sorry to read the mishap regarding the poor RasPi memory card. Don't worry too much about hitting the "correct" Pilot frequency - it doesn't matter as long as we know where it is, or the frequency word actually sent to the DDS. If you like you might just use the magic number 59055800 ( = 03851eb8 hex) to come back to the old 137499.99925 Hz.
Glitches seem rare enough not to be a big problem - I reviewed your files from the other night (Dec 10/11) and found only a single one shortly after 23 UT. Sometimes leaving the PC alone seems to help, avoiding unnecessary user interface activity.
Anyway LF conditions seem to be improving, and my favourite northern propagation indicaters (Loran-C Korea and Ussurisk) are picking up a bit of strength. Thought I might send a bit more EbNaut tonight, using same parameters as before:
137490 Hz, 2 sec, 8 char, 8K19A, starting on the hour plus about 0.3 s soundcard delay.
Still need to think up a message ;-).
Von: jcraig <[email protected]
An: rsgb_lf_group <[email protected]
Verschickt: Mi, 28 Dez 2016 7:01 pm
Betreff: Re: LF: EbNaut beaconing from JN64
Thank-you for your efforts. The pilot was generated using a raspi/ad9851
but there have been troubles with instability. Bypassing, shielding, etc
over the past few days yeielded no satisfaction and, sadly, a fried pi/SD
card! The arduino still works but the high precision math has to
be done on another machine as work continues on getting a replacement
73 Joe VO1NA
PS Stefan's coil is very compact. 0,4 mm wire? Hope it can take the kV!
PPS Paul's article greatly apprectiated. Thanks!
PPPS All the best to all in 2017
On Sun, 25 Dec 2016, Markus Vester wrote:
> Hi Joe,
> today I found the time to look into your files from Dec 19/20 (1220xx08.txt etc.). The pilot level is just right now. Assuming an accurate 10 MHz DDS-clock, the theoretical pilot frequency would be 137499.999255 Hz, consistent with the -0.8 mHz offset that I had found last time. This allows us to determine the course of your RX center frequency, which apparently drifted about 10 mHz upwards during that night (137499.8988 to 137499.9088 Hz).
> Some of the files (e.g. 0708 and 0608, see png) showed phase jumps, which are presumably due to occasional short audio dropouts in the soundcard processing. In principle it should be possible to correct these glitches a-posteriori but I have not attempted to do so. The 0308, 0408 and 0508 files looked good so I calculated the expected offsets for Ric's transmit frequency and attempted decodes. Unfortunately no useful result appeared yet.
> LF DX conditions haven't been too good lately. But I'm sure that once they improve, you will be able to pick up more than one European EbNaut station.
> Best 73,