Advanced Search

The 7I76 can have different HAL pins depending on mode (user settable) and firmware version. Also you can connect an external SSERIAL card to the 7I76 and that card has the same possibilities, though it will show up as a a SSERIAL device
under the main FPGA card.

Only the older SSERIAL cards like the 8I20 and 7I64 have fixed pins, all new cards have modes and new modes with new pins may be added.

Not yet.
(I don't know how to wire up the 7i76 yet)

Can anyone post a print out of the HAL pins when the 7i76 is connected to the 5i25.
Thanks.

and ngctouch to start the torch and set the pierce height, cut height and pierce delay.

John

It uses findtop to set the Z 0.000 for the material...

John

I created a subroutine for ngcgui that cuts a square with radius corners on my plasma torch.

John

wow- from a commit 4 years ago!

thanks for the prompt reply, -i'm still getting over some fever i had. so i ain't been online.

i only noticed the behaviour when trying to cut a small dia sq thread,and had to resort to
this....

####################################################
(info: G33 style threading Internal)
o<th_33> sub
#<ToolNumber> = #1 (= 1 Call Tool or 0)
#<Speed> = #2 (= 300 rpm)
#<diameter> = #3 (= 50.0 Diameter)
#<pitch> = #4 (= 2.0 Pitch)
#<inital_depth> = #5 (= 0.35 Inital Depth)
#<thread_depth> = #6 (= 1.2 Depth Px0.866)
#<spring_pass> = #7 (= 1 Spring Passes)
#<z_start> = #8 (= 5.0 Z Start)
#<z_end> = #9 (= -15.0 Z End)
#<x_offset> = #10 (= 0.5 X Clearance)
#<Coolant> = #11 (=8 8 is on, 9 is off)

(LOGOPEN,paramnum)
; Tool change
O120 if [#<_Tool:Number> NE #<ToolNumber>]
; Change Tool
M09 ; coolant off
O121 if [#<_Tool:Number> LT 20] ; front turret
G00 Z#<_Frontchange:Z>
G00 X#<_Frontchange:X>
O121 else ; Rear turret
G00 Z#<_Rearchange:Z>
G00 X#<_Rearchange:X>
O121 endif
G49 ; cancel toolength offset
T#<ToolNumber> M06
O121 if [#<_Tool:Number> LT 20] ; front turret
G00 Z#<_Frontchange:Z>
G00 X#<_Frontchange:X>
O121 else ; Rear turret
G00 Z#<_Rearchange:Z>
G00 X#<_Rearchange:X>
O121 endif
G43 G7 G97 S#<Speed>
M03
M#<Coolant>

#<_Tool:Number> = #<ToolNumber>
O120 else ; No tool change -continue.
M#<Coolant>
O120 endif

#<thread_depth> = [#<thread_depth> * 2]
#<x_offset> = [#<x_offset> * 2]
#<inital_depth> = [#<inital_depth> * 2]
#<Cut_Dia> = [#<diameter> - #<thread_depth>]
#<drive_line> = [#<Cut_Dia> - #<x_offset>]
#<Cut_Dia> = [#<Cut_Dia> + #<inital_depth> ]
#<Cut_increment> = [#<inital_depth> * 0.72]
#<spring_pass_num> = 0
G00 X#<drive_line> Z#<z_start>
(log,DrvLine = #<drive_line>)
(log,CutDia = #<Cut_Dia>)
(log,CutInc = #<Cut_increment>)
; in here
o101 while [#<Cut_Dia> lt #<diameter>]
G00 X-#<Cut_Dia>
G33 Z#<z_end> K#<pitch>
G00 X-#<drive_line>
G00 Z#<z_start>

O121 if [#<Cut_increment> LT 0.01] ; minimum cut depth
#<Cut_increment> = 0.01
O121 else
#<Cut_increment> = [#<Cut_increment> * 0.2]
O121 endif
(log,CutInc = #<Cut_increment>)
#<Cut_Dia> = [#<Cut_Dia> + [#<Cut_increment> * 2]]
O131 if [#<Cut_Dia> GT #<diameter>] ; maximum cut
#<Cut_Dia> = #<diameter>
O131 else o101 while [#<Cut_Dia> lt #<diameter>]
#<spring_pass_num> = [#<spring_pass_num> + 1]
O131 endif
(log,CutDia = #<Cut_Dia>)

O122 if [#<spring_pass_num> GT #<spring_pass>] ; spring cuts
#<Cut_Dia> = [#<diameter> + 1]
O122 else
O122 endif
(log,DrvLine = #<drive_line>)
(LOGCLOSE)
o101 endwhile
G00 X-#<drive_line>
G00 Z#<z_start> M09
o<th_33> endsub

################################################################

Which was a lot of writing for such a small item.

hopefully this can be sorted out, as on small diameter internal threads tool ridgidity/clearance is a big issue.

wanispissd wrote:

I think Andy is right, that the current limitation of emc is that it cannot go in reverse. I was thinking last night (a bit abstractly) that if you could output all the axis positions at a given interval (based on feedrate) and have the most recent ..say .. three positions retained, you could use those as destinations for a program that could reverse motion.

Can you read C?
I think that the code here
git.linuxcnc.org/gitweb?p=emc2.git;a=blo...4944b795c1747d56be4c
Is what calculates the actual positions, and it would be interesting to know how easy it would be to make it back-up.
The main question would be whether the previous path elements persist in the queue, or whether they are deleted after being completed.

This is also the code which would have to change to support pause/jog/touch-off/resume I think.

Peter do you mean for instance the 7i76 could have different components depending on something like dip switches on the board?
While I'd like PNCconf to read live systems ( and I think Andy is thinking on a mechanism) most things are predictable.
Of course it requires the user to select the right answers, which is unpredictable - but that is no worse then someone hand writing a HAL file.

For the 5i25 the user must know what firmware the board was programmed for and pick that firmware in PNCconf.
Then the user must pick the right sserial daughterboard in the right sserial channel.
then what choices are there in the sserial daughter board?

Do you see one 5i25 firmware / daughter board choice that will be heads above more popular?

Chris M

Yes, it has a local SSERIAL device for the 32 inputs and 16 outputs, spindle output voltage, direction, and enable flags and possibly 4 analog inputs (and one analog field voltage pin)

I think it may be difficult to completely support via pncconf unless pncconf can work with data from a
live system since some pins may or may not be there depending in the SSERIAL device mode.

The I/O pin names are Input and Output not sure about the other names yet

Peter can you explain how the 7i76 works? I'm guessing it uses one of the smart serial channels for the 48 I/O points? Which one?
Is the 48 I/O points similar to the 7i64 ? ie the inputs and outputs are designated (they are not GPIO)

thanks Peter I will work on supporting that.

All thats required there is deleting the "firmware=blahblahblah"

Chris,

I have a 5i25 and 7i76 coming next week and will test pncconf when it arrives.

John

I have pushed to 2.5 most of what is needed to support the 5i25, aside from the HAL code to actually load the 5i25
(it has to be different if it doesn't load firmware each time)

one also has to use the 'back-door' way of adding custom firmware to PNCconf, using a file with a python array.


Andy, would you be willing to test ?
You remember how you added custom firmware last time?

It should work aside from loading the card. (in other words you will need to hand-edit the loadrt hm2_pci line.)