Return-Path: Received: from mtain-me09.r1000.mx.aol.com (mtain-me09.r1000.mx.aol.com [172.29.96.145]) by air-de01.mail.aol.com (v129.4) with ESMTP id MAILINDE011-5ea24cbafe25211; Sun, 17 Oct 2010 09:46:13 -0400 Received: from post.thorcom.com (post.thorcom.com [195.171.43.25]) by mtain-me09.r1000.mx.aol.com (Internet Inbound) with ESMTP id D23CC380000EE; Sun, 17 Oct 2010 09:46:09 -0400 (EDT) Received: from majordom by post.thorcom.com with local (Exim 4.14) id 1P7TXW-0007cO-9a for rs_out_1@blacksheep.org; Sun, 17 Oct 2010 14:44:42 +0100 Received: from [195.171.43.32] (helo=relay1.thorcom.net) by post.thorcom.com with esmtp (Exim 4.14) id 1P7TXV-0007cF-Dw for rsgb_lf_group@blacksheep.org; Sun, 17 Oct 2010 14:44:41 +0100 Received: from imr-ma04.mx.aol.com ([64.12.206.42]) by relay1.thorcom.net with esmtp (Exim 4.63) (envelope-from ) id 1P7TXS-0001iZ-LB for rsgb_lf_group@blacksheep.org; Sun, 17 Oct 2010 14:44:41 +0100 Received: from mtaout-da04.r1000.mx.aol.com (mtaout-da04.r1000.mx.aol.com [172.29.51.132]) by imr-ma04.mx.aol.com (8.14.1/8.14.1) with ESMTP id o9HDiOes012274 for ; Sun, 17 Oct 2010 09:44:25 -0400 Received: from Black (nrbg-4d073bb1.pool.mediaWays.net [77.7.59.177]) by mtaout-da04.r1000.mx.aol.com (MUA/Third Party Client Interface) with ESMTPA id C7858E0000CA for ; Sun, 17 Oct 2010 09:44:24 -0400 (EDT) Message-ID: From: "Markus Vester" To: References: <8CD38567F4B0C6A-1C1C-116D@webmail-m021.sysops.aol.com> <2C4BE56C4BAB4EB984EBC865A2BBC338@Black> <4CBAE155.6030500@freenet.de> In-Reply-To: <4CBAE155.6030500@freenet.de> Date: Sun, 17 Oct 2010 15:44:07 +0200 MIME-Version: 1.0 X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Windows Mail 6.0.6000.16480 X-MimeOLE: Produced By Microsoft MimeOLE V6.0.6000.16669 X-Spam-Score: 0.0 (/) X-Spam-Report: autolearn=disabled,HTML_MESSAGE=0.001 Subject: Re: VLF: Re: Detections of 5 microwatt transmission Content-Type: multipart/alternative; boundary="----=_NextPart_000_000E_01CB6E12.2221D3A0" X-Spam-Checker-Version: SpamAssassin 2.63 (2004-01-11) on post.thorcom.com X-Spam-Level: * X-Spam-Status: No, hits=1.0 required=5.0 tests=HTML_30_40, HTML_FONTCOLOR_UNKNOWN,HTML_MESSAGE,MISSING_OUTLOOK_NAME autolearn=no version=2.63 X-SA-Exim-Scanned: Yes Sender: owner-rsgb_lf_group@blacksheep.org Precedence: bulk Reply-To: rsgb_lf_group@blacksheep.org X-Listname: rsgb_lf_group X-SA-Exim-Rcpt-To: rs_out_1@blacksheep.org X-SA-Exim-Scanned: No; SAEximRunCond expanded to false x-aol-global-disposition: G x-aol-sid: 3039ac1d60914cbafe2112e9 X-AOL-IP: 195.171.43.25 X-AOL-SPF: domain : blacksheep.org SPF : none ------=_NextPart_000_000E_01CB6E12.2221D3A0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable Hi Wolf, Paul was using his own dedicated receive software, with timestamps bas= ed on GPS 1pps pulses.=20 For transmitting, I employed SpecLab's signal generator, with samplera= te tracking to 10 kHz from a Conexant Jupiter GPS module. For receivin= g, I normally lock to 19.58 kHz, which seems to give good stability (e= xcept for a minor glitch when the skywave dips before sunrise).=20 As far as I have seen, the UK MSK signals (19.58 kHz, 22.1 kHz), the= Americans (24 kHz), and NWC (19.8 kHz, but too weak) seem to be phase= coherent to GPS. DHO (23.4 kHz) is stable but about 2.7 ppb high. The= French (HWU) and Turkish (26.7 kHz) are drifting and unusable. The DK7FC grabber is also based on GPS 10 kHz, injected into the recei= ve audio. Best 73, Markus (DF6NM) ----- Original Message -----=20 From: wolf_dl4yhf=20 To: rsgb_lf_group@blacksheep.org=20 Sent: Sunday, October 17, 2010 1:43 PM Subject: Re: VLF: Re: Detections of 5 microwatt transmission Congrats Markus and Paul for this achievement ! I think you used a GPSDO with 10 kHz output on both ends - Thunderbo= lt, Trimble, or similar ? (still thinking about a simple but reliable, non-GPS solution using= a 24/7 active MSK station, or maybe DCF77 / MSF with a soundcard runn= ing at 192 kSamples, or SDR-IQ running at 196 kSamples ). 