On 7/26/2011 2:43 AM, Greg Bentzinger wrote:
> Message: 2
> Date: Mon, 25 Jul 2011 21:32:52 -0400
> From: Dave<e...@dc9.tzo.com>
> Subject: [Emc-users] Machine re-control with two "X" axes
> To: emc-users@lists.sourceforge.net
> Message-ID:<4e2e1944.7090...@dc9.tzo.com>
> Content-Type: text/plain; charset=ISO-8859-1; format=flowed
>
>
> I am looking at re-controlling a machine that has two parallel "X" axes
> - one is called X and the other is U.
>
> The machine consists of two parallel sliding tables beneath a single
> head.   The head is mounted to an overhead Y axis and Z axis.
>
> Only one sliding table is used while cutting a part.  The idea was that
> one table could be loaded with a new part to process while the other
> table was being used to cut a part.
>
> The existing control, an 11 year old Osai control, has the ability to
> switch which between the hardware X and U axis.   The programs are
> written only in terms of X,Y, and Z.
>
> I believe there is a switch on the console to determine which table will
> be used prior to the part program being run.
>
> My question is:  What would be the best way to implement this two "X"
> axis control scheme with EMC2?
>
> Thanks,
>
> Dave
>
>
> It would seem that some fancy HAL programming might be able to address this 
> as either a Pallet code (more useful) or Table code (kinda unique to your 
> machine.
>
> Either way EMC could use a normal XYZ type program by defining which pallet 
> or table is active so that only one "X" axis exists at any given time.
>
> Pallets add another level of complexity to the machine coordinates because 
> axis boundries can be different for each. In a sense the G53 values could be 
> unique to each pallet. Adjusting and applying proper workshifts (G54-G59 
> ect.) allows the Tool table to use the same tool Height offset value without 
> concern for which pallet or table is active.
>
> I know as EMC2 exists (my not so current version) that to replacate the work 
> I do on a Kitamura HX400 4axis horizontal with cube pallets - I would be 
> replacing workshifts via G10 extensively - It was common to use 60-80 sets 
> during a run.
>
> At some future time I can see that the whole Tool Table / Workshift tables 
> and variables will need to grow dramatically.
>
> My Hal / pallet idea only works because you never are cutting both at the 
> same time.  A larger problem would be a dual turret lathe, and at this point 
> the easy way out for that might be running a second EMC PC slaved to the 
> master machine control.
>
> Greg out - must sleep!
>
>
>    
Thanks for the ideas..    Right now I think they are doing an "Osai" 
version of G92 after homing to change the origins, which is an odd way 
to implement an offset IMO.

I'm not confident that the guys who did the original machine did a great 
job on the controls.

Dave


------------------------------------------------------------------------------
Magic Quadrant for Content-Aware Data Loss Prevention
Research study explores the data loss prevention market. Includes in-depth
analysis on the changes within the DLP market, and the criteria used to
evaluate the strengths and weaknesses of these DLP solutions.
http://www.accelacomm.com/jaw/sfnl/114/51385063/
_______________________________________________
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Reply via email to