Remora - ethernet NVEM / EC300 / EC500 cnc board

I had problems with the new version of the board at the beginning. Flashing the board was not possible. I have tested some ST-Link and also Daplink. Actually the board was defective. I complained about it and got a new one and bought another one. Flashing worked with both without any problems.

At the moment when you couldn't flash it, did you try working with Mach3? If the controller itself is not working, it would hardly give signals to the output.

At the beginning I tested it briefly, but not in detail. seemed to work. In the end nothing worked.

I do not know if it is worth doing this fee at all. There is only one two-year-old firmware for LinuxCNC on it. Will there be any positive result for switching from Mach3 to Linux CNC with this board?

Masterdrive wrote:

scotta wrote:

Hi, there is no cpp source code for the NVMPG. I decompiled the extracted firmware and managed to find the serial transmission section. The communication protocol was derived from that. The attached spreadsheet simulates the protocol. You should be able to use that for your purpose or alternatively look at the code of the NVMPG LinuxCNC component.

github.com/scottalford75/Remora-NVEM/blo.../MPG%20protocol.xlsx
Hi Scott,
thanks for your time and dedication.
I've owned a NVEM 2 since 2018, but buried it long time ago since I was not able to make it work reliably.
www.cnczone.com/forums/controller-cards/...084.html#post2154084
At that point in time, and till now, I've been using Mach3 (yeah... no LinuxCNC... i know...)

2019, after migrating to an UC300ETH controller, I tried to make the MPG work with Mach3 without using the NVEM.
I didn't want to waste such a nice MPG.
At that time, I did some reverse engineering... not as professional as you did... mine was mostly trial and error and listening to the serial communication between the MPG and the NVEM after opening it and finding a MAX232. I captured hundreds of packages, and tried to find the similarities on what was displayed on the MPG and what was within the payload. I almost got the payload down to your level.
I was too dumb to know about headers and such -_-

I also tried to create an Arduino program for having the MPG communicate with a PC using MODBUS.
But again, I was too dumb to find a smart solution, like having the Button-presses being reflected as Output-Pin changes on the Arduino, and hook those up to my motion-controller. Or using the Arduino Pro-Micro as a HID. Instead I tried to make the communication work with MODBUS and Mach3 "Brain", which resulted in a complex, and unreliable solution.
So I buried that solution almost 4 years ago as well.

That being said... I'm surprised about the NVEM still being a topic after stumbling about this thread by accident.
Would you mind if I would revive my idea of a NVMPG to PC connection, utilizing some of the "wisdom" you've gathered throughout the years?
Don't get me wrong... I'm an "enthusiast", far away from being an "achiever", my programming days are more than 20 years in the past now... but it's been bugging me a lot looking at this "unfinished project" (a lot of them exist though -_-)

Kind regards,
David

Hi David, 
No problem from my side. This is exactly why I love open source projects. One builds from another.

Hi Scott,First, I want to thank you for your incredible work on the Remora firmware!Two weeks ago, I received my NVEM V5 board and started working on flashing it. Unfortunately, I couldn't find my STLink dongle (seems like I lost it), and I couldn't wait for a replacement. So, I decided to get creative and used what I had available. I flashed an STM32F103C8T6 using a Raspberry Pi Zero W with the "CMSIS-DAP" firmware you mentioned in one of your YouTube videos, using OpenOCD.As you described in the video, I then uploaded the "stm32f103xb_mimxrt1060_qspi_if.bin" to the mass storage device of the STM32F103C8T6, and it seemed to program itself as expected. I then connected it to my NVEM V5 board and dropped the firmware file onto it, but unfortunately, it didn't work.I spent the next three days troubleshooting but eventually gave up and purchased an STM32-Link, along with two CMSIS-DAP Probes (USB-A and USB-C). I received both of these items recently and tried again, but I’m still having issues.The CMSIS-DAP probes aren’t detected at all, neither by MCUxpresso IDE nor PyOCD. The STM32-Link is recognized by PyOCD, but I still can't get it to work as intended.Do you have any suggestions on where I should focus my efforts next? Any guidance would be greatly appreciated.Thanks again for all your help!

Similarly I've had no success with any purchased CMSIS-DAP probes. I'm still using the STLink converted with the firmware in the repo. I use this for all my debugging work in MCUXpresso.

I've also had good success with PyOCD and a Pico Probe.

What errors does PyOCD give with the converted ST-Link?

Yay, I managed to get the CMSIS-DAP probes to work with PyOCD! However, when I tried to flash the firmware, I encountered an error: E Error during board uninit: [session]

I've attached some pictures with more details of the issue. Any guidance on how to resolve this would be greatly appreciated!

Thanks again for your support!

Edit: For anyone facing the same issue where the CMSIS-DAP (WCH-Link) doesn't work as a DAP Probe, I was able to fix it using the "WCH-LinkUtility."  

Hey Scott,
Sorry to bug you again, but just wanted to let you know I finally managed to flash the firmware and ping 10.10.10.10! Next up is diving into the LinuxCNC side of things.

I’m planning to document everything I did to get it all working and might make a YouTube video showing the steps once my exams are out of the way next week. Also, I was thinking it’d be cool to contribute to the project documentation if there’s any way I can help out.


Thanks again for all your help – really appreciate it!

Great to hear and any help is greatly appreciated. Cakeslob as also been helping out the documentation.

It would be great to document how you got the probe up and running. This seems to be the hardest part for everyone with these boards, just getting the board to flash successfully can be a challenge.