The Meade 16" LX200R
                                                                                    
(click on images to enlarge)

After I finished building the observatory building and dome, for a number of years I used my 12.5” Dall-Kirkham. However, the RA and DEC drives, and telescope control software (none), was completely inadequate to do serious astronomy. It was a good telescope in its time, the mid 1970’s, but not now. In an attempt to bring it up to modern standards I purchased a Paramount. However, the positioning of the block pier in the observatory was designed for a fork mount. I attempted to make the Paramount German equatorial mount work by having fabricated a heavy rectangular steel platform that jutted out from the top of the block pier. I subsequently, after a lot of work, that this was nutty. Luckily, I managed to find a buyer in Australia for the Paramount. It had never been used. Software-Bisque was very nice in transferring a new Paramount warranty to the buyer. This all took place around 2004-2005. At this time, Jeff Hopkins, a renowned photometrist, suggested I buy the Meade 16”. As it turned out I had the money because I had just sold my Phoenix home in 2004. I am partial to fork mounts and so I bought the 16” Meade LX200R through Starizona. Starizona has been very good to me over the years. The 16” saw first light on February 17th, 2006. I am very happy with it. The optics are very good. The images through the eyepiece are just as good as through my 12.5” Dall-Kirkham and that is optically a good telescope. To date, the TCS appears like it will do everything I want it to. I believe I have a good telescope, contrary to its bad reputation, because I believe Scott Roberts, who now owns Explore Scientific, at the time worked for Meade and made sure I got a good scope. On the other hand, maybe Meade years ago always made good telescopes, with a very few exceptions that were sensationalized.
The permanent equatorial pier from Meade is very well built. The diameter of the pier is 10” and is made of ¼” thick steel. Top plate and bottom base are of ½” thick steel. Because my observatory already had a 32” square block pier I could not mount the Meade pier in the usual manner, the usual manner being bolting it to a concrete floor. Consequently, I had to come up with a different mounting method, which resulted in a much better method for polar aligning the telescope. The standard procedure for polar alignment is to apply brute force to move the pier in the azimuth direction and using shims where it bolts to a concrete floor for the altitude adjustment. This cannot be done on my block pier because the pier has a hollow central square hole formed by the six cinder blocks of the pier, and all the cinder block holes were grouted, each terminating with a ¾” J-bolt: there are twelve block holes and so there are twelve J-bolts at the top of the pier. The round base of Meade’s steel pier is 24” in diameter. To fit the Meade steel pier within the perimeter of these J-bolts, Meade turned down the diameter to 22.5”, at no extra charge.


These twelve J-bolts turned out to be an advantage in providing a good way to make an altitude adjustment. I bought a 1” thick steel plate from a salvage yard and had a machinist drill twelve 1” diameter holes to match the location of the twelve J-bolts. The Meade metal pier is bolted to the 1” thick steel plate. The steel plate rests on twelve nuts, one for each J-bolt. Before the nuts were screwed on the J-bolts I brushed on anti-seize compound over the J-bolt threads underneath the 1” thick steel plate.. Adjusting the nuts under the steel plate is how the altitude adjustment is made. Jeff Hopkins is in the image.
The azimuth adjustment is done by rotating the steel pier within the rotational slots in the base. The usual way to make this adjustment is to use brute force to rotate it. I did not like this and so I came up with a different method that does not require brute force and is very accurate. First, I drilled a ½” hole in the center of the base, and then put in a ½” diameter stud in the center of the 1” thick steel plate. Thus, the rotational axis of the pier is confined to the pivot point of the ½” diameter stud. Next, I bolted what I call a 1”square steel tangent arm to the top of the base of the pier. This is seen in the above image. You can also see the brackets I made (out of a ½” thick flat iron that I had to heat with a multi flame heating nozzle to be able to bend it 90 degrees) and the 3/8” set screws that push on the tangent arm.



How did I get the Meade Steel pier, which weighs about 225 lbs, and the OTA inside the observatory and on the block pier? In [INSERT IMAGE] You can Jeff Hopkins holding a frame I welded together out of 1/16” thick 2” square steel tubing that does this. The pier is hoisted through an opening in the deck. Once hoisted above the deck, the deck planks are put back in place and then the pier is lowered on them. Six inch diameter wheels are then added to the frame and this is how the pier (and OTA) is moved inside the observatory and erected. Jeff Hopkins is in the image.

Home Page