Request for Mesa 5i23_7i48_7i52 firmware

If your spindle control can set 0 speed as 1/2 way up PWM
than that should work but note that you will get full speed
when LinuxCNC is not running (since PWM would be off)
so you need a separate enable (perhaps from the 7I84).

This mode is supported in the hostmot2 driver for PWM as "offset mode"


 

On changing my ini file to try out this new firmware I get this error:
IDROM IOPorts is 3 but MD IOPort NumInstances is 6, inconsistent firmware, aborting driver load

I've created and attached the report file.

This one should be better:

 
 

That's good so far - existing system is perfectly happy and working as before with the new firmware in; it'll take a bit of time to integrate and check the new functions. Hasn't helped that the new servo drive/motor kit is now doing nothing but fault out, after a working brilliantly for about half a day (nothing to do with the firmware change, it's doing it regardless of how I send it commands). I think I can still try out the alternate PWMgen on my existing vfd.
Thank you once again for your help
Andy

Hello Peter,
In trying to get another couple of axes integrated with my mill, I seem to have killed my 5i23 - currently not found by mesaflash --device=5i23 --readhmid.

I stupidly unplugged the cable from a 7i52 to a stepper driver (which was unpowered) without powering off the system. I'm not sure if there's anything I can do about that short of buying a permanent replacement card for the mill, but in the meantime I need it back up and running and have 2x 7i80HD cards, one -16 and one -25 I can use.

I'd be grateful if you could build a firmware for me for one of the 7i80s that can support my config:

loadrt hm2_pci config="firmware=hm2/5i23/5i23_fmd.bit num_encoders=14 num_pwmgens=7 num_stepgens=2 sserial_port_0=1"

Currently this is for a 7i48 driving 6x +/-10V motors (4 servo, 2 spindle) and 6 encoders; a 7i52 with 1x sserial port (to 7i84), 2x stepper motors and 2x encoders, with space for one more pwm and 4 more encoders (for a future atc if anything).

thank you - also any advice on resurrecting the 5i23 gratefully received.

I can do this sometime in the next couple of days

Since the 5I23 is not found, it not a simple I/O issue but something
on the bus or power supply related. You might check (carefully) for hot parts

7I80HD-16 and -25 versions of 7i48_7i52 firmware:

 
 

Earlier today I removed the 5i23 and disconnected the other cards, rebooted, reseated, and then repeated this procedure another couple of times. mesaflash is now seeing the 5i23 again and my system is back up and running. I'm grateful for the 7i80 firmware though in case it happens again. Thank you as always for the support.

Back on the 5i23 with this firmware, trying to get two stepper motors working. Using a config I adapted from sim vismach xyzbc trt, combined with the hal files from my mill which works well as a 3 axis (XYZW, W on the knee).

I have step output but no direction control for B axis, neither step nor direction for C axis. Wiring from the 7i52 is:
SSerial on COM 0
Step0 on COM 1
Dir0 on COM 2
Step1 on COM 3
Dir1 on COM 4
and an as yet unused PWM on COM 5.
One single-ended encoder for each stepper motor wired in to ENC 0 & ENC 1.

Does this look like the right wiring for getting step/dir out of the 7i52?
Brown = step0 gnd
Red = step0 +
Yellow = dir0 gnd
Orange = dir0 +
Green = step1 gnd
Blue = step1 +
Grey = dir1 gnd
Purple = dir1 +

Oscilloscope shows a nice pulse stream on the Step0 connection but nothing on the rest - there's 0v and +5v where expected (i.e. pins 0/1 and 6/7 of each block of but nothing from what I think are the relevant TX pins.
The red alarm LEDs on the driver don't seem to mean anything; if I change things round so that the step0 pulsestream is sent to the C motor, that turns.

 

You need to set GPIO bit 24 low to get the TX3 output enabled

sudo mesaflash --device 5i23 --dbname2 7i52 --readhmid

should help with the pinout