search

Schildkroet / GRBL-Advanced

Grbl-Advanced is a no-compromise, high performance, low cost alternative for CNC milling. This version of Grbl runs on a STM32F411RE / STM32F446RE Nucleo Board. Now with backlash compensation, multi-axis and Tool Table support!
GNU General Public License v3.0
286 stars 84 forks source link

STM32F446RE stalling #49

Closed McCodie closed 3 years ago

McCodie commented 3 years ago

Hi, did you really test it with the STM32F446RE? Because the communication just seems to stall after a few commands. I tried to build the firmware with MacOS & Windows, both gave me the same result.

$H let's it crash immediately (nothing is connected to the Nucleo board).

Schildkroet commented 3 years ago

Yes, I'm actually running it on a F446. What do you mean by crash? Whats the raw terminal ouput? How did you compile it? Your $$?

McCodie commented 3 years ago

Hi,

thanks for your quick reply!

I somewhat think the problem is the $J parameter which is submitted by the universal gcode sender. Which application do you use for submitting the gcode? Candle?

$G [GC:G0 G54 G17 G21 G90 G94 G98 M5 M9 T0 F0 S0] ok $X [MSG:Caution: Unlocked] ok $$ $0 = 0 (Step pulse time, microseconds) $1 = 50 (Step idle delay, milliseconds) $2 = 0 (Step pulse invert, mask) $3 = 0 (Step direction invert, mask) $4 = 0 (Invert step enable pin, boolean) $5 = 0 (Invert limit pins, boolean) $6 = 0 (Invert probe pin, boolean) $10 = 1 (Status report options, mask) $11 = 0.010 (Junction deviation, millimeters) $12 = 0.001 (Arc tolerance, millimeters) $13 = 0 (Report in inches, boolean) $14 = 0
$20 = 0 (Soft limits enable, boolean) $21 = 0 (Hard limits enable, boolean) $22 = 1 (Homing cycle enable, boolean) $23 = 0 (Homing direction invert, mask) $24 = 50.000 (Homing locate feed rate, mm/min) $25 = 500.000 (Homing search seek rate, mm/min) $26 = 250 (Homing switch debounce delay, milliseconds) $27 = 1.000 (Homing switch pull-off distance, millimeters) $30 = 3000 (Maximum spindle speed, RPM) $31 = 0 (Minimum spindle speed, RPM) $32 = 0 (Laser-mode enable, boolean) $100 = 400.000 (X-axis travel resolution, step/mm) $101 = 400.000 (Y-axis travel resolution, step/mm) $102 = 400.000 (Z-axis travel resolution, step/mm) $103 = 10.000
$104 = 10.000
$110 = 1000.000 (X-axis maximum rate, mm/min) $111 = 1000.000 (Y-axis maximum rate, mm/min) $112 = 1000.000 (Z-axis maximum rate, mm/min) $113 = 10000.000
$114 = 10000.000
$120 = 30.000 (X-axis acceleration, mm/sec^2) $121 = 30.000 (Y-axis acceleration, mm/sec^2) $122 = 30.000 (Z-axis acceleration, mm/sec^2) $123 = 100.000
$124 = 100.000
$130 = 10.000 (X-axis maximum travel, millimeters) $131 = 50.000 (Y-axis maximum travel, millimeters) $132 = 500.000 (Z-axis maximum travel, millimeters) $133 = 360.000
$134 = 360.000
$140 = 0.010
$141 = 0.010
$142 = 0.010
$143 = 0.000
$144 = 0.000
ok $J=G21G91Z-500F1 ok $J=G21G91Z-500F1 ok $J=G21G91Z-500F1 ok $J=G21G91Z-500F1 ok $J=G21G91Z-500F1 ok $J=G21G91X500F1 ok $J=G21G91X500F1 $J=G21G91X500F1 $J=G21G91X500F1 $J=G21G91X-500F1 $J=G21G91X-500F1

Schildkroet commented 3 years ago

How do you compile it? With Embitz 1.11?

McCodie commented 3 years ago

I compiled it with Embitz 1.11 (Windows), also on MacOS using the Makefile, both gave me the same result.