73 and have a nice Sunday everyone, Wolf DL4YHF . Am 17.10.2010 12:56, schrieb Markus Vester:=20 Dear LF, I am very happy to announce that Paul Nicholson in Todmorden was= clearly able to detect each of my three test transmissions. His detai= led and most interesting report is at http://abelian.org/vlf/mv101009/ Hats off to Paul for his outstanding achievement! Best regards, Markus ----- Original Message -----=20 From: Markus Vester=20 To: rsgb_lf_group@blacksheep.org=20 Sent: Tuesday, October 12, 2010 8:57 PM Subject: VLF: Detections of 5 microwatt transmission Dear LF, for my VLF test transmissions on the last weekend (Oct 9: 8969.9= 98 Hz, Oct. 10: 8989.997 Hz), I have received reports from three recei= ving stations: - Walter DJ2LF near Erlangen (20.2 km) received the carrier agai= n in good quality, using 0.95 mHz resolution. Radiated power and recei= ved SNR were quite similar to our two-way QSO on June 4th. - Stefan DK7FC in Heidelberg (178.5 km) reported about 10 dB SNR= , and both dashes are still visible in the QRN minima on his 47uHz gra= bber window=20 http://www.iup.uni-heidelberg.de/schaefer_vlf/DK7FC_VLF_Grabber.= html This was the intended purpose of the experiment, and a nice coun= terpart to the earlier detection of a 200 uW kite transmission on Aug= 29th. Even though both of us were locking the samplerate to a 10 kHz= GPS-derived reference, the dashes appeared about two pixels low - per= haps due to a very minute rounding error in SpecLab's frequency scale= display. - To my utter surprise, Paul Nicholson (Todmorden, 1030.5 km) pr= oduced two spectra, taken over the duration of the transmissions: http://abelian.org/vlf/df6nm/2010_10_09a.gif (9:50 - 18:00,= 34 uHz) http://abelian.org/vlf/df6nm/2010_10_10a.gif (9:00 - 15:00,= 46 uHz) After taking a deep breath, we now have to deal with the questio= n whether this is a significant positive detection. Except for a known= central artifact on 8970, the highest peak appears in the correct fre= quency bin in both spectra. Naively, one could then propose that the= probability of this happening at random would simply be the inverse= of the number of displayed bins, ie. around 1:230 for Saturday and 1:= 170 for Sunday. Thus the combined probability of a false positive dete= ction on both days would seem to be only 1 in 40000. Certainly there= is a degree of arbitrariness in the choice of the display range (8 mH= z). Paul estimated that the signal was about 3 standard deviations= on Saturday (0.3% false detection rate), and 2 sigma in the higher no= ise on Sunday (5%), giving a combined false positive probability of 1= in 6667. We can also look at the plausibility of the absolute fieldstreng= th of the peaks (about 0.2 fT, equivalent to 0.06 uV/m). If I remember= correctly, Paul's first detection of Stefan's kite signal on March 15= was at about 3 fT, and Stefan was then radiating approx. 1 mW EMRP.= Scaling this down to my estimated 5 uW EMRP, and taking another dB fo= r the slightly higher distance, would theoretically result in 24 dB le= ss fieldstrength, or 0.19 fT - almost a perfect match. So by these lines, it would seem at least very likely that Paul= has indeed observed my feeble signal! We intend to do repeat the expe= riment in the near future for additional confirmation. Very many thanks to all involved in this work! 73, Markus (DF6NM) ------=_NextPart_000_000E_01CB6E12.2221D3A0 Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable
Hi Wolf,
 
