Advanced Search

Hi All,

I have finally reached the point in my Kitamura CNC retrofit that i can get the spindle turning.
I got a Delta 3.7KW C2000+ (3PH 380v) VFD + PG-01H encoder card. But I'm having some interesting issues....

So My Spindle motor is an old FANUC 2S 2.2 / 3.7KW 200V 4P 1500/8000rpm spindle motor:

Datasheet link:  s3.amazonaws.com/Icarus/DOCUMENTS/Fanuc_Manuals_1791.pdf So what I tried is to first figure out if this is an AC induction motor or an PM motor. From what I can find (or not find in this case) anything in the datasheet so that made me think it is an AC motor as i feel no "dents" when i turn the motor by hand.
So I connected it all including the OEM Fanuc encoder signals the the VFD.
And then I went through the setup wizard, (immediately went for the complicated method) of including the encoder feedback etc.
When i tried to do a test run it just immediately jumps in OL3 error. I tried some of the advisement in the datasheet such as changing accel time or carrier frequency but that did not help.

So i went back to the basics and just put it in V/F mode. Max output voltage 200V , poles: 4 , max freq 267Hz , base freq 50hz (1500rpm) .
Now when i pressed run it turned, but it did not sound good. The current was around 17A's when it runs with no load on like 500rpm. I tried to set it to higher RPM but the high load pretty much states around. After a bit the VFD throws and overload as it is pulling too much current.

And now... I'm quite lost, any idea what I'm doing wrong? I have worked quite a bit with vfd's and never had such troubles so I'm thinking there is something fundamentally wrong. (this servo spindle worked well with the FANUC drive it had before).

No, 7i95t_5abobd.bin should work.

Well that worked. No more dither error.
Thanks.

I don’t know where mesact expects the files either. I searched for the files in the file system with that name and replaced them. I know nothing about Linux so I’m sure made a mistake.
But if I ever need to flash again, I’ll know to use the command line in the directory with the file.

Sorry i could not help, i am sure it is something wrong with the "desktop explorer", that is the normal file explorer that does the desktop part, had this happen on Mate DE twice, it just requires deleting a hidden folder, but i do not know what folder is that on XFCE.

Yeah, that's the old file. not sure where mesact expects the files but you can program the new file with:

mesaflash --device 7i97t --addr 10.10.10.10 --write 7i97t_5abobd.bin --reload

run in the directory where the new file has been placed

I'm not sure i'm flashing the current file.

Do I need special firmware? Can you please make one for me?

Are you sure you are flashing the current file?

The change is the PWM version It needs to be > 0 for Dither to be supported

What does

mesaflash --device 7i97t --addr 10.10.10.10 --readhmid

report?

Yep. Tried reload in mesact, tried restarting the computer, power cycled the 7i97t, in every combination, together and separate.
but I also tried flashing 7i97t_d.bin and linuxcnc starts without the error every time.
Then I’ll try 7i97t_5abobd.bin and it I get the same error.

Thank you, I got it working now. Btw. yesterday I saw a list of all this modules that's supported, but I cannot find it again. It was not on GitHub, do you know where I can find the list?

Did you do a reload or power cycle after writing the firmware?
 

7i95t_5abobd.bin has 1 PWMgen and I think 4 or  or 5 additional stepgen channels

Its designed for the common "Sainsmart Mach 5 Axis breakout"
 

I tried to use the files you sent, but I’m still having the dither not found error.
I replaced both the .bin and .vhd files in the file directory.
I then used mesact to flash.
I assume I’m skipping a step.

Andy, thanks. That will definitely get used.
 

I am having an odd issue with the qtdragon_hd UI MDI. Doing a rapid on one axis causes one of the other axes to move. But it's only the axis that I've commanded previously. In the video example, I never sent commands to the Z axis so it didn't move here. I'm not sure what could be causing it. It is also changing the DRO, so it is sending the commands to both axes for some reason.

I tested in AXIS and it wasn't an issue, so it must be something to do with qtdragon. Is it possible the default gcodes are causing it in some way? The only one I can think of that  is active is G99 but I don't know if that would be capable of causing this issue.

Video of the error: Video

 
 

I decided to go with a buffered parallel port BOB to get more GPIO (Mesa 7i75 Breakout/FPGA protection card). But I also need a pwmgen to drive a Mesa SPINx1A for the VFD. I plans to use all 6 step/direction and encoder channels eventually, so I don't want to use one of those for VFD speed control if I don't have to. Leaving the sserial as is for future expansion.

So I need firmware which adds 1 pwngen to the P1.

Would 7i95t_5abobd.bin in the MesaCT do this? I assume it would make 5 pwmgen's available; and for those pins I want to use as I/O, I just declare them in HAL as GPIO. Is this correct?

Or do I need special firmware?