Mesa 7i39+5i20

nikos.chalikias wrote:

pin-out does not match to 7i39!

You appear to be correct, which is somewhat unfortunate.

I could edit this file and place the pins in the right places

It won't help, as you need to re-compile the firmware to change the pin allocations, and that requires software tools that you won't have.

I have sent a query to the developers list about the matter. It is also possible that Mesa have the correct firmware but it has not made it into the EMC2 repository yet.

In either case, it probably makes sense to wait for a software solution (which I expect to take no more than a day or two) than to make a breakout board.

nikos.chalikias wrote:

A Pins 1,3,5 and 25,27,29 ENCODER LINES ---> ENCODER

As expected

B pins 7,9,11 and 31,33,35 HALL LINES ----> NO-FUNCTION

Any pin with no special function becomes a GPIO (general purpose input / output) line. That means that you can read the values from the Hall sensor in HAL if you want to, or you can not bother if you are not using Hall sensors. The 7i39 has on-board Hall sensor inputs which are conditioned to suit the 5i20 inputs.

C pins 13,15 and 37,39 SENSE LINES ----> NO-FUNCTION, TPPWM, NO-FUNCTION, TPPWM

Pins 13 and 37 are SenseA for the two drivers, I am not sure what they do. 15 and 39 are fault status from the 7i39 SenseB output, and disable the PWM generator if the 7i39 detects a fault.

D pins 17 and 41 ENABLE LINES

---

> TPPMW, TPPMW

The Enable line from the 5i20 to the 7i39. The HAL file should connect this to Axis.N.amp-enable, you should be able to see this link in the HAL files I published.

E pins 19,21,23 and 43,45,47 PWM LINES

---

> ALL TPPWM

And these are the three PWM outputs which actually drive the motors.

A and E appear correct
B. no HALL lines?
C. 13,37 SENSE not used? 15,39 SENSE replaced by PWM?
D. ENABLE inputs replaced by PWM?

I think you are failing to read the "Pin Func" column of the PIN file. Everything looks right to me. If you specify 3 x 3ppwm and 3 x encoder in the config string then you will find that all of P2 and half of P3 are dedicated to 7i39 functions, and half of P3 and all of P4 are GPIO pins. (you can either ignore the GPIO pins on P3 or split the ribbon cable in half, with one half going to a 7i39 50-way plug and the other to other hardware.)

It all looks good to me.

Sorry for the delay, I have been on a yacht and internet-free for the last week.

Can you post the last dozen or so lines from dmesg after the failed launch? That should give some clues what the problem is.

It might be best to put both the INI and HAL files on www.pastebin.ca so we can try to reproduce the issue.