Mesa 7I43-4 first tests but no sucess

HI
It's me again with mychangeover to Mesa 7i43

I attched the zipped config

I read the manuals als did man hostmot2 and beiv to understand now.

and I have checked with simple GPIO config that all the pins can work as input and work fine as outputs..
Either pulled up external with 5V ( my normal mode) or via the boards switch with 3.3V...

My Problem:
If I use 4 stepgen I get no output signals, testet with osciloscope. Neither steps nor dir...

Also chargepup is not changing..

What have I done wrong?

Any hints? I tried some ours but now

Mike

Mike is it coming out of estop and the machine is on?
Does the display in AXISdisplay movement but the steppers don't move?
Do you have a physical signal connected to estop-ext (GPIO 39)

also did the steppers move in PNCconf's tune test?

Hi

Oops, I forgot two explain:

I get E-stop and limit switches correctly.

Yes the macine is runnig in axis, showing usual moves... Usinf the LinuxCNC engraving.

It's on my test configuration, witch was always runnig fine with parports.
Same L297 boars with small steppers to see things moving.
Checked all with Fluke hand oscillopscope.

NO output signal seams to be driven... Maybe a watchdog-issue ?

to make a test I used pure pncconf ... so it should be fine.

Mike

AND
NO also in tune test they do not move!

Does the built in hm2-stepper example config work? (as far as step and dir go)

you have edited the HAL file.
It mentions LEDs in the hostmot2 loading line - PNCconf does not configure LEDs
does the 7i43 have LEDS?
remove the num_leds=-1 from line 9 of the mesaCNC.hal file.
it is unnecessary anyways as -1 is the default

It shouldn't matter but lets start there...

The 7I43 has 8 user LEDs

Salu

Yes PCW, your idea was a good hint.

The simple demo runs fine...

Now I can do debugging myself why the other config doesn't like me. ;-()

THX
Mike

Hi to all
Another annoyance to get my configuration changed to MESA 7i43.

I already used the same HW with normal par port card and I use 5V with active high signalling. I have soldered 1K resistors to ground, parallel to all inputs and switches connecting to +5V when active. This was good enough to get stable signals, also using shielded cable.

Now I'm trying to switch to mesa and by first verifying the input signals with pncconf, the MESA "launch Test" . And I see all inputs flickering randomly from high to low.
If the mesa card was not yet loaded, all inputs stay at 0V, Is I want. After loading the pal program they go up. ( some 1.49 V ) That gives the random reading.
OK, I understood that there is a jumper W3 that can pull all pins to 3.3V. So I took it away. Gave not a better result. Restarted PC same behaviour.
I believe that I understood from the docs that it is possible to set the pull up function additionally by the sw.

My question:
Is it possible the way pncconf works is setting these pull ups always? And where can I change that behaviour?
Means I can keep 1K as normal resistor, as 5mA is enough. I don't want to install a heating just to see some input signals.

Mike

Heres the relevant 7I43 manual page (page 18):

OPERATION
STARTUP I/O STATE
When the 7I43 is used to control external equipment that is sensitive to the initial
pin states, the pre-configuration pull-ups should be enabled (W3 UP), and for most
configurations, pull-ups resistor should be configured on any I/O pin that is not driven when
configuration is complete (bi-directional pins and pins whose function is assigned by
software). Since the only definable pre-configuration state is with pull-up resistors on the
I/O, this means the proper I/O polarity is active low (so that all outputs are in the in-active
state at power up)

All pins default to High at startup. any other option will have problems
so any home built daughtercards should follow this standard.
If they do not you will have to have work-arounds on every pin and lots of trouble