CF1.0 problems

[dg1vs]

Hallo,
I hope someone can help me.

Maybe a similar problems as described in viewtopic.php?f=5&t=1825

In addition to a CF2.0 I still have a CF1.0. The CF2.0 works very well (cfclient latest version from the website, linux system, firmware CF2.0 crazyflie2-2015.08.1.zip).

The CF1.0 is airworthy and controllable, but has still an old firmware. Unfortunately I did not manage to turn the CF1.0 together with the cfclient (latest version from the website, linux version) in the boot loader mode. The error message is __init __ () missing 1 required positional argument: 'link'. Any ideas?

Second problem…
Using the virtualbox image, I was able to update the CF1.0 to the latest firmware crazyflie1-2015.08.1.bin from the website, but now I can't fly anymore the CF1.0. The CF2.0 works great. Any ideas?

The same behavior I encounter using the latest cfclient version from git. CF2.0 work great, CF1.0 doesn't fly.

Any hints are welcome

Kind regards
Karsten

[chad]

Hi Karsten!

Welcome to the forum.

Let's try to solve your second problem first as that might solve the first problem as well.

Using the virtualbox image, I was able to update the CF1.0 to the latest firmware crazyflie1-2015.08.1.bin from the website, but now I can't fly anymore the CF1.0.

I have a couple questions for you:

1) How did you update your CF1.0. Did you use cfclient? Did you use the command line?
2) What is the behavior of the CF1.0 now (after the update) when you turn it on?

Thanks!

[dg1vs]

Morning,

1) How did you update your CF1.0. Did you use cfclient? Did you use the command line?

With the cfclient in the vm.

2) What is the behavior of the CF1.0 now (after the update) when you turn it on?

So I'm at work and can't play with the CF1.0, but as far as I remember:
* I can hear the test-sound from the motors and some led are blinking.
* I'm able connect from the cfclient to the CF1.0, but often the connection get lost.

Kind regards

Karsten

[dg1vs]

Hallo

I have checked the status of the leds and also found a working cf1.0 with an very old firm with the help of an friend.

Working CF1.0

blue led on
red blinking 2 Hz

conected
blue led on
red blinking 2 Hz
green on and flying

Not-Working CF1.0 with new firmware

the funny motorsound
blue led on
red blinking 2 Hz

then it sounds like a reboot, since I can hear the the funny motorsound after 10-20 sec again

trying to conect
blue led on
flashing 5 times an a short break of 3 sec and flashing repeatedly
green on and not flying an no data from CF1.0

Hope this helps a little bit.

Kind regards Karsten

[dg1vs]

Just in order to supply more information.

The flying CF1.0 (old firmware) don’t report any actual values to the cfclient.

Hope this helps.

Kind regards Karsten

[dg1vs]

Hallo,

can you please help me!
Btw. the behaviour is the same for compiled firmware (make all PLATFORM=CF1) and the downloaded firmware from the website. Unfortunately I downloaded the buggy firmware also to the last flying (from my friend) CF1.0. So please help. I will try help you.

Kind regards Karsten

[chad]

Not-Working CF1.0 with new firmware


blue led on
red blinking 2 Hz

then it sounds like a reboot, since I can hear the the funny motorsound after 10-20 sec again

trying to conect
blue led on
flashing 5 times an a short break of 3 sec and flashing repeatedly
green on and not flying an no data from CF1.0

Hope this helps a little bit.

Kind regards Karsten

The "funny" motor sound is a feature added to newer firmware. The firmware you flashed has that. It's a start-up chime so-to-speak.

If the Crazyflie is rebooting and you're having the 5 flashes of the red LED, that doesn't seem to be a firmware issue. The CF booted and ran, so the firmware flash is likely good. The 5 flashes of the RED led means it failed some internal test according to this wiki page. Are you sure to keep it still and on a flat surface when you power it on?

The flying CF1.0 (old firmware) don’t report any actual values to the cfclient.

That sounds peculiar but possibly a version mismatch problem (i.e. old CF1 firmware, new client).

Unfortunately I downloaded the buggy firmware also to the last flying (from my friend) CF1.0. So please help.

