New roll, no luck, and some insight into the process

Over the weekend, I attempted to work the bird with only a handheld with a tiny rubber duck antenna, while on location in NH.  Unfortunately, the radio just disconnected from the laptop every time I attempted to transmit.  The handheld really needs a remote antenna or that happens on 5 watts.  I forgot my portable yagi, so I thought I'd give it a try.  I was able to remember all the commands I needed to type in every time, so that was good, but alas it didn't work.  52 degree pass.  Would have been neat if it did work.

Upon return yesterday I decided it was time to shoot a "new" roll since the satellite had had a few days of rest to fill it's battery.  I cleared the file counter and issued the commandle to mode 3.  I had gotten back late so I wasn't able to work it at the shack, had to do remote, and I lost elevation before I could confirm it accepted the command.

Today, I confirmed the file count and began the file size survey on the first pass, getting about 24 file sizes before the pass was over.  After entering those counts and identifying which files might be interesting, I was ready for the next pass (51 deg).  On this pass I retrieved the remaining files sizes that were "type-able", I began download of files 0x25 and 0x43.  I was missing block 2 from the 0x43 file so the file was too corrupt to view.  I remembered that the Kenwood seemed to have good signal at the beginning of the pass, so I checked the log and found it had the missing block.  I pasted it into a hex editor and saved it in the file-pit-build directory, ran the stitcher, and had a nice preview of a sunset picture of clouds.  The file was still missing several of the blocks towards the end.  I also previewed the other much larger file, 0x25, and see that it looked like an ocean/clouds picture...but this one is relatively very large (195 blocks) and missing quite a few blocks.

After entering the file sizes I retrieved and re-prioritizing the download schedule, I made a pass plan for the third pass.  Since the front part of 0x43 on was downloaded, I decided to start downloading it on the front end of the pass so that when it got towards the end, the pass would be at a better elevation and I'd have a good chance to catch the blocks toward the end of the file.  Then I dropped to KISS and retrieved the file sizes of the 4 "un-type-able" pictures, then dropped back to command and downloaded 0x25 some more.  This brought in about 25 files for 0x25 and left one file missing in 0x43.  I still need about 30 blocks (of 195) of 0x25 to complete the 0x25 picture.  I am only missing block 117 for 0x43, so I wrote a KISS program which I hope to run on tomorrow's pass and complete that file.  I copied the files I had additionally retrieved to the file-pit-build area and ran the stitcher on both pictures.  Here's what 0x25 looks like now, after 2 passes worth of work:



The stitcher output looked like this:
MISSING:
28,29,30,31,32,33,34,35,37,94,96,97,98,99,100,101,102,103,159,160,161,162,163,164,
165,166,167,168,169,170,171,187,188,194,195,
Total missing files 35, out of 195, 161 retrieved.

Since the first missing block is 28, you can see in the picture where the error starts to begin.  Tomorrow I will attempt to download this again, since there are too many missing blocks for me to retrieve if I dropped into KISS mode and tried to individually download each one.  The tradeoff is that 195 blocks will take most of the pass (at least all the quality part of the pass) to retrieve them individually.  Hope for lucky passes tomorrow (Tuesday 8/5).