[Next] [Up] [Previous] [Contents]
Next: First Draft of The Up: The Second Family: MIL Previous: Modeling Strategies

Part Construction

The second proposed method, the circular sweep, was chosen as the best method for creating this part. It is much simpler to maintain the proper shape of the part, and requires fewer operations.

The part was constructed such that centerline of the part was along the z-axis, with the keying flat on the positive y-axis. The x-y plane was chosen to be on the top surface of the mounting flange. This way, the origin of the part will coincide with the origin of the mounting hole. Figure 4.7 shows the cross-section and a top view of the part, each with the proper co-ordinate axes.


Figure 4.7: Cross Section and Top View of d38999-24

As in the previous part, the part was constructed using the top cplane, with most of the construction done using the front view. All geometry was created on layer 1. Figures 4.8 and 4.9 show various stages in the construction of the part.


Figure 4.8: Stages in Construction of d38999-24, Part One


Figure 4.9: Stages in Construction of d38999-24, Part Two

This model was actually created two times. The first method resulted in several problems referencing points, and created more parameters than were actually necessary. This required a redesign in the way certain points were specified. During this time, it was also discovered that what was previously assumed to be a fixed dimension, namely the diameter of the threaded portion on the back side of the connector, was actually variable.

The countersink was taken to be a standard 45 degree angle. Equations were developed to handle the positioning of the countersink. The variable RearHt was added to be the height of the threaded portion on the back of the connector. This was constrained to be at least the distance between the back of the connector and the mounting flange.

Because CADDS5 can not handle zero length lines, certain adjustments had to be made to the model. In three of the shell sizes, the dimension J is zero. This dimension had to be forced to be at least 0.0025'' in order for all the members to generate. This was done using constraint equations. This strategy also had to be applied in the creation of the countersink. The main flange line ends 0.001'' before the threaded section when there is no countersink.

Many of the dimensions used to create the cross-section of the part were radii, but the dimensions given in the SPEC were diameters. In order to differentiate between what variables were the diameter right from the spec, and which were divided by two to form a radii, the radii had ``_2'' appended to the variable names. Also, the information for this part was collected from several different sources, because no one source listed all the necessary information. Two of the sources used the name ``B'' for two different dimensions, so the dimension from the second source was renamed to ``B2''.

The portion of the _tbl file containing the variables used is shown below. The variable Min_RearHt is used in the table instead of RearHt. The value of RearHt is the smallest value that is at least Min_RearHt that will still produce a valid model. If RearHt is too small, the countersink would actually end up adding material to the model instead of removing it.

###############     RearHt:         If Min_RearHt is zero, no countersink is inserted.
# Family Table:     Master :        fpts.d38999.24
###############     Nut Master:     fpts.d38999.28
Member_name   A       B       B2      G_2     HH_2    J       P       Min_RearHt  S \
      T_2      W      Full_Filename                             Nut Nut_Filename

The variables Full_Filename and Nut_Filename are the full name of the connector, as in the previous family, and the name of the nut that matches it. The variable Nut is just the size code for the nut.


[Next] [Up] [Previous] [Contents]
Next: First Draft of The Up: The Second Family: MIL Previous: Modeling Strategies

Last Modified: Wed Aug 28 14:41:29 EDT 1996

Gregory Marr <gregm@alum.wpi.edu>