Operating under marginal viewing conditions: ============================================ When in doubt, record, and make copious notes about viewing conditions, in terms of both cloud and surface air clarity. To gauge the latter, I have found it useful to go outside, let my eyes dark-adapt for a few minutes, and then see how much light is scattered out of a vertical flashlight beam, while shielding my eyes from the direct light. It's qualitative, but quite helpful. Nightly routine operations made easy: ===================================== I have developed some tools to make our lives here a little easier. You'll encounter them in the following order. - EDS.BAT: I have used the practice of naming each night's log files according to the order of creation during the night, and the date. Thus the first log file on Feb. 20th would be KI_LOG1.220, and likewise the first temperature log file would be KI_TEMP1.220. I have done this by modifying the startup file KI_START.RL prior to each run of KI. To make this easier I wrote the above-named batch file (EDS stands for EDit Startup file), which accepts a single argument giving the location of the observations (Rabbit Lake = RL), and opens the startup file KI_START.%1 for modification. I told it to use VI, but you may want to change that to EMACS (if you decide to use it at all). You don't have to be in the subdirectory containing the startup files to call the batch program. - ASCIZE.BAT: KI produces temperature logs in binary format. Bernie wrote a utility called TEMP.EXE to generate a formatted ASCII list of the temperatures. ASCIZE.BAT makes it a little faster to create the ASCII list. Tell it the sequence number of the temperature log (like 1 above), and the date of the log (like 220 above), and it'll create the ASCII list for you, calling it ASCTEMP1.220 in the present circumstances. - CENSUS.BAT: This batch program generates a list of the image and log files created on a given night. For example, at the end of observa- tions on the night of the 20th, type CENSUS 20 to create a file C:\IMAGES\CENSUS.20 listing the number of files in each rele- vant subdirectory of C:\IMAGES that begins ?20?????.???, i.e., that was created on the 20th. Oh, yes, it also looks for all files in C:\USER\BERNIE\INTERP\LOG that end .?20. - CLEANUP.BAT: After backing up all the image and log files from a night, type CLEANUP 20 (e.g.) to delete them from the hard disk. This program goes through much like CENSUS does, only it deletes any files that match its template. I hope you find these batch files useful. I have. Oh, BTW, I've formatted the next DAT tape, RL9305. I'll take RL9303 and RL9304 back with me. Using the ephemeris program: ============================ I have generated lists of solar and lunar coordinates at 15-minute intervals throughout each of the nights we're here, to make it easier to find out when observations can take place. The files are named yymmdd.SUM (e.g., for this UT day, 930220.SUM), and are stored in \USER\BERNIE\EPH. The line starting S: gives the solar coordinates, and the M: line is for the Moon. I have also created a batch file (EPRL.BAT) to run the ephemeris program for the coordinates of Rabbit Lake. Without arguments it generates the output for the date/time given by the system clock. You can specify a date (format -d YYMMDD), time (-t HHMMSS), or other more esoteric things like atmospheric pressure, if you want to get fancier. The batch file is in the defined PATH so it can be used as a system call from within KI, where e.g. a DOSKEY macro could not be. Speaking of the system clock, it appears (both from the documentation, and from some experimenting I did) that the TIME command does in fact adjust the BIOS clock, contrary to the discussion we had in Calgary a while back. This is a sensible thing, in my view. Calibrating: ============ I estimated from your notes that the PLBS began to dim appreciably after about 105 minutes operation with a set of NiCd batteries. Since arriving I have ca- librated 7 times (so far), and the PLBS has been on for a total of 102 minutes. I borrowed a voltmeter (from the electronics technician, whose office is at B-Zone, north of here (check the map)), and found that the total battery voltage after 88 minutes operation was 7.62 VDC, i.e., perfectly nominal for these batteries. I had to return the voltmeter before the last calibration so I don't know the present, but there seems to be a discrepancy between what you experienced with it and its present behaviour. Update: as of 930219, 0850 UT, the PLBS has been on for 116 min, and according to the Micronta mini-DVM that belongs in the SkySwitch, the battery voltage is 7.59 V. I don't know how reliable the little mini-DVM is, but it looks like the batteries still have some juice left. If the Micronta is reasonably good, perhaps we should get one for the dedicated use of the Polar Camera project? I don't have a feel for DVM/DMM prices, but I'm guessing we could get the Mi- cronta for an order of magnitude less than a good Fluke, right? Boiling water and blowing breakers: =================================== I blew a breaker while boiling water, and finally had to go over to the hut by the navigation beacon to reset it. In the process I missed a splendid UARS overflight! :-( The lesson is that the kettle puts us perilously close to the limit of the available power. I have adopted the practice of turning off the circuit powering the truck block heater, and turning off the inside lights, while boiling water. So far, so good, although it's too soon to tell. I should mention that when I tried to restart KI after