Re: [digitalradio] Re: HF search text
On Sat, 28 Apr 2007 08:18:42 - cesco12342000 [EMAIL PROTECTED] wrote: Has anyone started using it, and had any success? Yes, it's useful. From user guide: Two numbers appear at the end of each line. The first number tells whether the soft-decision Reed Solomon decoder failed (0) or succeeded (1). The second number gives a relative confidence level on a 0 to 10 scale for results produced by the Deep Search decoder. 002300 2 -23 0.6 -35 3 * GM4SLV WD4KPD FM15 0 10 Looks like I've been the beneficiary of your deep search file Cesco! Full QSO with David when I thought 20m was closed for the night (01:23 local time). The power of a) your CALL3.TXT database and b) Andy's online sked machine! Cheers 73 from Shetland (IOTA EU-012) John GM4SLV
Re: [digitalradio] Re: HF search text
On Wed, 25 Apr 2007 01:47:30 - cesco12342000 [EMAIL PROTECTED] wrote: This should do it, but untested with wsjt deep search yet. http://www.qslnet.de/member/hb9tlk/deepsearch.html Does this deserve a mention in The Bozo's Guide? The more people that sign up the more useful it will be. Has anyone started using it, and had any success? I only have one SSB transceiver, my IC706, which is in the car most of the time (K1 K2 for CW only in the shack). I only operate digital modes when I bring the 706 in to the house. This tends to be weekends only, so I haven't had a chance to try out the new CALL3.TXT from Cesco's site yet, but a few people are signing up. Wonder what the critical mass is to make it useful? Cheers, John GM4SLV
Re: [digitalradio] Re: HF search text
On Wed, 25 Apr 2007 01:47:30 - cesco12342000 [EMAIL PROTECTED] wrote: This should do it, but untested with wsjt deep search yet. http://www.qslnet.de/member/hb9tlk/deepsearch.html Cesco, you are a genius! That seems to work perfectly and I see others are already using it. I'll add the ones I got sent by private email and then let it just build naturally! Cheers, John GM4SLV
Re: [digitalradio] HF search text
On Mon, 23 Apr 2007 14:35:40 -0700 WN1Z [EMAIL PROTECTED] wrote: To the person who is compiling an HF deep-search text: please post here when you do that. Hi Orrin and others, I'm adding every Call/grid square I get sent, or that I can deduce from reading messages on this list. I haven't had a lot of time over the weekend. Apart from working I also had to repair my AR7030 Receiver. It's long story inolving a splitter feeding my HF antenn signal to the AR7030 and my IC706 so I could monitor 2 bands at once. I then forgot I'd set this up and when I saw K1JT (the inventor of WSJT) calling CQ on 20m I gave him a call. Worked him no problem. Then noticed profoundly deaf AR7030. Have found several chip Ls and Cs in the font end filtering have gone open circuit. Made a temporory repar pending new components. You can see wht it's normal job is by looking here:- http://g4irx.nowindows.net/fivemegs/comparison.php Anyway, back to the CALL3.TXT file. The more calls/grids I get the better, because without a significant number the function will be useless - I'd guess with a small pool to choose from you'd get a lot of false positives? Cheers, John
Re: [digitalradio] Deep Search on JT-65a
On Mon, 23 Apr 2007 19:28:34 - c6alk [EMAIL PROTECTED] wrote: Folks: I have not been able to find how to turn off the Deep Search function in WSJT version 5.9.0. I think you need to upgrade to the latest version, which I think is 5.9.6 and you'll find it under Decode - JT65 Cheers, John GM4SLV
Re: [digitalradio] 12M Deep Search/ Straight talk on JT65a
On Fri, 20 Apr 2007 23:20:49 -0400 Andrew O'Brien [EMAIL PROTECTED] wrote: I like Shetland John's idea about deep search work on 12 or 10 M. I think I will give it a try tomorrow on 12 and see what I come up with. Will check prop charts for best time to north of John O' Groats... Andy K3UK Andy et al, I'm going to try CQ on 24.910 (DF 0hz) on Fist Minute for most of the day while I'm around the house. I don't expect to be overwhelmed with a pile up! My simple station at the moment is IC706mk2g, set to ~20w on the sync tone, and a random length doublet only 2m agl (it was put up to monitor the 5MHz beacons sytem we have over here for our 5MHz Experiment, for which it seems to wirk adequately even given its low height, but has been pressed into service for JT65A). I will look into creating a deep search text file for as many Digital Radio members as I see are active, and I'll guess the LOC if I have to and enable the function later in the day when I've built the file. It's off CQing now as I type (@ 0928z) Let's see what I see over the day. BTW I saw nothing of the Olivia you were trying on 20m in the early hours (over here) - 20m was closed completely. Cheers, John GM4SLV IP90gg
Re: [digitalradio] 12M Deep Search/ Straight talk on JT65a