Are you saying that the other CF1.0 that was working now has the same behavior as yours? The firmware available here on GitHub (cf1-2015.08.1.bin ) works fine with my CF1.0 (I tried it last night) and I am using the 0.6 VM with the latest cfclient on master. The output using git log on my crazyflie-clients-python clone is:

Code: Select all

bitcraze@bitcraze-vm:~/projects/crazyflie-clients-python$ git log -1
commit ca91d168951c0d457674c525564ef1c5e9bd412e
Author: Arnaud Taffanel <arnaud@bitcraze.se>
Date:   Thu Aug 20 16:11:53 2015 +0200

    Fixed zmq and input interfaces on windows exe build

So, I've verified this version combination (CF1, firmware, and client) as working. Can you verify you have the same combination?

[dg1vs]

Hallo,
it looks very complicated.

Just in order to check the pc-side. With very cfclient on my linux pc I can connect to an CF2.0 perform some flying with the CF2.0.

I can also with any cfclient download any firmware to any of the two available CF1.0.

The batteries are charged.

If the Crazyflie is rebooting and you're having the 5 flashes of the red LED, that doesn't seem to be a firmware issue. The CF booted and ran, so the firmware flash is likely good. The 5 flashes of the RED led means it failed some internal test according to this wiki page. Are you sure to keep it still and on a flat surface when you power it on?

The rebooting stuff drives me crazy. I just touch the button, the CF1.0 boots up, an I can hear the “start-up chime so-to-speak”. But approximately after 6-10 sec I assume that the CF1.0 reboots, since I can hear the funny sound. The 5 flashes of the red LED I see _after_ the reboot. The CF1.0 is located on the table and “hands off”, after starting the CF1.0 with the button.

I have the identical behavior on both CF1.0, with the new firmware. At least one of them was able to fly. I assume both did before.

Has the boot loader some impact?
Was someday the boot loader replaced?

Console output after reboot just after power up!

Code: Select all

SYS: ----------------------------
SYS: Crazyflie 1.0 is up and running!
SYS: Build 0:61e23e51d7d3 (2015.08.1) CLEAN
SYS: I am 0x871935375552864966FFF48 and I have 128KB of flash!
CFGBLK: Verification [FAIL]
IMU: MPU6050 I2C connection [OK].
IMU: HMC5883 I2C connection [OK].
IMU: MS5611 I2C connection [OK].
MPU6050: Self test [OK].
HMC5883L: Self test [OK].
MS5611: Self test [OK].
SYS: Free heap: 1840 bytes

Console output after reboot

Code: Select all

IMU: HMC5883 I2C connection [OK].
IMU: MS5611 I2C connection [OK].
MPU6050: Self test [OK].
HMC5883L: Self test [OK].
MS5611: Self test [OK].
SYS: The system resumed after watchdog timeout [WARNING]

There are any more debug output which I can support, in order the help to catch the bug....?

Kind regards

Karsten

[chad]

Has the boot loader some impact?
Was someday the boot loader replaced?

No, I don't think this is the case. It takes some effort to smash the bootloader. Yours seems OK.

Console output after reboot just after power up!

Code: Select all

CFGBLK: Verification [FAIL]

Console output after reboot

Code: Select all

SYS: The system resumed after watchdog timeout [WARNING]

These two things look to me to be the problem... Are the Crazyflie 1's you're using 6-DOF or 10-DOF?

I did a quick forum search and found this thread. You may want to simply compile the latest crazyflie-firmware code on master branch and flash that to see if it fixes the issue. You can follow the instructions in my post on the thread you mentioned in your original post to compile the firmware. Then flash it as you normally have.

[dg1vs]

Hallo chan,

first of all, thanks a lot! The CF1.0s are flying again. That is really cool.

Anyway, I think a discussion with the developers would be useful. As far as I understand the error message indicate, that something with storing the configuration is going wrong. The

Code: Select all

CFGBLK: Verification [FAIL]

log entry is generated during start-up calling configblockInit

There are two different files (see Makefile)

Code: Select all

PROJ_OBJ_CF1 += configblockflash.o
PROJ_OBJ_CF2 += configblockeeprom.o

with the same function. One is using the flash the other is using the eeprom for storing the information. I will try to understand the code a little bit deeper. But one short question... What's the logical connection between the configuration and the watchdog?????

One problem is still open, I can not configure the CF1.0, but this an other problem. I will first search the forum.....

Again, Thanks a lot!

Karsten

PS:

Are the Crazyflie's you're using 6-DOF or 10-DOF?

Both are 10-DOF CF1.0.

PPS: Sorry for the wrong post