the power failure, I had to cycle the clean AC about 10 times before AFB 0 responded. I surmise that the loss of all power (including heat) to the camera allowed it to cool down to a point where the RS-232 connection became unreliable. I didn't think to note the inside temperatures after getting KI going, though, so the above is not certain. VIEW.C: ======= I decided to attempt modifications to VIEW.C to rearrange the temperature labels, so they would be closer to their actual temperature order. I was successful in modifying both VIEW.C and PMTEXT.H. The temperatures are now displayed in the following order (top to bottom): tfw tei td tee tcf. I suppose I could have tackled the hard-wired "AIRGLOW" label, but I didn't. Emboldened by this success, I modified the way KI opens its command and temperature log files. I modified the "open" call for the temperature log to have the attribute O_APPEND rather than O_CREAT, and modified the "fopen" call for the command log to have the argument "a" rather than "w" (append rather than write). The modified version is KI.(C,EXE), and the unmodified version is OLDKI.(C,EXE). Now the trouble is, KI doesn't know how to open the log files if they're not already there. I had to run OLDKI first to create them, and then KI to append to them. I did some snooping in the very limited C documentation I had, and concluded that the "access" function could be used to decide whether to open new log files or append to existing ones. My boldness was not sufficient to induce me to tackle implementing "access", so it's up to you. Script hiccups: =============== I have noticed that sometimes during the repeated execution of a script like WFAURORA or WAURORA, the interior loop through the desired filters (e.g., WFAUROR1) will get interrupted in mid-stream, and a new loop will begin with filters 0/0. This can be detected either inside, from the sequence of images displayed, or outside, from the unexpected pattern of filter wheel motions. I haven't been able to figure out why this is happening. In itself it's more a nuisance than anything else, but to the extent that things can happen in KI that I don't understand, it worries me. See if you notice the same sort of thing, and if you can make any sense of it. Cables: ======= On arriving I verified operation of the cameras, and then tried to lift the cables out of the snow where we could see them. I found that they are frozen to the ground just south of the cameras. They had melting weather here while we were out, and some water must have collected in a low spot. It will take some time and care to extricate the cables. Probably breaking up the neighbouring ice with a hammer and screwdriver would be satisfactory, and the frozen parts could be brought inside the building to thaw out before packing up. Packing up: =========== On reviewing the schedule of satellite overpasses and moonrises, I've concluded that you'll probably have to write off the last night here, and spend the time getting the equipment packed up and back to the warehouse. You'll probably want to think about how to schedule this activity, and make arrangements ahead of time for someone to come out with a loader to help get the crates on the flatdeck. There is a small service crew on at nights, so you wouldn't have to wait until morning to load up and take the crates to the warehouse. I've been waiting for the N2 cylinders to arrive from Saskatoon, but they're still not here, and now won't arrive before you do. Keep asking at the warehouse for them. Of course, with each day that passes now there is less chance that we'll need them, so you might as well leave them at the warehouse unless you actually need one. Once the stuff is at the warehouse, I *believe* you only need to ask them to send the stuff back to Saskatoon. Everything should still have the address labels from the trip here, and you can use the permanent markers to change the "From" and "To" designations. I've given Ray Fournier, who's in charge of the warehouse in Saskatoon, all the information he asked for, so the transfer from Cameco/Saskatoon to Canadian Freightways/Saskatoon should happen smoothly, and then CF should have the stuff back to us within a day or so after that. I expect we should have the stuff back by Monday the 8th. I'll call Canadian Freightways while I'm in Saskatoon, to make sure they know what they're getting, and how to handle it. Printer: ======== I couldn't get the printer to work properly. There were two areas of trouble: the print quality was poor, possibly because the ink cartridge is toast, and the sheet feeder couldn't keep the printer happy - I kept getting "paper" errors. If you've brought the docs on the sheet feeder, maybe you can get it to work. If not, it's no big deal, unless you had planned to use the printer. After I wrote the above, I tried to change the printer cartridge. On beginning to do this I finally noticed that what I brought up with me on my first trip was not a replacement cartridge but a refill kit! I faithfully followed the directions provided but just made a mess, so you won't be able to use the printer (unless you are able to succeed where I failed - feel free to try!). From now on let's just buy replacement cartridges. Reaching me in Saskatoon: ========================= I'll be in the Physics Department at U. of S. from Monday to Thursday of next week. I can be reached either by phone (call Pat Wilson at 966-6403 and leave a message) or fax (966-6400, same as Nick Rolheiser's #). From Friday the 26th till Monday March 1st I'll be incommunicado, so plan to have any emergencies before then, OK? :-)