[step-manufacturing] Minutes for 3rd conference call

[Martin Hardwick]

Attendees
---------------
Alain Brail, AlBavis
Martin Hardwick, STEP Tools
David Loffredo, STEP Tools
Mikael Hedlind, KTH
David Odendahl, Boeing
Bengt Olsson, Sandvik
Fred Proctor, NIST
Leon Xu, Boeing
Larry Maggiano, Mitutoyo
Ian Stroud, EPFL
Stephen Newman, University of Bath
Brian Kindilien, CCAT
Chris Pfeifer, CCAT
Sid Venkatesh, Boeing


Apologies for Absense
--------------------------------
Aydin Nassehi, University of Bath
Magnus Lundgren, KTH
Bob Ericksson, Pratt & Whitney

The latest version of the Moldy data includes flute count and spindle 
power data. We discussed how to use this data to generate appropriate 
feeds and speeds for each CNC that needs to machine the part. The 
following formulas have been suggested by Sandvik

Parameters to adjust:
      flute count   -goal is to maintain chip load by adjusting speed 
and feed.
      speed&feed override   -goal is to maintain chip load by 
adjusting speed and feed
      feed override   -goal is to change chip load by adjusting the feed

Machine tool limitations:
      speed max   -goal is to maintain chip load by adjusting speed 
and feed down to speed max
      feed max   --goal is to maintain chip load by adjusting speed 
and feed down to feed max

Logic
=====

Change of flute count:
      speed = speed_old * (flute_new / flute_old)
      feed = feed_old * (flute_new / flute_old)
      run logic for changed machine tool limitations

Change of speed&feed override:
      speed = speed_old * overr_speed&feed
      feed = feed_old *overr_speed&feed
      run logic for changed machine tool limitations

Change of feed override:
      feed = feed_old * overr_feed
      run logic for changed machine tool limitations

Changed machine tool limitations:
      if speed > speed_max then
            speed = speed_max
            feed = feed_old * (speed_max / speed_old)
      if feed > feed_max then
            feed = feed_max
            speed = speed_old * (feed_max / feed_old)

The data generated by these algorithms needs to be saved in a file 
that is specific to the use of this STEP-NC program on this machine.

A new version of the Boxy part has been generated by KTH. This 
version contains extensive in-process CAD models showing the "state" 
of the part before and after each workingstep and showing the volumes 
removed in each workingstep.

We discussed two sets of challenges. The first challenge is how  to 
document the two setup's in one STEP-NC file.
1. The simple STEP-NC solution is to put each setup in a different 
STEP file. The better solution is to model both setups in the same 
file. There is some support for multiple setip's in the STEP-NC DLL. 
A new version of the STEP-NC data needs to be created with these 
setup's to verify the functionality.
2. Tolerances need to be defined for the in-process models so that 
the system can check the configuration of the second set-up before 
machining begins and possibly check other models during the machining as well.

The second challenge is how to improve the accuracy of the machining 
simulations.
1. The simulator always put the part in the middle of the table, but 
in practice most CNC engineers would put the part off-center to avoid 
singularities.
2. A simulation of the part on the Trunnion machine shows that the 
default axis configuration is a poor choice. Post processors allow 
the user to select the "front" or "back" of the axis. Similar 
constructs are available in STEP-NC and we need to test them.

We discussed meeting logistics. We decided that SC4 T24 should meet 
for two days - Thursday September 24 and Friday September 25th with 
the meeting on the 25th possibly finishing early to allow for plane 
connections. The demonstrations will be given on the first day. 
Stephen Newman will produce a list of the equipment available for a 
live demo. The second day will be used for planning follow-on 
meetings and activities. Martin Hardwick will produce an agenda.

The agenda for the SC1 meeting on Monday, Tuesday and Wednesday is 
being developed by Ian Stroud. The Wednesday agenda will focus on the 
machine tool model and will be a joint SC1/SC4 meeting. Martin 
Hardwick and Ian Stroud will coordinate the agenda for this part of 
the meeting.

The T24 meeting on Thursday and Friday will be open to all visitors. 
Previous meetings have included representatives from CAD, CAM, CNC 
and Machine Tool vendor companies. Overseas visitors are always 
welcome. There are no nationality restrictions for this meeting.

The next conference call will be at the regular times on Wednesday July 15.

A WMV archive of the meeting is at 
ftp://ftp.steptools.com/private/Bath/stepmanuf_telecon_20090603.wmv

Action items
------------------
1.   Boeing to test the new feed-speed generation by changing the 
flute count of a tool and machining a Moldy part.
2.   STEP Tools to investigate creating a STEP file for Boxy with two setup's.
3.   KTH to define tolerances for the TO-BE model of setup 1 which is 
also the AS-IS of setup 2.
4.   David Odendahl to investigate STEP-NC method for restricting axes.
5.   Stephen Newman to develop list of equipment at Bath that can be 
used for a live demonstration
6.   Ian Stroud to make agenda for SC1 meeting
7.   Martin Hardwick to make agenda for SC4 meeting
8.   Ian and Martin to agree on agenda for joint meeting.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.steptools.com/pipermail/step-manufacturing/attachments/20090702/dbf99cd6/attachment.htm>