Higher end Chinese controllers?

[John S]

Best one I ever saw was a reference to a water buffalo ???

Couldn't manage to work that one out though so got the spares list and cross referenced the picture.

It was a hydraulic ram

[JoeHarris]

Best one I ever saw was a reference to a water buffalo ???

Couldn't manage to work that one out though so got the spares list and cross referenced the picture.

It was a hydraulic ram

Haha love it!
I feel a thread dedicated to chinglish coming on.

[Boyan Silyavski]

quote from other thread:

However there are many potential pot holes it could fall down on like Canned cycles etc, long cycle times etc.
Lack of memory is also another issue. Even the higher spec controller is memory limited. 32Mb or even 64mb isn't enough for Large 3D programs and will require DNC to drip feed the code. For some 32Mb may seem plenty but I can tell you some the people I help have 3d files in the Gb's. Pauses or temp locks while dripping code, even if position is not lost, is problem on 3D surface because it leaves mark which wreck the job or at best cause extra work hand finishing.
Dripping from USB isn't always reliable and RS232 just defeats the point of stand alone Controller. All these things have to be tested and influence the final outcome.!! . . . . Unfortunately again we are back to Cracking open the Good or Bad Egg.?

After talking with a couple of manufacturers unfortunately that seems to be the case. The typical answer i received was:

Our CNC controllers are using standard G code, the system can be directly copied from the U disk file to the system processing, but can not directly use the U disk online processing.The file size depends on the disk space of our system.The largest disk space of our system is 128M.

Its incredible but the mini controller DDCSV1 beatas them on that front. I will continue its review in the relative thread. Just would like to say here that it reads and executes 10 million line file without a glitch. From start to end. There were some glitches at first and at the end i found that was because i was not using the original pen drive. Once i started doing that, 2 weeks of perfection till today, working 8h per day at least non stop.

If you wonder how i managed to make 10 million line file. Very simple. Milling 1300x1000, trochoidal using 6mm bit. it was not big on size though. So at the end that memory limit could not be such big problem as it seems.

[magicniner]

There were some glitches at first and at the end i found that was because i was not using the original pen drive.

And when "The Original" dies what do we use?

- Nick

[Boyan Silyavski]

And when "The Original" dies what do we use?

- Nick

I think brand ones made for capturing HD video on phones or cameras will work right / card + usb adapter/ . Will buy one original kingston and will try. Problem comes from that i have 10 pendrives which are branded and i am not sure if original or no. But when i fit and SD card inside my Samsung Edge 7 and start recording video at HD or 4k it will become pretty obvious if original or not

[JAZZCNC]

So at the end that memory limit could not be such big problem as it seems.

But how large was the file size.? 10 Million very small lines won't take much memory. Try complex 3D surface at that physical size and see how it performs. If it loads that file then I'll be impressed..

[Boyan Silyavski]

But how large was the file size.? 10 Million very small lines won't take much memory. Try complex 3D surface at that physical size and see how it performs. If it loads that file then I'll be impressed..

That good enough for you? I don't think a bigger file can make any difference evaluating the controller. I paused it in middle and changed tools, re zeroed and started again/ broken tool/. I stopped it later, remembered the line and when i had time started from line 2 500 000. Controller was busy 15 minutes until it went to that line and started perfectly from there. I don't have doubt the DDCS V1 is a working controller. I know how to square with it the gantry, how to tool change and all that stuff. It homes perfectly and mind that, can be made to repeat 5 times at the sensor and you control the backoff.
I will write all that in detail when i have time, but i am impressed.

[Clive S]

Boyan what is the significance of the 233mB file shown in the pic.

Thelargest disk space of our system is 128M.

What is the actual gcode file size?

[Boyan Silyavski]

Boyan what is the significance of the 233mB file shown in the pic. What is the actual gcode file size?

The significance is that for the cheap controller which processes from USB directly, without downloading the file to the memory, there is no problem with the file size which as seen is bigger than its memory at least twice. Meaning also the DDCSV is quite ahead of its bigger brothers, at least on some implementations.

That's the actual size of the G code in MB. Its trochoidal with small cutter, so millions of lines.

[JAZZCNC]

That good enough for you? I don't think a bigger file can make any difference evaluating the controller.

Yes that's good enough for me and agreed impressive if it can buffer without pausing in the cut. Have you tried cutting at high feeds/accel while the buffering from USB.? IE 3D cutting. This will be the Acid test.

I know how to square with it the gantry

Explain how it does this please.?
Does it home to separate switches for each side/axis or are you using the Index pulse from your Servo drives and just sending signal to one input.? This will make difference to Stepper users who don't have Index pulse and use slaved motors.