PDA

View Full Version : Galil "Jog Buttons" program



Karl_T
12-28-2008, 09:48 AM
I have a nine input button setup for my four axis mill with galil card. The buttons are arranged in three rows of three buttons and are connected to inputs 1-9. My present host level program uses the center button for hi/lo speed jog, and the perimeter buttons for axis and jog direction.


I'm looking to write a Galil program for a better "jog Buttons" routine.

I've been reading the Galil command reference, and such a program is certainly doable. But, it will be a bit of a challenge for me as I've never programmed in native Galil.

This must be a common need. By chance has anyone seen a similar Galil "jog buttons" routine?

Karl

HuFlungDung
12-28-2008, 11:20 AM
Karl,
You lookin' for a 'crash and burn' jog mode? :D

I say that with tongue in cheek because I've had past experience with Bandit and Shadow which made extensive use of such a jog system using a number pad. While its functional, I most often seemed to end up crashing the tool after about a dozen changes of direction.

Much as the Camsoft implementation of keyboard jog sucks (in machine response time), I've learned to settle for it. My 'crash and burn' rate is about the same, either way :D

You have a copy of my Galil threading program which runs in Galil memory. Now this looks nothing like what you would use for a jog routine, but what comes to my mind as a difficulty, is that I had to use the Camsoft SUSPEND I/O commmand (something to that effect) for the duration of time in which the threading program runs in Galil hardware. Now it so happens that all of my I/O used onboard Galil I/O so I could still have the Galil program identify and take over any of the inputs that I required while threading was going on. This included buttons like Estop, Feedhold, Spindle On, Spindle Off.

In your case, you are going to need to make your Galil card identify the offboard I/O of your keypad. I don't have a clue how you go about that.

Karl_T
12-28-2008, 11:39 AM
...
Much as the Camsoft implementation of keyboard jog sucks (in machine response time), I've learned to settle for it. My 'crash and burn' rate is about the same, either way :D
....
You have a copy of my Galil threading program which runs in Galil memory. Now this looks nothing like what you would use for a jog routine, but what comes to my mind as a difficulty, is that I had to use the Camsoft SUSPEND IIn your case, you are going to need to make your Galil card identify the offboard I/O of your keypad. I don't have a clue how you go about that.

yes, your program is a big help to show how to run galil under camsoft, thanks. from this I can see how to suspend camsoft and run from the inputs on the Galil ICM board.

my Camsoft jog buttons is WAY better than keyboard jog. but it hangs the control too often. There must be a better way. I'm sure a low level (Galil) program won't hang the Camsoft host. Now, writing a good galil "jog buttons" looks to be a fair challenge.

Karl

keithorr
12-28-2008, 01:39 PM
I already have an app and can work with mouse or keyboard input. fast acting. can run concurrently with camsoft as long as a handle is available. (galil has eight, camsoft should only use two at the most) Murray, let me know if you want an eval version. Need to know how/if it affects the camsoft commanded position.

CarbideBob
12-28-2008, 03:42 PM
I have a nine input button setup for my four axis mill with galil card. The buttons are arranged in three rows of three buttons and are connected to inputs 1-9.

Karl

I have zero experience with camsoft so please excuse the following stupid question.
What are your pushbuttons connected to?

If you have something like a DB-10096 IO expansion which connects directly to the Galil card via a ribbon cable this is pretty straightforward.

If your IO is on a separate card there is no way for the Galil card to talk to it in native commands.

IMO you'd have to be a little nuts to implement a jog routine through the keyboard. The update time from the cpu in the keyboard to the main cpu is way too slow and the delay is going to give you problems for sure.

Bob

keithorr
12-29-2008, 12:44 AM
IMO you'd have to be a little nuts to implement a jog routine through the keyboard. The update time from the cpu in the keyboard to the main cpu is way too slow and the delay is going to give you problems for sure.

Bob

NOT TRUE. I use keyboard inputs for I/O outputs, motion starts and jogging and the machine response is to the eye as fast as typing letters in a message board reply form.

Speak of what you know, not what you suppose or guess. Keyboard jogging also works without appreciable delay in Mach3; either tapping for incremental move or keydown for continuous jog. I know the camslog jog routine is defective but that is no reason to assume all jog methods are deficient.

keithorr
01-04-2009, 03:58 PM
I have uploaded a jog text application. Only the X axis is enabled. It should handle all exceptions (no card detected, amps off etc). Use it with your mouse cusor.

Left click=motion in positive direction
Right click=motion in negative direction
Cursor leaves the control area=jog stops
Release Left or Right mouse button=jog stops

If using with camsoft, test by jogging + or -, then use the camsoft mdi to return to the original postion with a slow feedrate (start at X=0 before jogging)

Let me know what happens.

Zip File Location www.keithorrblowpipes.com/GalilJog.zip

This application requires the microsoft NET Framework to ensure platform compatability. You may need to download files from Microsoft before using the Galil Jog Test application.

Feedback appreciated.