Background: The Mark IV consists of two telescopes mounted side-by-side on a single mount. Each telescope is a 5-element refractor with diameter 10 cm and focal length 400 cm. One telescope has a Bessell V-band filter, the other a Bessell I-band filter. The light from each telescope is focused on a 2048x2048 CCD, model CCD442A (made by Ford/Loral/Lockheed/BAE). The telescopes point to the same patch of sky, imaging an area roughly 4 degrees on a side simultaneously; the scale is 7.5 arcsec per pixel. In a one-minute exposure, the system measures accurately stars between seventh and fourteenth magnitude.
You may want to read a collection of Tom Droege's messages about the Mark IV hardware and software as background. Herb Johnson collected this set, which span the period Jun 5, 1997 to Feb 12, 1998.
There are several pages created by other TASS members with information specific to the Mark IV:
Show-and-Tell presentations provide lots of pictures.
Trouble-shooting the Mark IV units are stories which may help new users.
Further down on this page, you will find sample images documenting the development of the Mark IV. There is information from
There is also a a brief analysis of one of the first night-time Mark IV images, taken by Tom Droege from his rooftop on Dec 15, 1998. Another Tech Note describes a Mark IV image in Virgo, taken March 17, 1999; it includes a number of galaxies in the Virgo cluster.
Michael Richmond has tried to keep track of the various CD-Roms of Mark IV data which Tom Droege has distributed over the years. Take a look at Michael's list of the Mark IV CD-Roms.
Mike Salmon has put a copy of a whole CD-Rom's worth of Mark IV images taken by Tom Droege onto his FTP site (which is in the UK, so access might be slow for people across the Atlantic):
OK, things are almost making sense. The attached file, whatisit.gif is a 150x200 pixel piece out of a 2k x 2k exposure. The exposure was 4 minutes long at -15 C. This was (I hope I have all the light leaks covered up) a dark exposure. So what are the spots? Are these the "traps" that I have heard about. They were also visible on a 2 minute exposure, but fainter. A one minute exposure did not show the spots, though they may have been burried in the noise.
I spent the day making a box of synthetic stars. I got tired drilling little holes, and I don't have the right tools so they are not very even. It is just a box painted white inside with a little light and a black outside,white inside, metal cover with the holes drilled in it. The attached fits file has been cut down to the bare minimum so as to not use up bandwith to Tahiti, don't ask me why. The holes are drilled on a 0.125" grid. The average position should be fair, but I don't have the proper set up to make the drill land exactly on the right spot if my cheap milling machine would get it there.
OK, this is just a few pixels cut out of a 2k x 2k file. I notice that the fits header tell the history of how it was cut from the large file. Just be aware that it was in no special place in the image. Away from the edges, etc.. It is looking at an array of 7 x 7 #60 drill holes at 23 feet. A #60 drill ia about 1 mm. The lens is 135mm focal length. The pixels are 15 microns.
.040/(23x12)=1.44 e-4 radian. This is 30" of arc. So the holes are 30" of arc wide. 15/135000 = 1.11 e-4 radian. This is 23 " of arc. So the holes are 1.3 pixels wide.
Experts please take a look, and tell me if you see anything you don't like. No flat or dark field adjustments, this is just raw data. With some pain I could put on a short focal length lens (the pain is in finding a focus - takes the better part of a day) which would make the spots 1/2 pixel or so wide.
Download a FITS version of the image.
This is a view with the mark IV V camera of my light box. There is a metal plate with 0.040" holes in it on 0.125 centers. The plate covers a "white box" which is lit by a fluorescent night light. Note that the holes are not on even centers but the mean position should be pretty good as I did it on my milling machine. I just don't have the attachment that would allow accurate drilling with a small drill. Other things to know: lens 400mm focal length f/4. Pixel 15 microns. Full open aperature. Exposure about 1 second. Peak light about 1/2 full scale. Distance to light box about 20 feet.
Download a FITS version of the image.
I now have the Mark IV all set up and programmed to scan the sky. Rube Goldberg would be proud. It is sitting there on my lab bench running the following program:
1) Home the declination axis at the southern horizon 2) Move up to 42 degrees 3) Move up 3 more degrees 4) Turn on the RA drive and note the time 5) Take an exposure 6) Transfer exposure from memory to disk 7) Go to 3) unless past 46 degrees 8) Check time and rewind the RA drive at high speed to get to starting position 9) Go to 1)
I will soon have a full disk.
The longest operation is 6) where I read out the memory and write it to disk. It takes about 100 seconds to transfer 16 Mbytes. This is QBasic at work. There will be lots of improvements for programmers to make. ;^)
This is just great fun, and what I have been working a couple of years to achieve. It is getting harder and harder to find something I have not tested. But there is a lot of stuff, and I will never be sure I have it all.
The one pain is the mystery 10 second pause that something in the system inserts between writing to the serial port and being sensitive to reading from the port. It seems to appear without reason. It is sensitive to DOS version, Windows version, and all that other good stuff.
My plan is to soon get the prototype up on the roof so that I can take and distribute sample data. Then I will send out a board or two to Chris so that he can start writing code.
Once that is done, I will hope for bad weather so that I am not tempted to stay up all night taking data. Then the production line will run.
The Mark IV is taking pictures. I came home from a dinner party to see clear skies. So I turned everything on and it is now 2:30 AM and I am ready to quit. Besides, I am about to be staring at the moon.
I got a rough focus on both cameras. One (I) will have to be adjusted as I cannot quite get it to focus with my limit setting. At first the stars looks like little birds. Gulls with their wings pointing N-S. A few attemps at rotation of the whole mess narrowed down the spot size. Lots to do, but mostly everything works. There are the usual funny things to be investigated. I also need another clear night to get the focus right.
It is a good sign that the entire evening was spent taking measurements. i.e. nothing broke! And the first night of operation too.
Well, the Mark IV vibrates in RA due to the pulsed drive of the stepping motor. I have done what I can with the magic gunk, and there is still too much vibration. I see two ways to proceed.
I lean towards 2). The system is plenty good to take engineering data. Possibly even some real measurements. For a typical star see the attached IStar.FTS.
Editor's Note: click on image to download FITS file. MWR
In this image, North is to the left and West is up. The N-S is smeared out because I do not yet have a good north alignment. E-W (Up-Down in the image) is smeared out from the vibration. I suspect in the summer it will be better when the rubber is warm. I am tempted to put a heater on the rubber coupling to see what it does.
My present plan is to switch to a synchronous motor for the RA drive. This would get away from the 8/1 rewind to forward speed problem as I would expect to be able to drive a synchronous motor faster. This will take an additional electronics card and testing, and ... and would take a few months.
So should I continue the production line, get the Mark IVs into the field, and then work on a fix, or should I stop now and work till it is fixed?
I would particularly like to hear from all who are due to get a system.
Meanwhile, I will spend a few days working on other schemes. Giant dash pots come to mind. Something to take energy out of the system. It will be hard as there is a big mass on the RA axis and it is only moving of order 0.004" peak to peak. Sigh!
Back to TASS home page