| my account | login-logout | resources | support | catalog | home | get webcard |

Online Classroom


Post New Topic  Post A Reply
search | help desk | commons
  next oldest topic   next newest topic
» Online Classroom   » Celestial Navigation   » Public Discussion of Cel Nav   » Accuracy of Weems Parallel Plotter #120

   
Author Topic: Accuracy of Weems Parallel Plotter #120
Henry Breyer


 - posted March 07, 2005 11:55 AM      Profile for Henry Breyer           Edit/Delete Post 
I just bought a rolling plotter and measured a series of parallel lines, each 2 ft. apart. They were out of parallel by anywhere from 3/64" - 3/32" over the 2 feet. Is this an acceptable tolerance? I noticed that there is some play between the bar that the wheels are attached to and the housing that the bar sits in. Do they all have a bit of play? Are the Bi-rola rules any more accurate?
David Burch


 - posted March 07, 2005 08:52 PM      Profile for David Burch           Edit/Delete Post 
I am not sure i understand the data. Do you align the ruler carefully with one line, then roll to another parallel to it that is 2 feet away, and then discover that if one end of the 12" ruler is on the line the other end is off by the amount you quote?

It would seem that this type of measurement would have as much to do with the operator as with the tools, or am i missing something?

Maybe a related note: i have heard some navigators like to add small o-rings into the grooves of the rollers so that they get better grib and ultimately better mesaurments, but i have not tried this myself.

I have heard of the bi-roller but have not used it myself.

(we can look at the accuracy acceeptance level once the measurement is clear to me.)

From: Starpath, Seattle, WA
Henry Breyer


 - posted March 08, 2005 05:30 AM      Profile for Henry Breyer           Edit/Delete Post 
David,

Sorry I was not very clear about my method of measuring. First I noticed that as you roll the plotter along the paper, a "bubble" of paper can form in front of the plotter which could have been the reason for the misalignment so I tried to eliminate as many sources of error as possible. That is why I went to such lengths to measure the accuracy of the plotter as described below:

Realizing that a warped table surface could make the plotter roll out of parallel, I took a piece of medium density fiberboard (which is usually amazingly flat) and measured it for flatness with a machinists straight edge. It was so flat that I could not detect any light between the straight edge and the MDF in any direction.

I then lay the plotter on one side of the MDF and drew a line along the edge of the plotter from one end to the other (12 or 13 inches). Then I rolled the plotter to a point 2 feet away from the original line and drew another line along the entire length of the plotter’s edge. When pushing the plotter, I held it at the center of the grip and carefully pushed the plotter across the MDF. Before drawing the first line, I checked that the wheels of the plotter had a decent grip on the MDF. Using the machinist’s straight edge, I measured the distance from the right hand corner of the 1st line to the right hand corner of the second line and did the same on the left hand corners of the two lines.

I did this a number of times and each time, I got different measurements out of parallel. They ranged from nearly “perfectly parallel” to as much as 3/32” out of parallel. The typical range was from 3/64" - 3/32" out.

Thanks,
Henry

David Burch


 - posted March 09, 2005 11:01 PM      Profile for David Burch           Edit/Delete Post 
Ok, i see what you did... and you find 3/64" - 3/32" error on a 12" length. that is about 0.05 to 0.09 out of 12, which is 0.05/12 to 0.09/12, which corresponds to angles of arctan(0.05/12) to arctan(0.09/12) = 0.2° to 0.4°.

I would have to conclude that after rolling some 24 inches to be off less than half a degree would be more than acceptable for standard plotting. It is in fact rare to have to roll that far anyway. usually we just roll to the nearest meridian and read the angle, which we can read to about ±0.5° but usually not much better.

In passing, if you do want to do the most accuracte plotting, then chances are two large navigator's triangles are the best approach. This works very well, but some find their use awkward — many old timers sware by them; many new to their use sware at them.

We always try to use the weems rolling plotter, but if it won't work (edge of the table or chart, folds in the chart, wrinkled charts, etc) we do not argue with it, and just grab a set of parallel rulers.

From: Starpath, Seattle, WA


All times are Pacific  
Post New Topic  Post A Reply Close Topic    Move Topic    Delete Topic next oldest topic   next newest topic
Hop To:

Starpath School of Navigation

Copyright, 2003-2021, Starpath Corporation

Powered by Infopop Corporation
UBB.classicTM 6.3.1.1