Advanced Search

Hi Rick,

This is pretty cool. I never would have thought that ngcgui would be so powerful... I love it on my lathe. I can't wait to get my BP Series 1 over to my new shop and convert it to EMC.

John

501 sample file

Now that we have the ngcgui as a tab in AXIS it is just so easy to run sub programs that we are using them more and more.

Often we have a completed ngc file for a part we machine and it works fine...if we need just one, however sometimes we need multiple copies of the same parts and we have large enough stock to machine several.
The concept of this untested sub is to take that part and duplicate it as many times as necessary in what ever direction you wish using the ngcgui.
This was set up for XYZ machine such as a mill/router/plasma.
In order to avoid error with G10 command G40 is used before G10.
Tool offset should be O.K. but will need to be applied in the part file.


With this sub you should be able to take your existing part file and duplicate it.

You can select:
What file to run.
Where to start the first copy.
How many copies.
The distance between and direction of the copies.
Where the machine should park when done.


To try just make your part file a sub by..

Name the part file with a numerical name (say 501.ngc)
Make the file a sub by adding

o<filename> sub
to the beginning

and adding

o<filename> endsub
to the end.

Be sure to have all files in the correct directory.

For example the file 501.ngc is

o<501> sub
F20
g0 x0y0
g1 z-.125
g1 x1.5
g1 y-1.5
g1 x0
g1 y0
g0 z.025
g0 x1.5
g1 z-.125
g1 x0 y-1.5
g0 z.025
o<501> endsub

Because ngcgui allows you to string several subs together before you finalize you should be able to set up several rows and columns of the part.
Again this is untested and you will probably only want to use this as a starting point to fit your needs.
You probably need at least EMC 2.4 or later.

Rick G

Hi,

Thanks for the effort. I also tried to get some information from Panasonic, and from Sodick, the machine manufacturer, without luck. I have made a chart of pinouts, taken from the machine electrical diagrams, and have matched it with some information regarding the current controller.

The chart is available at my dropbox: dl.dropbox.com/u/8975057/CURRENTCONTROL.xls

The controller is able to output both PWM and analogue signals. I am currently testing signals on a working machine to find out what connections are being used.

My Panasonic guy can not find anything on those part numbers.

John

ioiotutu wrote:

Another question, can i see in a external soubroutine a global variable?

Yes

However, to do so violates all the principles behind object oriented programming, that give rise to the use of discrete subroutines in the first place.

This may sound pedantic, but there is good logic behind it.

A subroutine should not be reliant on any external data, just the parameters passed when it is called.
That way if you find a far better way of implementing the moves in the subroutine, you can rewrite it and so long as the parameter interface is the same, it will continue to work with all the code which previously used it.

If you use a global variable then you might just as well write the whole lot in-line and dispense with the separate file.

The subroutine you wrote will not work unless each piece of code you use it with, has the same global declaration

So what you should do is just pass as a parameter

O<posi> call [20] [1000]

regards

Not sure how much power you need for the relays but the usb is limited.
But of course find the location of the problem with testing.

Rick G

Andy,
Thanks for the idea, it did not occur to me to get clean 5v that way.

A spare USB cable would get me good 5v also, right?

I will give it a try.
~*Brian

Rick,
Good idea.
I will try switching pins and see what happens.
thanks
~*Brian

The buzzing might be from the wall-wart, they are often not the most stable or well regulated supplies.

Do you have any other power supplies you could test with? (Black and Red on a spare hard drive power plug in the PC, for example)

If you exchange the output pins to the two relays does the problem follow the parallel port pins to the other relay or does the same relay buzz when on the other signal?
If a power supply problem would expect both relays to be a problem, if so a cap on the power supply may help.

You may also want to look at the parallel port tester...

www.linuxcnc.org/component/option,com_ku...d,4743/lang,english/

Rick G

The machine I am retrofitting came with encoder cables, connected from the encoder directly to the controller. The Panasonic manual I have found to best describe my drive/motors show that the encoder should connect to the drive.

To add to the confusion, the included encoder cable is comprised of the correct plugs, a 25 pin dsub on the controller side, that the manual says should connect to the drive. But, the pinouts described in the manual are very different than what I am observing.

For others, John is referencing the encoder diagram I am working off of:

I will try to contact Panasonic as well. Thanks for the help.

just relay #2
yes, I guess about a 60 cycle hum
power supply is a 5v wall wart, it powers both relays
I think the 5v supply is fine (rated 5v 700ma, but measures 5.6v actual)
this is hal config...
Owner 6 Type bit Dir OUT Value FALSE Name motion.spindle-on ==> spindle-on

~*Brian

I'm trying to contact my Panasonic friend that I get my servos and drives from now to get some info.

I wonder if the RX RX/ are feed back to the drives like a tacho velocity feedback. (from the IRC this morning)

John

I am guessing that the problem is that you don't have the cables? Otherwise it ought to just be a case of wiring them together and applying the control voltage to the drives.

A direct email to Panasonic might help.