On Sat, 21 Apr 2007 10:32:25 +0100 John GM4SLV [EMAIL PROTECTED] wrote: I'm going to try CQ on 24.910 (DF 0hz) on Fist Minute for most of the day while I'm around the house. I don't expect to be overwhelmed with a pile up! My simple station at the moment is IC706mk2g, set to ~20w on the sync tone, and a random length doublet only 2m agl (it was put up to monitor the 5MHz beacons sytem we have over here for our 5MHz Experiment, for which it seems to wirk adequately even given its low height, but has been pressed into service for JT65A). I ran for several hours this morning and saw no responses or other signals. I need to go out for a few hours so have stopped the CQ calls at 1312z. I'll restart when I return later this afternoon (UK time). Knowing me I'll miss the tiny 1/2 hour long opening when signals appear, but hey, there's always another day! John GM4SLV IP90gg
Re: [digitalradio] 12M Deep Search/ Straight talk on JT65a
On Sat, 21 Apr 2007 09:37:31 -0400 Andrew O'Brien [EMAIL PROTECTED] wrote: I'm not sure which Shetland island you are on John, I checked the Lerwick market web cam but did not see you out doing your shopping. I will try you later. Andy K3UK Hi Andy, Just returned and see absolutely no evidence of any signals on 24.910 while I was out (left in on MONITOR). Now restored to CQ on first minute. I'm on the main island, and was in Lerwick, but didn't go to the Market Cross so kept myself incongito! If 12m fails to show anything perhaps we could try 15m another time. It would probably need to be an afternoon UTC for UK/USA path? I'll leave it running for now. BTW what's your (and anyone else who's interested) Locator to 6 figs and I'll start a new deep search file. Cheers, John GM4SLV IP90GG
Re: [digitalradio] 12M Deep Search/ Straight talk on JT65a
On Sat, 21 Apr 2007 12:58:12 -0400 Andrew O'Brien [EMAIL PROTECTED] wrote: John, name is Andy and am in FN02hk . I have been monitoring 12M while watching footy on the telly but just checked and see no received signals from anyone on 12M. I will transmit for a while and see if anyone picks me up. Andy. Thanks for listening/looking Andy. I'll start an HF oriented CALL3.TXT file and distribute it via the files section on the yahoo group page. Already people are starting to email me Callsign Locator pairs. The file actually has more fields available for each entry (it's a comma seperated text file), if you look through it, and some fields are for personal/equipment details too, as well as the EME field. I've been calling CQ on 24.910 since 1608 UTC (time now 1730 UTC) and in that time have seen nothing during the second minute RX periods. I'll leave it for a while longer... John GM4SLV IP90gg
Re: [digitalradio] Deepsearch on JT65 HF --turn it off
On Thu, 19 Apr 2007 20:58:01 -0500 w6ids [EMAIL PROTECTED] wrote: Well, Bill, I'll thank you for that. It answers a nagging question. I've seen instances where I've seen myself being called by a number of stations over time periodically.and I wasn't even keying the rig. Can you explain what is happening - I DID have Deep Search ON as I recall. Only recently did I switch it off. Howard W6IDS Richmond, IN Hi Howard, As far as I can see the deep search function uses the CALL3.TXT file and the MYCALL value and tries to correlate the incoming signal with every combination of these. The one with the highest correlation factor is presented as its guess as to what the incoming signal is. This is great for decoding weak signal way down in the noise - it's always easier to hear something if you know what's been said in the first place - but since most of the people active on HF aren't already in the CALL3.TXT file you're more likely to get a false positive. This explains seeing your own station being called when you've never transmitted a picowatt of JT65A. Yours was the best correlation to the incoming data. Turning off Deep Search prevents this action. As others have mentioned overnight, a dedicated HF CALL3.TXT should allow us to use this feature. All that needs to be done is for active users to start with a blank CALL3.TXT file and add stations as they are worked. That's what the ADD button does on the To Radio area. Unfortunately it seems to require a full 6 figure Locator and unless you know the operator, or exchange this information at the end of the QSO I suppose you could stick in a standard dummy pair of letters for the small sub-squares (what's the letters for the middle small square - we could standardize on that?)for HF we don't really need the minute accuracy of a 6-figure locator. Once people have collated their own CALL3.TXT we could share them and combine them to include all the unique callsigns from all the individual files. A small bit of excel bashing? I have found a few HF operators I've heard are actually in the existing CALL3.TXT file - so they've been active on JT65 on V/UHF (poss EME too?) for a while now, and HF is just another string to their bow. Cheers, John GM4SLV