I modified the Makefile a little bit. CFLAGS := -O2 -g1 -std=c11 -Wall -Wextra $(INCLUDE) -fno-common -fsingle-precision-constant -fdata-sections -ffunction-sections -fomit-frame-pointer -mlittle-endian -DUSE_STDPERIPH_DRIVER -DSTM32F446RE -DSTM32F446xx -D__FPU_USED -DARM_MATH_CM4 -Wimplicit-fallthrough=0

CFLAGS := -O2 -g1 -std=c11 -Wall -Wextra $(INCLUDE) -fno-common -fsingle-precision-constant -fdata-sections -ffunction-sections -fomit-frame-pointer -mlittle-endian -DUSE_STDPERIPH_DRIVER -DSTM32F411xE -DSTM32F411RE -D__FPU_USED -DARM_MATH_CM4 -Wimplicit-fallthrough=0

Schildkroet commented 3 years ago

Did you also change the linker flags? https://github.com/Schildkroet/GRBL-Advanced/blob/software/Makefile#L40

McCodie commented 3 years ago

I tested your hex file and it's not working. yes I also modified the LDFLAGS (as mentioned I also switched to STM32F446 in Windows and the result was exactly the same as with the macos binary).

LDFLAGS := -lm -flto -Wl,--gc-sections -T../stm32f446re_flash.ld -Wl,-Map=$(OUTPUT).map --specs=nosys.specs -nostartfiles --specs=nano.specs

Is this a stock NUCLEO-F446RE board? Did you install a crystal on X3?

Schildkroet commented 3 years ago

I tested your hex file and it's not working.

Not working at all? Garbage output? Same behaviour as yours?

McCodie commented 3 years ago

it's not returning anything via the serial interface did you set the serial to 115kbaud?

st-flash --format ihex write GRBL_Advanced.hex the other compiled image I'm loading with: st-flash write GRBL_Advanced.bin 0x8000000

st-flash 1.6.1 2020-11-28T00:13:38 INFO common.c: F446: 128 KiB SRAM, 512 KiB flash in at least 128 KiB pages. 2020-11-28T00:13:38 INFO common.c: Attempting to write 57748 (0xe194) bytes to stm32 address: 134217728 (0x8000000) EraseFlash - Sector:0x0 Size:0x4000 2020-11-28T00:13:38 INFO common.c: Flash page at addr: 0x08000000 erased EraseFlash - Sector:0x1 Size:0x4000 2020-11-28T00:13:39 INFO common.c: Flash page at addr: 0x08004000 erased EraseFlash - Sector:0x2 Size:0x4000 2020-11-28T00:13:39 INFO common.c: Flash page at addr: 0x08008000 erased EraseFlash - Sector:0x3 Size:0x4000 2020-11-28T00:13:39 INFO common.c: Flash page at addr: 0x0800c000 erased 2020-11-28T00:13:39 INFO common.c: Finished erasing 4 pages of 16384 (0x4000) bytes 2020-11-28T00:13:39 INFO common.c: Starting Flash write for F2/F4/L4 2020-11-28T00:13:39 INFO flash_loader.c: Successfully loaded flash loader in sram enabling 32-bit flash writes size: 32768 size: 24980 2020-11-28T00:13:40 INFO common.c: Starting verification of write complete 2020-11-28T00:13:41 INFO common.c: Flash written and verified! jolly good!

Connected to cu.usbmodem14103 @ 115000 baud

McCodie commented 3 years ago

I have 2 NUCLEO-F446RE boards here both show the same behaviour.

Schildkroet commented 3 years ago

Did like 100x jogs and then homed. Since nothing is connected, it fails after about 1min.

