WA5IYX FM Station Data Sources and station logged data maintenance =================================== My Very Early Years -------------------- After first getting the 1949 Stewart-Warner AM-FM-Phono set fixed in Aug 1962 so that it worked on FM I began DXing with it using an indoor dipole of twin-lead that could be oriented in several directions in the dinette using clothespins and string to attach it. With that system (set had no r.f. amplifier in it but three 10.7-MHz i.f. stages before the 6AL5 ratio detector) all that I got was in-state tropo, the best of that being Houston (c. 200 miles) with some hints of Dallas. Why I never connected the 6-element hi-low VHF TV (1955) Yagi (14' AGL and fixed-aimed south east) back then is beyond me. That had then already been used for three summer seasons of Es TV-Dxing. My only source of US FM station data came from the Whites Radio Log included in Radio TV Experimenter issues. Raw logs were kept on various loose-leaf notebook paper, later compiled into tyewritten log sheets. A few reception reports were mailed out, with some being answered by QSL card or, more often, by letter. Despite numerous attempts to get Es in 1965-1969 when evidence of it was seen on Ch 6 that system never did until July 1970 when, amongst many unIDs, 93.7 from Minneapolis was nailed. Modern Equipment ---------------- A more modern FM set, a Wards Airline GEN-1451A, was gotten that October and my FM DXing really took off. My practice of keeping the raw FM-DX log on legal pads (first of various sizes & colors, now uniformly yellow) came into effect within the next couple of years. It provided an easy way to spot any sheets that had become detatched from the pad (and often the log sheet for any sudden 144-MHz Es event was simply one ripped from the pad).** It was 1972 though before I got my first of many issues of Bruce Elving's FM Atlas. I then proceeded to underline (in red ink) the stations that I'd logged on the maps and the station listings therein. Updating the data in them was done in pencil from the Elving monthly column in the VUD, soon making them a sloppy mess if it was too long before a new issue of it came out. And, when a new issue came out also came the arduous and time-consuming ritual or underlining all the logged stations again. The maps were especially onerous to do as stations often overlapped onto more than one map. I had copies of a receiption report form offset printed to expedite my sending of those. (Something that I had also done, with the cumbersome gelatin stencils, for my AM BCB DXing ten years before.) Those reports were regulalry sent off to new stations logged until 1974-75. One snooty chief engineer at a 104.9 in Wisconsin once replied that reports had to be sent in within 24 hours of a logging in order to be considered for verification - WTF ? In mid 1972 I also started to maintain a 3x5 index card file (typewritten) on the stations logged and relogged (eventually setting a minimum of 250 miles on relogs to avoid the Houston/DFW mass of stations). These were initially organized by frequency and call. Much later, due to the vast ongoing changes in calls, it was re-sorted by frequency, state, and city of license. I eventually had to build a large box to hold all those cards (giving up on the smaller metal ones, which kept growing in number). Managing those cards - just aligning them in a typewritter was such a chore itself that I would often wait until the Es season was over with to tend to that - became just too much and was halted c. 1997.* Enter the Computer ------------------ An Atari 800 (8-bit system with 64k RAM) was obtained in Nov 1983. After I got a Word Processer Cartridge and a printer for it in Aug 1984 I started to compile my TV/FM reports to the VUD using those. It wasn't until Nov 1984 that I got a floppy disk drive (5.25") to replace the tape recorder mechanism that I'd suffered with for a year. In Feb 1985 I got a copy of Synapse's Synfile+ database manager for it and began to enter my first-log FM station data into it (by that time having over 1500 via Es alone). I had that on three floppies - each sorted by frequency then by state-city, by mileages, and then by propagation modes. Since each data base used 4 files in it, and the maximum number of files on one side of an Atari floppy under their DOS 2.5 was 64, it was a constant stuggle to divide up the 100 frequency chunks into ways that would permit their db files to fit (by size AND number) onto each side of a floppy. Adding new stations to those db floppies was often arduous, since after so many new ones were added it would proceed with a time-consuming reindex of that particular database (four files). Fragmentation over time on the floppies would also slow the process down. Doing just a COPY file of the db files to a new floppy wouldn't work as the software used index pointers to precise disk sectors that would no longer be valid if that was done. Instead a copy database had to be used with much floppy swapping back and forth (as only 64k of RAM). Quantum Leap in Mileage Determinations -------------------------------------- In the summer of 1992 Dajja Enterprises announced a hardcopy set of databases that contained the coordinates of US TV/FM transmitters (TLs). Before that I'd been using a Feb 1968 National Georgraphics US Map (mounted on a large piece of cardboard) to determine distances (to the city of license, which was often tens of miles from the TL). I used