Advanced Search

My suggestion would be to use a normal config and the thcud component . The THC300 example is way too complicated IMHO.

John

OK - Thanks for the replies. I did read a post about the performance of the THCAD using a parallel port, but am thinking that it's better to have a slow thc compaired to no thc. Maybe it just won't work, but right now the changes I need are not much and want to be hopeful that this can work out. Even temporarily until I decide how all this is going to play out for me.

John - Thanks for the files. I have been going through them, but didn't know how much of the 5i25 config files I can learn from to implement into my parallel hal file. I did see your Sheetcam post file, but didn't think it included torch touch-off reference so a touch-off didn't have to happen every pierce. Also, I plan on using a scribe someday which this post incorporates along with some thc control, but I have that part turned off for now.

I will plug away and see how far I can go. Thanks for all of the information all of you have provided. I wouldn't be this far without it.

PK

Good suggestion John. I will build it into my first design. I am on number three already.

One important thing is the plasma cutter itself. If it does not provide a pre build reference voltage it could be dangerous to open the plasma cutter and create yourself the reference voltage as the voltage / amperage combination are dangerous.
I am currently waiting to my plasma cutter ends warranty to make this myself!

Paul,

The biggest changes will be the pin names and you only need the section that connects the output from the THCAD card to the computer. Just change the pin name to match the parallel port pin your using.

John

Thanks emcPT. Fortunately, my plasma cutter has a CNC port for 1/50 voltage reference, torch activation and ARC-OK. I think the Proma THC unit is designed to be mounted in the plasma cutter if a reference voltage is not available, but like you said - that would be a wait for the warranty to expire.
Thanks...

Maybe you already knew this but if you load the 'Plasma-5i20' sample configuration in Linuxcnc. The 'Plasma-demo.hal' and 'Plasma-demo.ini' files are set up for parallel port. They should get you very close to what you want.

Thanks Coolmo32. I did save those samples and went through the parallel files. I need to sit down again when I get a chance again and go over the 5i20 files, too. As of last week, I haven't been smart enough to see any pin assignment samples getting the THC info in. I just haven't had the time the past few days to sit down and go over John's files for the 5i20, either. I'm not saying the pin assignment isn't there, but I haven't found it yet. I see the pin assignments for everything else: Touch off, torch on, etc.

I was kind of expecting to find a line, generally, like:
net thc-data-input <= parport.0.pin-10-in-not

...or something like it.

in a terminal window, type:

man thc

this is the THC manual and lists the THC components pins

Note that no parallel port pin will connect directly to the THC component except ARC-OK.
The actual voltage reading is done by measuring the frequency that the THCAD generates
This frequency is measured by using the encoder components velocity output pin.

So voltage measuring connections are: THCAD --> Parallel port pin --> encoder component 'A' input --> encoder component velocity out --> THC component

Paul wrote:

Thanks Coolmo32. I did save those samples and went through the parallel files. I need to sit down again when I get a chance again and go over the 5i20 files, too. As of last week, I haven't been smart enough to see any pin assignment samples getting the THC info in. I just haven't had the time the past few days to sit down and go over John's files for the 5i20, either. I'm not saying the pin assignment isn't there, but I haven't found it yet. I see the pin assignments for everything else: Touch off, torch on, etc.

I was kind of expecting to find a line, generally, like:
net thc-data-input <= parport.0.pin-10-in-not

...or something like it.

I think you need somthing like this:

# Set up the Encoder for the THC
setp encoder.0.counter-mode 1
# added by tc
net freq-in parport.0.pin-11-in => encoder.0.phase-A
net thc-vel-in encoder.0.velocity => thc.encoder-vel
# these will need to be adjusted
setp thc.scale-offset 119000
setp thc.vel-scale 0.00037866834
#
setp thc.velocity-tol 100
setp thc.voltage-tol 1.0

I think it also needs this for z-axis:

# position command and feedback hijacked by thc comp
net emcmot.02.pos-cmd thc.z-pos-in <= axis.2.motor-pos-cmd
net thc-pos-cmd thc.z-pos-out => stepgen.2.position-cmd
net motor.02.pos-fb axis.2.motor-pos-fb <= thc.z-fb-out

Oh wow - Thanks Coolmo32! At one time I thought I might be able to work through this, but I never would have pulled that out of my ...head. I think I will get a chance to hook this up this weekend.

PK

Coolmo32 wrote:

net emcmot.02.pos-cmd thc.z-pos-in <= axis.2.motor-pos-cmd

Just a comment, and an opinion, but I am not fond of this style of signal naming. "emcmot.02.pos-cmd" looks too much like an "offfical" label that has a meaning to LinuxCNC, whereas all signal names are completely arbitrary.

andypugh wrote:

Just a comment, and an opinion, but I am not fond of this style of signal naming. "emcmot.02.pos-cmd" looks too much like an "offfical" label that has a meaning to LinuxCNC, whereas all signal names are completely arbitrary.

Understood. Thanks for the info.

John,
I am using your thcud component. I changed what I had to in order for thc to work. One question please. In the touchoff.ngc I set a G92.1 ond then Z50. As my Z max limit is set at 5mm, I get an error that the move will exceed the max limit. In actual fact the torch is 200mm away from the limit. Is there a way to overcome this situation?

That would be the expected behavior when you reset the G92 offset.

linuxcnc.org/docs/html/gcode/gcode.html#sec:G92_1-G92_2

John