Difference between revisions of "Recans A0"

From iagent
Jump to: navigation, search
(Debug)
 
(11 intermediate revisions by the same user not shown)
Line 1: Line 1:
 
Recans A0 has two controllers, one is based on RP2040, the other on MCP2515.  
 
Recans A0 has two controllers, one is based on RP2040, the other on MCP2515.  
 
To bring up the RP2040 CAN controller:
 
<pre>
 
sudo ip link set can0 type can bitrate 1000000
 
sudo ip link set up can0
 
sudo ifconfig can0 txqueuelen 128
 
</pre>
 
 
  
 
The MCP2515 CAN controller<br>
 
The MCP2515 CAN controller<br>
Line 19: Line 11:
 
After booting:  
 
After booting:  
 
<pre>
 
<pre>
 +
sudo gpioset 1 165=0
 +
sudo modprobe -r mcp251x
 +
sudo modprobe mcp251x
 
dmesg | grep mcp
 
dmesg | grep mcp
 
</pre>
 
</pre>
  
 
<pre>
 
<pre>
sudo ip link set can1 type can bitrate 1000000
+
sudo ip link set can0 type can bitrate 125000
sudo ip link set up can1
+
sudo ip link set can0 txqueuelen 128
sudo ifconfig can1 txqueuelen 128
+
sudo ip link set up can0
 
</pre>
 
</pre>
  
Line 46: Line 41:
 
</pre>
 
</pre>
  
===Errata===
+
<pre>
 +
candump -tz -Ddex can0,#FFFFFFFF
 +
</pre>
 +
 
 +
To compile the RP2040 with klipper:
 +
<pre>
 +
cd ~/klipper
 +
cp test/configs/rp2040.config .config
 +
make menuconfig
 +
-- Communication interface = RP2040
 +
-- RX = GPIO4
 +
-- TX = GPIO5
 +
make
 +
sudo apt install libusb-1.0-0-dev
 +
sudo FLASH_DEVICE="2e8a:0003" make flash
 +
 
 +
</pre>
 +
 
 +
 
 +
 
 +
===Hardware Errata===
 
* SPI_CLK and SPI_MOSI are interchanged and need to be fixed.  
 
* SPI_CLK and SPI_MOSI are interchanged and need to be fixed.  
 
* /RESET is floating.
 
* /RESET is floating.
  
 +
===Error messages===
 +
When sending:
 +
- write: No buffer space available
 +
 +
Got error -1 in can write: (105)No buffer space available
 +
 +
After that I get:
 +
                                         
 +
DROPCOUNT: dropped 109 CAN frames on 'can0' socket (total drops 576)
 +
 +
(302.908159)  can0  RX - -  20000004  [8]  00 01 00 00 00 00 00 00  ERRORFRAME                 
 +
        controller-problem{rx-overflow} 
 +
 +
And the INT pin remains low. Needs a restart.
  
 
===Debug===
 
===Debug===

Latest revision as of 00:28, 25 February 2024

Recans A0 has two controllers, one is based on RP2040, the other on MCP2515.

The MCP2515 CAN controller

sudo mount -o remount,rw /boot
sudo armbian-add-overlay /boot/overlay-user/sun50i-a64-mcp251x.dts
sudo mount -o remount,ro /boot

After booting:

sudo gpioset 1 165=0
sudo modprobe -r mcp251x
sudo modprobe mcp251x
dmesg | grep mcp
sudo ip link set can0 type can bitrate 125000
sudo ip link set can0 txqueuelen 128
sudo ip link set up can0

Interrupt: STEP7 = PL11

sudo apt install net-tools can-utils
cansend can1 5A1#11.2233.44556677.88
ip -details link show can0
~/klippy-env/bin/python ~/klipper/scripts/canbus_query.py can0
candump -tz -Ddex can0,#FFFFFFFF

To compile the RP2040 with klipper:

cd ~/klipper
cp test/configs/rp2040.config .config
make menuconfig
-- Communication interface = RP2040
-- RX = GPIO4
-- TX = GPIO5
make
sudo apt install libusb-1.0-0-dev
sudo FLASH_DEVICE="2e8a:0003" make flash


Hardware Errata

  • SPI_CLK and SPI_MOSI are interchanged and need to be fixed.
  • /RESET is floating.

Error messages

When sending: - write: No buffer space available

Got error -1 in can write: (105)No buffer space available

After that I get:

DROPCOUNT: dropped 109 CAN frames on 'can0' socket (total drops 576)

(302.908159) can0 RX - - 20000004 [8] 00 01 00 00 00 00 00 00 ERRORFRAME

       controller-problem{rx-overflow}   

And the INT pin remains low. Needs a restart.

Debug

During init/reset:

TX: 0xC0 = RESET

TX: 0x03 0x0E 0x00 = READ CANSTAT
RX: 0x00 0x00 0x80 = Device is in Configuration mode

TX: 0x03 0x0F 0x00 = READ CANCTRL
RX: ? 

TX: 0x02 0x0F 0x20 = WRITE CANCTL = Sleep Mode


When trying to set interface up:

TX: 0x02 0x0F 0x80 = WRITE CANCTRL REQOP2 = Configuration Mode
RX: 0x00 0x00 0x00 
TX: 0x03 0x0E 0x00 = READ CANSTAT
RX: 0x00 0x00 0x64 = OPMOD[2:0] = 0b011, ICOD[2:0] = 0b010

Device is in Listen-Only mode, Wake-up interrupt


It seems that the bring up sequence interferes with the initialization. Must make sure the mcp2515 probing is done after power is applied.