rulers with 0.1" marks on their 72 miles/inch scale map, rounding off to the nearest 5 miles. I was actually just on my second such map, as the first had been much frayed. But, I could now dispense with all of that and calculate very accurate mileages (rounding off to the mile). The hard copy was used to enter the coordinates of my FM stations logged (by then well beyond 2000 by Es alone) into a BASIC program designed by K5KS to calculate distances (and bearings) and modified by me to print out the mileages (one at a time) on the bottom of all those 3x5 index cards. Computer Upgrade ---------------- The next month, Aug 1992, I "upgraded" to a Packard-Bell 386-SX-16 (85-M HD, 1M RAM, Win 3.1). Their included database in Lotus Works was totally unsuited to handle my FM log station databse as it would only sort on one field. Eventually, PC File 6.5+ (a DOS-based program) was settled on. However, the anticipated transfer of DIFs from the Atari didn't work out (connecting the modems from each at 1200 bps and using ATA commands on each's terminal software was very iffy most of the time). I even arranged for a while for me to upload DIFs to a local BBS with the Atari and download them with the 386. I finally had to resort to re-entering all that first-log FM data by hand into the PC File 6.5+ dbase. Later on Dajja offered their data set on floppies. But, that came in one huge chunk (too large for DOS EDIT to handle) so had to be split into arbitrarily-equal sizes, then cut-n-pasted into 100 files (one for each frequency). Those files then had to be adjusted so that the coordinates would be on the right spot on a line in a text file for a QBASIC program that I'd written (based on the earlier K5KS one) to read them and batch process the mileages out to another text file. It litterally took weeks to just set up those 100 files to be properly read. By that time I'd also gotten bd.exe by W9IP, which reportedly was very accurate (for very long distances taking into account that the earth is not a sphere), but it had no way to batch process inputs, and I was totally incapable of modifying it (DOS-based) so that it could. Getting Really Online --------------------- I'd first used modems (300 bps, soon 1200 bps) with the Atari in Feb 1989 to dial up various BBS. Then with the 386 (and later machines) it'd been 2400 bps to 14k kps. I finally got with an ISP (SWBell) in late January 1997, and that opened up a whole new world of sources for FM station information. (It also meant that I ended up spending more time trying to track down all those unIDs from pieces of program bits than I ever did or even could have done before.) Dial-up was finally left behind for the current DSL in Jun 2002. ========================================================================================================= * Sometime c. early 2005, I transcribed ALL of the data from those index cards (and subsequent logs) into text files, one for each frequency - producing these (sample) from which station counts by propagation modes could be easily seen (and double checked). Those files are updated at the end of the year (using a PC File 6.5+ output text file sorted on freq, state, city - from an LD input file made using my custom QBASIC program on my finalized FM-DX text log for that year). ========================================================================================================== ** For the past several years I've only used one side of a sheet since ink bleedthru had made some of the older logs very difficult to read now. The full station call, frequency, city, state are added on later review (once in red ink for new, now in blue pencil for all). The later program detail info gleaned from the C-90 tape review are added in No. 3 pencil, as is the tape counter index (in parentheses) with an asterisk denoting a likely candidate for making a digital audio clip of later on. The different colored writing enables one to see at a glance what was the original raw data and what later was added upon tape review. Until late 1979 almost all of my FM-DX tapes were done on a 1968 reel-to-reel (1-7/8 ips) Radio Shack Model xxx machine (originally gotten to use with my Russian language course at UT Austin that spring). Over the years its speed slowed down (unbeknownst to me) so that any digital dubs now made from those has to have Wavepad do adjustments for that which vary with the years involved. So accurate tape counter indexing is only mostly available for the cassettes made after late 1979, and those counters seem to vary by machine (of which there were MANY) for some reason(s). Tape counters on r-to-r machines are affected by how much leader tape may or may not be used on the ends of the tape in a given reel. With the advent of rapid RDS/HD idents on FM (now noted in the log by a circled R or H) there is far much less use/reliance of/on the C-90 tapes for deducing idents. Often, the memory or direct frequency buttons are just quickly pushed if a given signal is not decoding or obviously near some ident or program content (local ad, PSA, weather, news) that would easily lead to an ident. Thus my collection of C-90s to review now are but a fraction of what they were just a few years ago. The old maxim of "tune for talk" in FM DXing no longer applies since the arrival of RDS/HD and online playlists and live streaming of stations. ======================================================================================================