I've got a series of circuit board diagnostic tests, where each test is
uniquie for a specific part number.  A complete test for a single part
number may consist of say 20 individual tests, each returning values with
varying units (ohms, volts, hertz, sec, etc).  I'm not sure if what I'm
attempting will produce useful results, but it can't hurt while becoming
familiar w/ DX.  Our goal is to identify trends within this data to predict
failures of components, before they actaully fail, providing a visual
display for quick identification.  I wanted to try placing the test numbers
(of the 20 individual tests) on the y-axis, and use the x-axis for varying
serial numbers.  The z-axis could be used for multiple instances of the same
serial number.  This may produce an uncomprehensible display, but you can
never have too many techniques tucked under your belt.  Since the 20
individual tests will have a broad range of values and units, I do not want
them to be connection-dependent.  Varying serial numbers, however, will
always have the same range of limits and units for an individual test, and I
would like to make these connection-dependent.  I'm envisioning the data
values to be expressed as deviations from the nominal value, taking into
account the range of upper and lower limits.  I'm hoping that this will
enable us to identify hot-spots (or a combo thereof) within the data which
may indicate a future failure.  I'll play with the multigrid thing in the
meantime...

thanks david,    jason


> Jason,
>
> I think we will need more information about what you are trying to
> accomplish. While this could be done with multigrid, I'm not sure how
> you are trying to render such as beast. Remember that connections
> dependent data doesn't interpolate where as positions depended data
> does. Also if you have positions dependent data (ie data on a grid
> intersection) this is going to be seen in both x/y.
>
> See if you can explain what you're trying to do by having the two
> different dependancies.
>
> David
>
> >i want to make data connection-dependant in one dimension, x for
> >example, and position-dependant in another dimension, y for example.
> >How do I accomplish this??
> >
> >Thanks,   jason
>
> --
>
............................................................................
.
> David L. Thompson                          The University of Montana
> mailto:[EMAIL PROTECTED]                 Computer Science Department
> http://www.cs.umt.edu/u/dthompsn           Missoula, MT  59812
>                                             Work Phone : (406)257-8530
>
>

Reply via email to