This page contains pictures of my version of James Lerch's automated Foucault tester. Visit his site at the link provided at the end of this paragraph.
The first one I made used a high resolution, low lux, black and white CCD camera with a 5 to 50 mm zoom lens. Despite many problems I had to over come in its construction it was a failure. The camera for some inexplicable reason clipped off the left and right sides of the reflected light from my light source. A vignetting effect of some sort. Since the area the software analyzed is along the x axis of the mirror it effectively removed zone five from analysis. I will not mention the financial cost to yours truly. I started over from scratch and used James proven method of using a Cam Corder. Both his camera and mine are Sharp cameras of similar make and model. My unit uses the head moving mechanism from a SyQuest model SQ 312RD. A 10 megabyte removable platter hard drive of late 1980's vintage, to move the knife edge in and out of the light cone. The units used a linear guide to move the heads and was fairly easy to convert and mount to the carriage assembly. Basically I cut out the part of the drives chassis to which the motor and linear guide were attached and mounted it to the front of the testers chassis. Such a deal! The first image is the KE assembly. The second shows the KE blade and the slitless opening for the light source. The third and forth image are of the LED selector guide. Like James I wanted to experiment using different light sources. I have a red, white, blue, and green LED for a light source. The LED assembly was made from the head guide mechanism from a 5 1/4 floppy drive. The LED's are soldered to a pcb which is attached by two screws to the floppy drive assembly that moved the heads across the platter. Two quad LED printed circuit board holders are stacked together over the bodies of the four LED's. The device was a bit touchy to make as the slide has to have enough friction between the wood mount and the plastic guide to keep the LED in position but not so much to impede ease of movement. Drilling holes in the two wood pieces to exactly match the distance between the two guide rods was an interesting exercise. For those who have no such interest stick to one LED. It is much easier on the nervous system. I apologize for the out of focus images. The zoom function of my digital camera is less than satisfactory. Especially with me behind the lens. I am no photographer.
Links to all the software mentioned on this page will be at the bottom of the page with the exception of RTAFT. The software is not yet available for public use as it is under development. If you are interested in Beta testing please contact James from his web site.
|The following photos are of the upper and lower carriage assembly. Care should be taken in the construction of the guides for the threaded rod. Most important is their proper alignment with one another and the motor. Any misalignment can cause excessive binding which can stall the motor or effect step distance adversely which will have considerable impact on your test results. Careful as I was I still had to shim with thin pieces of plastic on two of the guides. The center guide is a 1/4-28 threaded coupling press fit into a 3/8 inch hole. Softwood is preferable and drill your hole perpendicular to the wood grain to minimize splitting.The coupling I used was hex shaped and required no glue to hold in place. The guides were made from 1" x 1" wood squares glued together then cut together to the proper length. The easiest way to insure alignment is to thread a rod through the threaded guide. Place an unthreaded guide on each end. Use a clamp to hold them in position and cut of the excess on a table saw. The first and fifth guide have a top hat shaped 5/16 OD and 1/4" ID nylon sleave inserted inside them. The second and forth guide have a 1" long by 3/8" OD by 1/4" ID nylon standoff mounted inside them.|
|The following images are of the printed circuit board for the LED light source. It consist of a power switch, LED selector switch and a current source control for light density. The LED's are powered by a 9 volt battery that has current limiting resistors commensurate with the current and voltage requirement for each LED. The last image is the Red laser module. The laser is used to align the light source with the camera lens. It uses one three volt coin battery for power.|
|Both motors in the following images are 400 step bipolar motors. I used them because such motors have more torque for their size and in my first design I took the easy route and chose muscle over finesse. My first platform motor was a unipolar motor and had to over come to much resistance in the mechanical threaded rod assembly. It would stall out badly. The bipolar replacement did the job but barely. I had to do considerable reworking of the platform before the binding issue was reduced to an acceptable level. The platform you see in these pictures uses bipolar motors but since my present design turned out so well a unipolar would work fine. However the KE motor was a bipolar from the hard drive and my driver circuit is made to drive bipolar motors so there was no reason to use unipolar motors. The message I am trying to convey here is, if unipolar motors is what you have or can obtain, by all means try using them. James uses that type motor in his test platform to good effect.|
|The following images show the cradle to keep the camera in place and the front and back of the face of the upper platform assembly where the KE motor is mounted. The nine volt battery is the power source for the LED's.|
|The following images are of the completed tester although I removed the motor driver so it would not block the view of the back of the camera in a couple of the images.|
|These images are of the bipolar stepper motor controller. It is essentially Mel Bartels Alt/Az circuit for controlling a Dobson mounted telescope via a computer running his software. Block diagram wise it is the same circuit that was designed around a isolated computer bus using an isolated five volt power source, TTL eight bit wide buffer IC, and opto coupler/isolator's to keep the drivers separated from the computer parallel port. Instead of power bipolar transistors, or power mosfets, it uses two L298 motor control IC's from ST-Micro Inc. Control of the motors is via RTAFT by James Lerch.|
|Below are images from my first test of a F/4.5 mirror of 6.5 inches diameter made from Pyrex glass. At this stage no figuring has been done. It is fresh of the grinding machine after polishing out. The first set of images are from David Rowe's Figure45 program for reducing the test data. The biggest disadvantage I have found using Figure 45 is the lack of a method to make hard copy of the test results using a Windows 9X based OS and Windows based printers that have no DOS printer drivers. As a result I have been forced to take pictures of the images with my digital camera from my monitor screen. All an all not to bad but a far cry from a good JPEG or GIF image. One of the nice features of RTAFT is you can point to the location of these two programs and it will start them up once RTAFT has completed testing the mirror.|
|The next group of images are from the Windows version of Sixtest by Jim Burrows.|
|The next group of images were captured by RTAFT and shows the mirror shadows as seen in each of the five zones.|