Advanced Search

Sounds like either a power supply issue (marginal 5V)
or corrupted flash (wrong bitfile written, which is possible
with older mesaflash versions)

I guess it could be a failed flash chip, but this is pretty rare.
 

With your existing hardware, lowering the base thread value
is the only thing you can do. How low you can lower it depends
on the kernel (turbo mode etc) and isolcpu settings.

Did you run a latency test? (why did you choose such a long base thread period?)

Hi,
I have a 7i96s that worked fine for a day and then during a reboot it was no longer recognized by linuxcnc. Looking at the LEDs on the card I noticed that  the init, done and power LEDs remain on and I don't have the ethernet link up.
Following the manual I put the w1 and w2 jumpers in the up position, but the board remains with the LEDs on and ethernet link is down!
What can I check?
Michele
 

It sounds like a large voltage spike (>25V) is being coupled into the encoder input.
I guess this could be a wiring error. I would disconnect the encoder connector at the 7I96S
and verify that there is no connection from any THCAD digital side pin to the
plasma source GND

Is the cable shielded? If so the shield should be terminated at 7I96S TB2 pin 9 (GND)

 

Hi everyone,

I am part of a research group here at Iowa State and we are using LinuxCNC to build a laser based directed energy deposition system. We have some Omron G5 R88D-KN01L-ECT drivers that a few other people have seemed to have gotten working on here and I was hoping you might be able to provide some input if this describes you.

Right now I'm just trying to get position exchange information in the HAL viewer for a single axis.

Thanks to the Ethercat installation from repositories - how to step by step , and EtherCAT: How to use the dedicated OmrG5 lcec driver? I think I have the XML setup correctly, as the driver is happy and displays "--" (preop, waiting, no error) until I launch LinuxCNC, which about 1-2s after the main UI loads the driver drops into the '90' error. This occurs regardless of if I have the drive configured as a "generic" device or if I use the OmrG5 driver. I've also tried changing the servo thread cycle time through all of the supported cycle times listed in the documentation (0.5, 1, 2, & 4ms) and no change in behavior.

I've also tried Beckhoff's Ethercat Configurator to make sure that the cycle time was set appropriately

I've attached a snippet of the documentation for the communications synchronization error. Any help/thoughts would be incredibly appreciated as I've been banging my head against this wall for a few days now. The XML, INI and HAL files have also been attached.

Thanks!
 

Hi PCW,

My connections are in twisted pair cables and are connected as follows:
 

From TB4 (200XT plasma system) to THCAD-300 to TB2 (7i96s)

Edit:
 
 

If in question about hardware pin, parameter names, type this
at a terminal when LinuxCNC is running:

halcmd show all hm2

or to save all the names in a file (as a copy_paste source) :

halcmd show all hm2 > hm2names.txt

5:

For faster normal encoders, here is firmware that adds 6 encoders on P1:

 




 

I would expect the THCAD300 to connect to the 7I96S TB2 pins:

7  A+
8  A-
9 GND
12 +5V
A+ and A- should be a twisted pair

Its also very important that none of those pins connect to _anything_
inside the plasma source.

Hello everyone.
I have an error with the reading of arc voltage (-45 V) after one piercing.
Before piercing I was reading 0 V.

My plasma source is AutoCut 200XT from Thermal Dynamics.
I am using a 7i96s mesa card with THCAD-300 also from mesa.
The interface I am using is QTPlasmaC.

The THCAD-300 is mounted inside of the plasma source connected directly to raw Arc Voltage.
The output from THCAD-300 is connected to the encoder from 7i96s (TB2.7 and TB2.8).

Already switch the THCAD-300 and 7i96s to brand new ones and at start the readings of arc voltage seems to be ok.
But when I put Torch On the Arc Voltage goes to -45 V, after that the value doesn't change. No matter if plasma is on or off.
If I look to halshow I get no readings at all in encoder counts signal.

Looks like for some reason that I cannot understand I am damaging the encoder input on the 7i96s board.
I've made the exactly same connections on Hypertherm Plasma Systems and works fine.