Remora - ethernet NVEM cnc board

Hmm, having to restart LinuxCNC twice..? What's the behaviour on the first restart?

This beyond-cool project should be ported to the Nucleo-F746ZG. Thus eliminating the hardware constraints that are set by the manufacturers of those boards you are using. Although nothing wrong with them for those looking for a turnkey, I could see custom boards & hats being far more useful to many. Just a suggestion...

Hi Scott

As per post
forum.linuxcnc.org/18-computer/44828-rem...rd?start=1480#288965

I'm getting joint following errors (on 2 different setups)
Power cycle EC500 & then start LinuxCNC. The first time I try to jog I get joint following errors. Exit LinuxCNC & restart (not the EC500) & the fault dissapears.

--- QtVCP Screen Started on: Wed, Dec 27 2023 17:41:22
--- QtDragon Version 1.1 on Linuxcnc 2.10.0-pre0-2499-g1b0a72c1f2
---Wed27 17:41: Machine OFFWed27
17:41: Hard limits trippedWed27
17:41: Tool 0: No ToolWed27
17:41: Unexpected realtime delay on task 0 with period 1000000This Message will only display once per session.Run the Latency Test and resolve before continuing.Wed27
17:41: Machine ONWed27
17:41: joint 0 following errorWed27
17:41: Machine OFFWed27
17:41: command (EMC_JOG_STOP) cannot be executed until the machine is out of E-stop and turned on 

Alex

frayja, needing to restart linuxcnc to recover from estop several times is very abnormal behavior. I would be willing to bet this behavior is most likely due to your hal estop latch logic. I suggest verify the integrity of your hal estop latch logic before moving forward.

Can you recover from estop at all? physical or software

Use halshow to watch all those pins to see exactly what is happening when you cannot reset estop

linuxcnc.org/docs/stable/html/hal/tools.html#_halshow

Mecanix wrote:

This beyond-cool project should be ported to the Nucleo-F746ZG. Thus eliminating the hardware constraints that are set by the manufacturers of those boards you are using. Although nothing wrong with them for those looking for a turnkey, I could see custom boards & hats being far more useful to many. Just a suggestion...

This version of remora is very hardware specific. We have nucleo/dev board in other versions of the firmware, but nothing for the PHY ethernet version. The NVEM and EC500 that use the f207/407 could probably be ported to the nucleo equivalent versions

"estop several times is very abnormal behavior"

Yes I have posted a message elsewhere on the forum about this.
As I said this seems to only be happening on V2.10.
The need to enter the software twich on power up of the EC500 seems to be universal since I updated to the new firmware. IE On linuxCNC 2.9 the Estop works correctly but I still need to enter the software twice to get rid of the joint following errors.

I have erased & reflashed one of the boards I have (& re uploaded to config) & the behaviour seems to be the same.

Is no one else having this issue?

As always any help appreciated.
Thanks
Alex

fintech wrote:

Downloaded the ZIP, I have that processor SDK installed. The issue seems ethernet related. I'm trying a few ways to get the firmware on the MCU. I'm trying The daplink process get The RT1052 Drive in windows. When I copy the remora firmware I get a text  file called FAIL.TXT  after its reboots

error: Failed to initialize the target MCU
type: target
 

Hi Scotta & fintech,
I have the same error from my EC300 firmware drag'n'drop using the DAP-link, 'Failed to initialize the target MCU'; is the remedy to use 'pyocd'?
(I couldn't get the CMSIS-DAP-STLINK21.hex to work, I made the USB DAP-link from Github dev directory .bin files)

Taken from 03 Feb 2023 00:25 #263562 above:
python -m pip install -U pyocd
pyocd flash .\remora-rt1052.bin --target mimxrt1050_quadspi

If pyocd is the recommended route, the .bin file would now be 'remora-rt1052-EC300-0.2.0.bin'.
Would the target still be 'mimxrt1050_quadspi' for the EC300?

Apologies if this issue is particular to the defunct DAP-link route; please also point me to the latest ST-link to CMSIS-DAP creation method.

Hi,
I am trying to upload remora-rt1052-EC300-0.2.0.bin to the RT1052 EC300.
I have moved on to trying the pyocd method.  Using either CMSIS-DAP or ST-link V2 clone and pyocd list, I get Target 'N/A' and I assume that this is no correct.

(base) C:\Users\...\Remora-RT1052-main\Remora-RT1052-main\firmware>pyocd list
  #   Probe/Board              Unique ID      Target

---

  0   X893 ARM CMSIS-DAP HID   0001A0000000   n/a

(base) C:\Users\...\Remora-RT1052-main\firmware>pyocd list
  #   Probe/Board    Unique ID                  Target

---

  0   STM32 STLink   B65B5A1A0000000047F90A00   n/a

(base) C:\Users\...\Remora-RT1052-main\Remora-RT1052-main\firmware>pyocd flash .\remora-rt1052-EC300-0.2.0.bin --target mimxrt1050_quadspi
0004572 I Loading C:\Users\...\Remora-RT1052-main\firmware\remora-rt1052-EC300-0.2.0.bin [load_cmd]
100%
0010256 I Erased 0 bytes (0 sectors), programmed 0 bytes (0 pages), skipped 117760 bytes (460 pages) at 20.25 kB/s [loader]

The pyocd flash terminal output looks just like those shown in earlier replies, but for the 0 bytes programmed.
I have checked the connections as Target 'N/A' could indicate incorrect hook-up.
Is the other experience of getting N/A?

I have a USB to Serial module of as close a match to that shown by Scott in the YT video.  I have tried to use this to verify whether Remora is present and healthy on the EC300 but I am not confident in the settings in Putty or Realterm, I have tried both.
In Win10 Device Manager, connecting the USB-to-serial module creates 'USB Serial Port (COM4)' so I use that in Realterm (or Putty); what is the correct Baud rate, I'm defaulting to 9600?
Display -> 'ascii'
Port -> Baud '9600', Port '4 = \VCP0'

What is the robust method of establishing the serial/UART feedback?
I assume from the YT video that I should see the 'Entering Start State' etc without connecting ethernet to RPi.

I has read pretty much all 150 pages here, skimming some sub-topics, and I don't see guidance on the serial/UART set-up (I may be wrong though).

I am looking for help please to diagnose and understand each stage towards the fantastic prospect of PRi + step/dir timing from within the EC300 motion controller.  I appreciate any help available.

Wishing everyone a Happy New Year with lots of CNC'ing
Tim

Hi Tim,

For the serial output the baud rate is 115200.

Firmware is up to version 3.1.2 now.

github.com/scottalford75/Remora-RT1052-c...ora-rt1052-3.1.2.bin

If you power on the board without an Ethernet connection the boot will wait for the network and will not move on. You will get a serial output though.