scottlawsonbc / wii-esc Goto Github PK

- correct calculations of on/off times, the 0xFF value for timer means 1 cycle. 
Probably better to use NEG instead of COM
- process last 1 or 2 steps without exiting ISR
- in case of no major improvement, rollback changes.. ;)

What steps will reproduce the problem?
1. Flash Plush30A firmware to HobbyWing 40A ESC
2. Power up the ESC

What is the expected output? What do you see instead?

I expect some sort of boot up beep sequence

Hi there,

if I compile the project, either in CodeBlocks or using make -f Makefile.avr in 
Ubuntu 12.10 or current Arch Linux, I get the following compile error:


lib-avr/arduino/HardwareSerial.cpp:120:15: error: attempt to use poisoned 
"SIG_UART_RECV"
lib-avr/arduino/HardwareSerial.cpp:133:4: error: #error No interrupt handler 
for usart 0

Am I missing some configuration option or defines?

I'm using
avr-gcc (GCC) 4.7.0 in Ubuntu 12.10
avr-gcc (GCC) 4.7.2 in Arch Linux

Thanks, Julian

To Archive maximum torque, the angle between the stator and rotor magnetic 
fields must be in 90deg±30deg

- Normal align (0 deg)
- 2 short commutations (probably CommStart/8)
- Normal startup with BEMF (90 deg)

Slowly decrease the input down to 900us (with servotester) The motor is making 
(little) sound (startup active?).

If I decrease fast everything is shut down correctly. 

I flashed a ZTW12a with bs_nfet, OneShot-Option and activated Bootloader via 
USB TurnigyLinker.

I've get no error only ok messages.

Since that time, I can't flash this ESC anymore (with BlHeli). I've read fuses 
from other usable ZTWs (low 2e high cf lock ff) and write it manually via ISP. 
I can't reflash they via BlHeli. Then I reflash they manually via ISP with 
blheli-hex, but I can't connect it anymore - only with ISP I could read, fuses, 
eeprom, hex;

Any Ideas?

I think the chip is to heaven...

The "FET Control" part could be changed from 
#define ApFET_on  cbi  PORTB,3
to
#define ApFET_on  cbi  PORTB,ApFET

I've lots of ESC's lying around whilst I was building my octo (which turned 
hexa). Have been flying happily with SimonK firmware for a while (never used 
bell-motors)

Steps followed
1a.Upload latest (conventional) firmware via supplied tool (ESC's have been 
selected on Pads, but used the Atmel header-tool)
1b.Upload MultiWii throttle settings
2. Removed BEMF's (from my F30 and F40's, couldn't find on TGY Plush/Sentry 
18A's)
3. Solder motors to ESC
4. Spool-up
4a.Watch (great!) throttle response
4b.Listen to annoying winding/grinding noise
4c.Shut down, check motors on SimonK ESC...
5. Relative silence!
6. Upload wii-ESC Complementary PWM version
7. Spool up
8. Scared of motors going to s**t
9. Continue flying with SimonK (for now)

Will try this with D3536 motor's I've not connected and (old) 2213's I've lying 
around. But also KDA's made same weird sound.

Normal? BTW It's not cogging, they spool-up and respond great. Either via FC or 
RX. When I say normally, initial start-up takes quite some throttle, even via 
regular RX. When throttle down, motor stops, WOT again, motor only starts after 
certain throttle point.

Have video to show this (and compare), can upload if necessary.

Hi, I'm trying to open the project and it doesn't do anything. I've downloaded 
all the files from the source (all header files and so on),I've changed the 
compiler to avr-gcc and I've even made the same file hierarchy as in the 
repositorybut I just get a "Opening D:\Hobby\CodeBlocks\wii_esc_ng.cbp" message 
in Codeblocks log and it's like I didn't open anything.
What am I doing wrong? Is there another way or must I make a new project and 
start adding files?

Hello!

Your firmeare does not fit for new HobbyKing RedBrick 50a BEC. It is based on 
mega48V.

Can you adopt your firmware for it?

Tehnical desription:
=== BOM ===
1x ATmega48v
3x IR2102
12x P0603BD N-Channel MOSFET

=== Defenitions ===
H = IR2102 PIN 2 (!HIN)
L = IR2102 PIN 3 (!LIN)
S = from phase wire over 47k+(47k to fround)

=== RC transmitter PWM INPUT ===
PD2 over 4.7k

=== INTERRUPT ===
PD6
from a/b/c phase wire over 3x47k+(3x47k to ground)+3x47k

=== PHASE A ===
H - PD0
L - PD1
S - PC2

=== PHASE B ===
H - PD4
L - PD3
S - PC3

=== PHASE C ===
H - PD5
L - PD7
S - PC4

This method has an advantages:
1) High sensitivity
2) Not require filterind

Disadvantages:
1) Bad at short off cycles (high freq or high power)

Todo:
1) implement method
2) automatically switch between methods
   off-cycle >= 10-20 ticks (off-cycle long enough to stabilize current)
   zc_time   >= power_range (we have atleast one transition) 

Hello

Could it be possible to add a setting for compiling a reverse firmware like for 
Simonk ? 

I usually solder motor wires to esc directly, it's really usefull to flash them 
with the bootloader & linker once everything is wired to reverse motor 
rotations.

Regards.

What steps will reproduce the problem?
1. Flash wii-esc firmware
2. Start increasing throttle on several ESCs

What is the expected output? What do you see instead?

Expected that all ESCs would start at the same time, but instead some seem to 
start at a different PWM input. I think this has to do with the internal 
crystal in the ESC...? 

Please provide any additional information below.

Perhaps a live throttle calibration would help? I have gotten around these 
inconsistencies on simonk's firmware by calibrating each startup.

Currenntly we have ZC blanking time: 15 deg. The recommended value by 
literature is 20 deg.

ToDo:
- Increase blanking time by 15/2=7.5
- Decrease timeout by 15/2=7.5
- Test results

Hi Dmitry,

nice rewrite, it works very well!

I tried to add some PPM signal checks, but failed with adding a power off on a 
signal loss (see filter_ppm_data() and rx_ppm_callback()).

What should be added:
- Detect signal loss (not working with my patch)
- Require some valid ppm frames to arm the ESC
- Require some valid ppm frames to start the motor

Cheers,
Heiko