]