DDCSV1.1 4 Axis controller

The Shop > CNC

<< < (37/202) > >>

Benedikt:

--- Quote from: Simon0362 on November 18, 2016, 03:29:55 AM ---
--- Quote from: Benedikt M. on November 17, 2016, 03:11:15 PM ---

I have no idea how to get the bootloader to do anything more than to try to boot from USB, there is absolutely no documentation on it and I am not able to get it's source code nor can I actually dump the bootloader. I would not be surprised if it actually resides in a mask rom in the CPU.

--- End quote ---

I don't doubt your analysis but nonetheless I am somewhat surprised that they would go for a masked ROM - would it have been done on the grounds of cost/volume production? or as a security protection?

I was involved in this sort of production many years ago and there was a high risk seen for having a large stock of junk in
the event that the ROM program needed modification - but

that was a good few years ago and the economics have probably changed a lot since then...

--- End quote ---
Great to have somebody from the industy in our team.
My assumption is that the bootloader is installed by the chip manufacturer by masking it. However i was not yet able to prove this. Atmel does the same in some of there arm chips which is having a known to be working bootloader in rom.
I assume this is the same here for three reasons:

- Other devices with this CPU have the same bootlog from the loader
- the manual which i have not yet read due to lack of time states that the cpu can boot from usb which might indicate a factory installed bootloader.
- the device shows a bootloader even without any flash connected and the chip is not advertised to contain any flash

The benefits are robustness of the first stage bootloader, easier loading of a kernel or second stage bootloader like uboot.

Also they probably got it right however there is of course a risk that the bootloader has some issues.

Simon0362:

--- Quote from: Benedikt M. on November 18, 2016, 07:02:42 AM ---
My assumption is that the bootloader is installed by the chip manufacturer by masking it. However i was not yet able to prove this. Atmel does the same in some of there arm chips which is having a known to be working bootloader in rom.
I assume this is the same here for three reasons:

- Other devices with this CPU have the same bootlog from the loader
- the manual which i have not yet read due to lack of time states that the cpu can boot from usb which might indicate a factory installed bootloader.
- the device shows a bootloader even without any flash connected and the chip is not advertised to contain any flash

--- End quote ---

Hmmmm.....not going to disagree with such a well presented set of arguments :smart:

And to back up your case (and further destroy mine!), I found this on the Nuvoton site:
"Mask ROM Technology

By virtue of 0.5um and 0.35um logic process architectures, Nuvoton has developed 0.5um 5V Logic embedded 0.37um Flat Cell and 0.35um 3.3V/5V Logic embedded 0.32um Flat Cell processes to support ROM needs. Nuvoton offers the most competitive Flat Cell ROM Device size in the industry. This allows us to provide an efficient and cost-effective solution for Code Mask requirements and flexible production management for ROM based products." http://www.nuvoton.com/hq/foundry-service/technologies/mask-rom-process/?__locale=en

Simon

Dirty Engineer:

--- Quote from: Cncdane on November 17, 2016, 07:00:28 PM ---Just bought this, and this has become a bit of a nightmare for me. I really like the feel of the controller, but whenever I try to run a programme, will it stop after a few lines, and the controller stays busy and speed goes to zero. Most often, I have to power cycle the controller to make it respond to the buttons. I can reset, but most often can I not even jog the axises. Restarting or resuming the program just make the controller go busy.

I have not wired up the controller to my CNC (got no endstops, and fears a crash, given the problems I have so far). Am I missing something?

Tried code generated with several post processors from Fusion360, as well as F-Engrave, and the outcome is the same.

--- End quote ---

FYI you have to use Fanuc/Gcode code in MM not inch. Mach3 style code will not work. More info starting around 3:30 minute mark and going forward in this video.
[ Invalid YouTube link ]

--- Quote from: machineshopohio on November 17, 2016, 10:19:36 PM ---I also just bought this, and it takes time to learn it. but i'm still moving forward.

--- End quote ---

This is a great video of setting up this controller.

Merlin201314:
The mm/inches are decided by the G20(imperial) or G21(metric) command at the beginning of every G-code file, and the controller is able to accept both codes.
Even the pure Fanuc G-codes are not working, you need a generic G-code generator, or it doesen't work.
This controller has a very limited set of instructions available in memory, so do not aspect big jobs, but it works and if you manage to compile your own g-code file by merging several routines it will do the job as a professional one.
This is my poor opinion :thumbup: .
Cheers

Benedikt:
If we were going totally crazy, we could try to improve the program. But that would most likely end in a full rewrite. However it is absolutely possible.
What we would need to do is get a development toolchain going and then we would need to reverse engineer the motion driver interface.
Of course, we would then need to get screen output and button input working and develop proper routines. It would be a ton of work and it might not be worth it.
I doubt we can get the original source, but maybe somebody knows the people who made it. We never know.

Navigation

 Message Index

Go to full version