[step-manufacturing] Minutes of January 4th Conference Call

[Martin Hardwick]

Attendees
--------------
Martin Hardwick, STEP Tools, USA
David Loffredo, STEP Tools, USA
Vincent Marchini, Ameritech, USA
Mikael Hedlind, KTH, Sweden
Alain Brail, AIrbus (retired), France
Bob Erickson, Pratt & Whitney, USA
David Odendahl, Boeing, USA
Fred Proctor, NIST, USA

Apologies for Absence
-----------------------------
Sid Venkatesh, Boeing USA
Tom Scotton, CCAT, USA
Bengt Olsson, Sandvik, Sweden
Aydin Nassehi, University of Bath, UK

1. Planning for the ISO STEP Meeting, June 10 to 15 in Stockholm

ftp://ftp.steptools.com/private/CAM_exchange/TC184_SC4_Stockholm_Juni_2012_(T24_presentation_2011-11-18).pdf 


The STEP-Manufacturing team (Wg3/T24) will meet for three days during 
the SC4 meeting. On Tuesday we will meet with the rest of STEP and the 
agenda will focus on common issues including our requirements for 
tolerances and kinematics, and possible mutual interest in an event 
driven implementation method for JavaScript. On Thursday we will meet at 
KTH (see map in the pdf - the distance between the two locations can be 
walked in one hour). This day will focus on demonstrations and industry 
requirements for the second edition of AP-238. The last day will review 
the proposed extensions for the second edition AP-238 data model.

We began planning two demonstrations for the industry day at KTH.

The first demonstration will show machine tool simulation on a 
holographics display. For this demonstration we will drive six 
projectors with 3D images of a machine tool executing a machining 
program.  The images will be generated from the JavaScript/WebGL/XML of 
the web demos, or from the desktop ST-Machine depending on the relative 
efficiency of the two methods. The JavaScript data is preferred because 
it allows custom user interfaces to be generated for different applications.

The second demonstration will show how the new kinematics model can be 
used to predict machine tool accuracy for a given setup. KTH will extend 
the AP-242 kinematics model to include error property data for each link 
in each kinematic chain. STEP Tools will  generate minimal and maximal 
tool path coordinate data from this error property data. A TBD piece of 
analysis code with then analyze the path volume to determine where this 
is a possibility of tolerance failures. The kinematics link error 
property data will be generated from the ASME 59.5 model and double ball 
bar tests.

2. Kinematics data exchange

a. Setting the cutting tool in the spindle

Many of the STEP-NC models have the cutting tool set too deep into the 
spindle with the result that only a small fraction of the tool appears 
to be available for cutting. We discussed how to fix this problem and 
there was agreement that in the future the machine tool models should 
include a gauge line/diameter definition. We will then be able to match 
this gauge line up against a similar line on the cutting tool holders to 
set the tool holder into the spindle in future simulations.

b. Tool Retract Motions

This can be a three part issue. The STEP-NC program may include a 
retract motion that takes the tool to a point that is safe with respect 
to the part geometry. This is followed by a subprogram motion which 
takes the tool to a point that is safe for the setup or considered safe 
by the enterprise. Finally the machine tool vendor defines a motion that 
takes the tool to the best rest point for the machine.

We discussed the second two motions. In the new HTML5 environment we 
should be able to define the subprogram motions as JavaScript listener 
code that is set to execute whenever a retract operation is found in the 
STEP-NC data. The third type of motion needs to be defined as part of 
the machine tool model and is discussed in the next section below.

c. Tool Change Motions

Today most enterprise are not defining the tool change motions because 
the effort is not worth the payoff. However, there was agreement that if 
the machine tool vendor could include a definition of the tool change 
motions in its machine tool models then this would be of benefit because 
the enterprises would then be able to include them in their error 
checking at near-zero cost.

We discussed how to define such motions. One possibility is to define 
them in AP-238 but a better solution may be already defined in the Part 
105 kinematics model included in AP-242. This solution has the advantage 
of allowing all the machine specific motions (including the retract 
defined above) to be included in one master model of the machine.

3. Implementation Method for JavaScript

STEP Tools has extended the WebGL demonstrations to include simulations 
of five axis machines and robots. In the next phase it will extend them 
to show the machining programs running on the different machines.

http://www.steptools.com/demos/

The JavaScript programming environment is very interesting for machining 
because it is event driven. Consequently programmers may be able to 
extend the machine tool simulations to include event driven actions such 
as G-code generation.

The next meeting will be held at the regular times on Wednesday January 
25 one week later than usual due to a prior commitment by STEP Tools. A 
full recording of this meeting is at the following URL.

ftp://ftp.steptools.com/private/Undecided/stepmanuf_telecon_20120104.wmv

Action Items
----------------
1. STEP Tools to publish the JavaScript and XML data being used for the 
WebGL demonstrations.
2. KTH to investigate the feasibility of using this JavaScript and XML 
to drive the holographic display for the machine tool simulation 
demonstration.
3. KTH to develop scenario for the toolpath accuracy demonstration.
4. Boeing to obtain machine tool and fixture model data for a high 
accuracy application.
5. KTH to develop cutting tool and machine tool data with gauge line 
data for setting the tool into the spindle.
6. STEP Tools to extend the web demos to run the machine programs on the 
machine tool models.

Martin Hardwick
Team Leader ISO STEP-Manufacturing