Paul was using his own dedicated rece= ive=20 software, with timestamps based on GPS 1pps pulses. <= /DIV>
 
For transmitting, I employed Spe= cLab's signal=20 generator, with samplerate tracking to 10 kHz from a Conexant=20 Jupiter GPS module. For receiving, I normally lock = ;to 19.58=20 kHz, which seems to give good stability (except for a minor= glitch=20 when the skywave dips before sunrise).
 
As far as I have seen, the UK MS= K signals=20 (19.58 kHz, 22.1 kHz), the Americans (24 kHz), and NWC (19.8 kHz, but= too weak)=20 seem to be phase coherent to GPS. DHO (23.4 kHz) is stable but about&n= bsp;2.7=20 ppb high. The French (HWU) and Turkish (26.7 kHz) are drifting an= d=20 unusable.
 
The DK7FC grabber is also based on GP= S 10 kHz,=20 injected into the receive audio.
 
Best 73,
Markus (DF6NM)
 
----- Original Message -----
From:=20 wolf_= dl4yhf=20
Sent: Sunday, October 17, 201= 0 1:43=20 PM
Subject: Re: VLF: Re: Detecti= ons of 5=20 microwatt transmission
=
Congrats Markus and=20 Paul for this achievement !

I think you used a GPSDO with 10= kHz output=20 on both ends - Thunderbolt, Trimble, or similar ?

(still thin= king about=20 a simple but reliable, non-GPS solution using a 24/7 active MSK stat= ion, or=20 maybe DCF77 / MSF with a soundcard running at 192 kSamples, or SDR-I= Q running=20 at 196 kSamples ).

73 and have a nice Sunday everyone,
&nb= sp; =20 Wolf  DL4YHF .

Am 17.10.2010 12:56, schrieb Markus Veste= r:=20
Dear LF,
 
I am very happy to announce = that Paul=20 Nicholson in Todmorden was clearly able to detect each=20 of my three test transmissions. His detailed and most in= teresting=20 report is at
 
 
Hats off to Paul for his outstand= ing=20 achievement!
 
Best regards,
Markus
----- Original Message -----
From:=20 Markus Vester
Sent: Tuesday, October 12= , 2010 8:57=20 PM
Subject: VLF: Detections= of 5=20 microwatt transmission


Dear LF,
 
for my VLF test= transmissions on=20 the last weekend (Oct 9: 8969.998 Hz, Oct. 10: 8989.997 Hz), I= have=20 received reports from three receiving stations:
 
- Walter DJ2LF ne= ar Erlangen=20 (20.2 km) received the carrier again in good quality, using = ;0.95 mHz=20 resolution. Radiated power and received SNR were quite similar= to our=20 two-way QSO on June 4th.
 
- Stefan DK7FC in= Heidelberg=20 (178.5 km) reported about 10 dB SNR, and both dashes are still= visible in=20 the QRN minima on his 47uHz grabber window
 
 
This was the = ;intended=20 purpose of the experiment, and a nice counterpart to the ea= rlier=20 detection of a 200 uW kite transmission on Aug 29th. Even though= both of=20 us were locking the samplerate to a 10 kHz GPS-derived reference= , the=20 dashes appeared about two pixels low - perhaps due to a ver= y minute=20 rounding error in SpecLab's frequency scale display.
 
- To my utter sur= prise, Paul=20 Nicholson (Todmorden, 1030.5 km) produced two spectra, take= n over the=20 duration of the transmissions:
 
   http://abelian.org/vlf/df6nm/2010_10_09a.g= if  =20 (9:50 - 18:00, 34 uHz)
   http://abelian.org/vlf/df6nm/2010_10_10a.gif  =20 (9:00 - 15:00, 46 uHz)
 
After taking a de= ep breath, we=20 now have to deal with the question whether this is a significant= positive=20 detection. Except for a known central artifact on 8970, the high= est peak=20 appears in the correct frequency bin in both spectra. Naively,= one could=20 then propose that the probability of this happening at random wo= uld simply=20 be the inverse of the number of displayed bins, ie. around 1:230= for=20 Saturday and 1:170 for Sunday. Thus the combined probability of= a false=20 positive detection on both days would seem to be only 1 in 40000= .=20 Certainly there is a degree of arbitrariness in the choice of th= e display=20 range (8 mHz).
 
Paul estimated th= at the signal=20 was about 3 standard deviations on Saturday (0.3% false detectio= n rate),=20 and 2 sigma in the higher noise on Sunday (5%), giving a combine= d false=20 positive probability of 1 in 6667.
 
We can also look= at the=20 plausibility of the absolute fieldstrength of the peaks (about= 0.2 fT,=20 equivalent to 0.06 uV/m). If I remember correctly, Paul's first= detection=20 of Stefan's kite signal on March 15 was at about 3 fT, and Stefa= n was then=20 radiating approx. 1 mW EMRP. Scaling this down to my estimated= 5 uW EMRP,=20 and taking another dB for the slightly higher distance, would=20 theoretically result in 24 dB less fieldstrength, or 0.19 fT -= almost a=20 perfect match.
 
So by these lines= , it would seem=20 at least very likely that Paul has indeed observed my feeble sig= nal! We=20 intend to do repeat the experiment in the near future for additi= onal=20 confirmation.
 
Very many thanks= to all involved=20 in this work!
 
73, Markus=20 (DF6NM)

------=_NextPart_000_000E_01CB6E12.2221D3A0--