McCodie commented 3 years ago

can you show me your settings? $$

Schildkroet commented 3 years ago

`#define DEFAULT_X_STEPS_PER_MM 400

define DEFAULT_Y_STEPS_PER_MM 400

define DEFAULT_Z_STEPS_PER_MM 400

define DEFAULT_A_STEPS_PER_DEG 10.0

define DEFAULT_B_STEPS_PER_DEG 10.0

define DEFAULT_X_MAX_RATE 2400.0 // mm/min

define DEFAULT_Y_MAX_RATE 2400.0 // mm/min

define DEFAULT_Z_MAX_RATE 2400.0 // mm/min

define DEFAULT_A_MAX_RATE 12000.0 // °/min

define DEFAULT_B_MAX_RATE 12000.0 // °/min

define DEFAULT_X_ACCELERATION (100.06060) // 306060 mm/min^2 = 30 mm/sec^2

define DEFAULT_Y_ACCELERATION (100.06060) // 306060 mm/min^2 = 30 mm/sec^2

define DEFAULT_Z_ACCELERATION (100.06060) // 306060 mm/min^2 = 30 mm/sec^2

define DEFAULT_A_ACCELERATION (200.06060) // 1006060 mm/min^2 = 100 mm/sec^2

define DEFAULT_B_ACCELERATION (200.06060) // 1006060 mm/min^2 = 100 mm/sec^2

define DEFAULT_X_MAX_TRAVEL 300.0 // mm NOTE: Must be a positive value.

define DEFAULT_Y_MAX_TRAVEL 400.0 // mm NOTE: Must be a positive value.

define DEFAULT_Z_MAX_TRAVEL 900.0 // mm NOTE: Must be a positive value.

define DEFAULT_A_MAX_TRAVEL 360.0 // °

define DEFAULT_B_MAX_TRAVEL 360.0 // °

define DEFAULT_SPINDLE_RPM_MAX 1250.0 // rpm

define DEFAULT_SPINDLE_RPM_MIN 0.0 // rpm

define DEFAULT_X_BACKLASH 0.05 // mm

define DEFAULT_Y_BACKLASH 0.0 // mm

define DEFAULT_Z_BACKLASH 0.06 // mm

define DEFAULT_SYSTEM_INVERT_MASK 0x3F //

define DEFAULT_STEPPING_INVERT_MASK 0 //

define DEFAULT_DIRECTION_INVERT_MASK 0 //

define DEFAULT_STEPPER_IDLE_LOCK_TIME 255 // msec (0-254, 255 keeps steppers enabled)

define DEFAULT_STATUS_REPORT_MASK 1 // MPos enabled

define DEFAULT_JUNCTION_DEVIATION 0.02 // mm

define DEFAULT_ARC_TOLERANCE 0.001 // mm

define DEFAULT_REPORT_INCHES 0 // false

define DEFAULT_INVERT_ST_ENABLE 0 // false

define DEFAULT_INVERT_LIMIT_PINS 0 // false

define DEFAULT_SOFT_LIMIT_ENABLE 0 // false

define DEFAULT_HARD_LIMIT_ENABLE 1 // true

define DEFAULT_INVERT_PROBE_PIN 0 // true

define DEFAULT_LASER_MODE 0 // false

define DEFAULT_HOMING_ENABLE 1 // true

define DEFAULT_HOMING_DIR_MASK 0 // move positive dir

define DEFAULT_HOMING_FEED_RATE 35.0 // mm/min

define DEFAULT_HOMING_SEEK_RATE 600.0 // mm/min

define DEFAULT_HOMING_DEBOUNCE_DELAY 250 // msec (0-65k)

define DEFAULT_HOMING_PULLOFF 2.0 // mm

define DEFAULT_TOOL_CHANGE_MODE 1 // 0 = Ignore M6; 1 = Manual tool change; 2 = Manual tool change + TLS`

McCodie commented 3 years ago

Thanks, I think it is due to some very wrong settings. Seems like I have some progress, I just ran the setup wizard in Universal GCode Platform and set the default values.

I did not do any real configuration yet, I will wire it up with my machine and do all the real setup. I'll update this discussion once I'm done with that.

thanks a lot for your quick reply!

Schildkroet commented 3 years ago

Did you test my last file?

McCodie commented 3 years ago

yes, same result - so it's very likely due to the wrong settings... as mentioned I have updated it again with my build but ran the wizard for the settings. It's not stalling anymore now (but I only did some smaller tests) I will continue tomorrow with the real wiring.

