Startup trouble

The original grbl controller had a step time of 10000 which is double the default of lcnc. So I went with that already. But isn't the mesa card controlling this? So it should work even if I had super crappy jitter?

I'm sure about the DB25 pinout. Double-checked with the manual.
But I'm not sure if I have to ground the limit switches L1-L5 or pull them to 5V or if I can leave them open. Do you know that?

I checked the direction signals. I get 5V for one direction and 0V for the other.

Thanks.

Are the motors locked? could be some kind of enable issue
 

What do you mean by 'locked'?
They are not mechanically locked. I can turn them freely by hand. I also don't hear a 'stressed' sound when I try to turn them. Just nothing.

Any idea about the limit switch inputs?

If they are not locked, that means the drive is not enabled
which suggests a wiring error

Not sure if the limit switches are relevant at this point

 

(this forum drives me nuts - third time it has eaten my message because of a timeout...)

Again: it may be a wiring error, but I have double-checked all connections. The TB6560 boards lights up its respective LEDs for an axis whenever Linuxcnc is trying to move them. So that seems to work.
The cabling from the TB to the motor is still 'original' from before the retrofit. So that should work, too.

That leaves the possibility that the TB has died on me.

Is there a way to test it? I'm not an EE, but I have an oscilloscope available.

What would be a recommended replacement for that board? The motors have 3 A and I need 4 axis in total.

Thanks.

Are you sure the enable polarity is correct?

Its possible the drive is bad but having all 4 axis fail
at once does seem somewhat unlikely (unless the enable circuitry
was damaged somehow).

Can you try with your GRBL setup?

 

I got myself a TB6600 (like this: www.roboter-bausatz.de/p/schrittmotortreiber-tb6600-4a-9-42v) for testing.
This one has STEP & DIR +- and was wired accordingly.
I wired Enable+- directly to the 5V/GND of the Mesa card (which I guess I will have to change eventually (the wiring, not the card)).

After powering up, the motor immediately locked up and started making a low noise that sounds like white noise or like water running through a pipe. The motor did not get warm, but also didn't move when ordered to.
The TB6600 has been configured to 3A and 16 microsteps.

Thanks.

You might try a longer step pulse (say 20 usec = 20000)

Thanks, that worked!
I had to set it to 40000 and then it worked. I tried 30000 and it stopped working. Is that value unusually high and do I need to be worried?

Another question: I have wired the enable pin to 5V/GND of TB1 on the mesa card. So when I power the machine up, the motors immediately get the enable signal. Is that desired behavior, or should I use an extra output to switch that enable signal directly or via relay? Linuxcnc would set that output only when a motor is supposed to move.

And another one: the motor runs a bit 'hard'. In a PID controller, I would assume that P is too high. I see that there is something like that in pncconfig. Is this the place to do something about this? And is there something better than try and error?

Thanks!

Edit: forgot to add that this is with the TB6600. With the new setting, I rewired it to the 6560 and it stopped working. I have not yet tried if the grbl solution still works.

I think that's typical for these drives

You could use a 7I96S SSR output to drive enable if you wish
(or even a spare step/dir output if you dont need all 5 axis)