DHTML Menu, (c)2004 Apycom
TPAS Case Study - Monty's mount 1-Sep-06 First 10 stars

Recently Argo Navis user Monty S. forwarded Wildcard Innovations some data from a sampling run he performed and I would like to take the opportunity to provide some feedback, which hopefully may be useful and interesting to both Monty and other readers.

The mount model was a Losmandy G-11 German Equatorial. The mount was approximately polar aligned but the SETUP MOUNT selection was set to GEM EXACT ALIGN. A one-star alignment followed by a star pointing test was performed.

Of the 32 objects sampled, 14 were stars, 16 were deep sky objects and 2 were planets. Of the 14 stars, one was a user-defined object.

The objects listed in order of observation appear here -

NGC 7479
NGC 7515
NGC 2281

Generally we only recommend using stars from the bright star catalog or brighter planets. For various reasons, it is often difficult to judge the center of many deep sky objects. For example, some objects are quite extended and others have centers that were measured from photographic plates whose emulsions in some cases are more sensitive than the human eye to certain wavelengths. However, I appreciate that Monty may have wished to combine some sampling with some observing and of course observing should always be the number one priority.

Therefore I will split this report into two parts, firstly showing the data from just the first 10 objects, all of which were stars. In a second posting, I will report on the results when all objects are considered.

This report includes links to graphs which are in Scalable Vector Graphics (SVG) format. To view SVG content, you will need to load a free Adobe SVG viewer plugin, which is available for Windows, Linux and Mac OS-X from here.

As we will be presenting not only this but future reports using SVG data, it is recommended interested readers take just a few minutes to load the free plugin into their browsers now.

The first graph shows the residuals of the first 10 stars as error vectors on an orthographic projection. An orthographic projection is a distant view of the celestial sphere as seen looking down on the zenith. The yellow squares represent the stars and the tails projecting from the squares represent the direction and magnitude (not to scale) of the error residual.

The utility of this graph is that also gives some feel for the distribution of the sampled stars in an Alt/Az sense. As one can see, the sampled points cover a reasonable area of the sky from near the zenith to down toward the horizon.

It may be useful at this point to point-out some features of the SVG viewer.

Object names and residuals: If one passes the mouse over the center of the yellow squares, the name of the sampled object and its pointing residual is shown.

Zooming: Hold the control key and click the left mouse button to zoom in at the mouse pointer location. Hold the control key and click-and-drag the left mouse button to select a region to zoom into. Hold the control and shift keys and click to zoom out. You can also use the zoom commands in the context menu which you can access with the right mouse button.

Panning: Hold the alt key and click-and-drag with the mouse to pan an SVG image. Hold the shift key as well to pan only in a vertical or horizontal direction.

The next graph is a scatter diagram. The raw residuals (i.e. no pointing terms have been applied to the data) are represented by the radial distance each point is from the center of the blue circles. The inner blue circle represents the Root Mean Square (RMS) value of the data. In other words, the RAW RMS, which in this case has a value of 13.6'. The outer circle is equivalent to the radius of the largest pointing residual, which was 19.1'.

When viewing this graph, keep in mind that the mount is not polar aligned and that TPAS is making no corrections. In other words, the graph represents the approx. performance that most Digital Setting Circle (DSC) devices could achieve on the same mount, with the same alignment using the same one star alignment method.

The next graph is also a scatter diagram. It shows the residuals of the same 10 stars when TPAS has been used to fit the index terms ID and IH and the polar misalignment terms, MA & ME.

Notice that the RMS has dropped to 8.2'. This figure gives some feel for the actual raw pointing performance of the mount itself. In other words, when polar mis-alignment is allowed for, this is how well this mount/OTA is capable of locating those 10 stars when no other TPAS error correction assistance is applied. Notice the two distinct groupings of stars, with four toward the left-half of the graph and six toward the right. These two distinct groups occur due to one or more errors within the OTA/mount that cause the error residual to change their apparent direction when the OTA is flipped from one side of the mount to the other. Classic examples of such errors are CH and NP, the collimation in hour-angle error and the non-perpendicular axes error respectively.

The next graph shows what happens when the CH and NP terms are added.

The RMS drops down to 2.5' with the largest residual, that of the star Kocab, at 4.8'. Notice the left-right cluster groups have now largely merged.

The next graph shows the addition of three more terms, in this case, the harmonic terms DCEC, DCES and HCES.

Notice that the RMS drops to 1.1'. Given the 8192-step effective resolution of the encoders fitted provide at best a resolution of 2.6' a step, this is an excellent result. The largest residual of the 10 stars was 1.9'.

Special thanks to Monty for making the raw data available.

Best Regards

Gary Kopff
Managing Director
Wildcard Innovations Pty. Ltd.
20 Kilmory Place
Mount Kuring-Gai NSW 2080
Phone +61-2-9457-9049
Fax   +61-2-9457-9593

Argo Navis is a trademark of Wildcard Innovations Pty. Ltd. All other trademarks belong to their respective owners.
© Copyright 2006 Wildcard Innovations Pty. Ltd.