Schildkroet commented 3 years ago

BTW don't write to $0. It contains the system flag, not step pulse time

McCodie commented 3 years ago

mm I just did some more tests with it and it's not reliable.

$$ $0 = 14 (Step pulse time, microseconds) $1 = 25 (Step idle delay, milliseconds) $2 = 0 (Step pulse invert, mask) $3 = 7 (Step direction invert, mask) $4 = 0 (Invert step enable pin, boolean) $5 = 0 (Invert limit pins, boolean) $6 = 0 (Invert probe pin, boolean) $10 = 1 (Status report options, mask) $11 = 0.010 (Junction deviation, millimeters) $12 = 0.002 (Arc tolerance, millimeters) $13 = 0 (Report in inches, boolean) $14 = 0
$20 = 0 (Soft limits enable, boolean) $21 = 0 (Hard limits enable, boolean) $22 = 0 (Homing cycle enable, boolean) $23 = 0 (Homing direction invert, mask) $24 = 25.000 (Homing locate feed rate, mm/min) $25 = 500.000 (Homing search seek rate, mm/min) $26 = 250 (Homing switch debounce delay, milliseconds) $27 = 1.000 (Homing switch pull-off distance, millimeters) $30 = 1000 (Maximum spindle speed, RPM) $31 = 0 (Minimum spindle speed, RPM) $32 = 0 (Laser-mode enable, boolean) $100 = 250.000 (X-axis travel resolution, step/mm) $101 = 250.000 (Y-axis travel resolution, step/mm) $102 = 3650.000 (Z-axis travel resolution, step/mm) $103 = 10.000
$104 = 10.000
$110 = 3000.000 (X-axis maximum rate, mm/min) $111 = 1500.000 (Y-axis maximum rate, mm/min) $112 = 500.000 (Z-axis maximum rate, mm/min) $113 = 10000.000
$114 = 10000.000
$120 = 50.000 (X-axis acceleration, mm/sec^2) $121 = 1.000 (Y-axis acceleration, mm/sec^2) $122 = 1.000 (Z-axis acceleration, mm/sec^2) $123 = 100.000
$124 = 100.000
$130 = 200.000 (X-axis maximum travel, millimeters) $131 = 200.000 (Y-axis maximum travel, millimeters) $132 = 200.000 (Z-axis maximum travel, millimeters) $133 = 360.000
$134 = 360.000
$140 = 0.010
$141 = 0.010
$142 = 0.010
$143 = 0.000
$144 = 0.000
ok $3=0 ok $J=G21G91Z-1F1 ok $J=G21G91Z-1F1 ok

Canceling file transfer.

$J=G21G91Z1F1 ok $J=G21G91Z1F1 ok

Canceling file transfer.

$J=G21G91Z1F100 ok $J=G21G91Z-1F100 $J=G21G91Z-1F100 $J=G21G91Z-1F100 $J=G21G91Z-1F100 $J=G21G91Z-1F100 Connection closed Connected to cu.usbmodem14203 @ 115000 baud Connection closed Connected to cu.usbmodem14203 @ 115000 baud Grbl 1.1f ['$' for help] $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$ Grbl 1.1f ['$' for help] $G $$

as you can see it just stalls again.

I was only moving the Z axis up and down, I have done some STM32 firmware myself and moving like 10 motors is stable on the STM32 controller with the same cabling (although I have only used one output of my application) so either some settings are messing around with it or this just doesn't work reliable at the moment. For sure it's not the fault of the controller.

If you have any further idea please let me know :-) I did not have a closer look at the grbl code, are the pulses software generated or dma/timer based?

to go forward I'll also set up a linuxcnc system in parallel so I can go on with other things in the meanwhile.

Schildkroet commented 3 years ago

What happens if you use a terminal (like hterm) and enter the commands manually?

McCodie commented 3 years ago

Hi,

I used coolterm on macos and grbl is crashing after a few commands, I cannot imagine that it will be stable for you if you use the same controller and eg. g-code sender because there aren't that many parameters that can be different. Did you really cut something with that grbl port?

I have been using linuxcnc for a few days now and there are no such issues.

McCodie commented 3 years ago

just use universal g-code sender and jog around a few times and it will crash. it makes no difference if you copy / paste the commands a few times.

Schildkroet commented 3 years ago

I send like 100 cmds with ugs and everything worked fine (Win and Linux). Did you tried with newest commit without changing any settings?

jielemon commented 2 years ago

just use universal g-code sender and jog around a few times and it will crash. it makes no difference if you copy / paste the commands a few times.

Hi, which universal g-code sender did you use? Could you please provide the download link?

sundtek commented 2 years ago

this is old ... the problem were EMC issues.