Re: [digitalradio] Re: Deepsearch on JT65 HF --turn it off
On Fri, 20 Apr 2007 01:25:12 - Bill McLaughlin [EMAIL PROTECTED] wrote: Hi John, Sure you already know this but the key was the 0 3 082600 1 -14 5.2 584 3 * GM4SLV SM2LKW KP15 ? 0 3 Hi Bill, I only started with JT65 yesterday but it didn't take me long to dig through the docs to explain this one and turn Deep Search off. I read through the CALL3.TXT file and think we should adapt it for our need if everyone was willing. Cheers, John GM4SLV
Re: [digitalradio] Re: Straight talk on JT65a
On Sat, 21 Apr 2007 01:00:52 - expeditionradio [EMAIL PROTECTED] wrote: --- In digitalradio@yahoogroups.com, Leigh L Klotz, Jr. [EMAIL PROTECTED] wrote: My conclusion is that this mode is about 10-15 dB worse than it appears to be.. After a week or so of monitoring JT65A on 14MHz and 7MHz, I tend to support your conclusions. Nonetheless, the recent fad flurry of activity on 14076 with JT65A demonstrates the enthusiasm for any potential new extreme weak signal modes during the bottom of the solar cycle. I agree that it's not all it seems to be. I've been following the discussions over the last month so a few days ago I started to spend time receiving JT65A on 20m, and today started having contacts. I must thank David WD4KDP for my first QSO, this afternoon UK time. I was initially impressed:- GM/Shetland to WD4 with only 10 watts to a random length dipole hastily errected at 6' agl (there are no trees here to use as makeshift supports) was a nice introduction to the possibilities of a new mode.. However, on reflection his signal on the waterfall would have been decodable on Olivia or MFSK ,and quicker with more information. He was showing as -10dB S/N but as mentioned in previous posts, the reported S/N level a moot point. My feeling is that, brilliant as it might be for its intended purpose, with the deep search function, operating within organized skeds for EME on V/UHF, it just is too sensitive but lacks the dynamic range for HF. Bonnie is right that we're at the bottom of the solar cycle, but 20m is still not the barren wasteland of 70cm, where you are poining a beam out to space. We aren't trying to sniff out a tiny signal in an otherwise silent band, a job that JT65 must excel at. I've seen signals today on 20m that were registering as -20dB and others as -1dB, with the strongest signals showing S9 on the rig's S-meter. Strong signals just seem to upset things, the software is much happier with a nice clean, but very weak, possibly fading and multipathed, signal. I find that if I adjust levels to keep RX Noise to ~0dB on a reasonably quiet band (a few -10dB signals) then as soon as a big signal appears the red warning lights start flashing, the waterfall turns white and signals are shown as -1dB but with no message recovered. The mode doesn't seem capable of the dynamic range we see on the HF bands. Perhaps we need to try it on the notionally closed higher HF bands like 12m/10m and work on prearranged skeds, with a valid deep search file to test propagation during this low point in the solar cycle? Just my observations after a few days of messing about with the mode du jour. Cheers, John GM4SLV (I'll still be giving it a go though, it's quirky enough to be interesting!)
Re: [digitalradio] My Own JT65A Monday - FINALLY
On Thu, 19 Apr 2007 18:13:35 -0500 w6ids [EMAIL PROTECTED] wrote: I know the thing works, but there's this thing about a learning curve, time accuracy, etc. So, I was kind of surprised when I actually pulled off the sequence of making a contact that is located elsewhere besides CONUS: 222800 0 -21 7.1 -127 3 222900 10 -5 1.5 -124 0 * CQ JS1OYN PM951 0 223200 4 -7 1.8 -16 3 * W6IDS JS1OYN PM95 1 0 223400 4 -9 1.1 -16 3 # W6IDS JS1OYN PM95 OOO 1 0 223600 10 -20 -17 3 RRR ? 223800 10 -19 -17 2 73 ? Howard, Coincidentally I started on the learning curve yesterday for HF JT65A along with another local radio club member, Steve GM7GWW. We're currently in QSO together on 2m while we both watch 20m or 40m for JT65A activity while we get to grips with the mode. For your information I received the following from you today on 20m. I was using an untuned random length dipole only 2m agl :- 162200 7 -18 1.3 -35 3 * OH5VG W6IDS EM79 1 0 223100 5 -16 1.6 137 2 * JS1OYN W6IDS EM79 1 0 I'm in the Shetland Islands (IP90) so I guess that counts as outside CONUS! I hope to start transmitting JT65 in the next few days once I'm comfortable with driving the software, and when I remove my IC70611g from the car. At present I'm receiving on my K2 (no SSB module so TX). There certainly seems to be a steady trickle of activity on 20m, and some familiar callsigns for lurkers to the digitalradio group to spot. Cheers from Shetland (IOTA EU-012) John GM4SLV
Re: [digitalradio] 30M OLIVIA
On Thu, 28 Dec 2006 11:42:43 -0600 John Bradley [EMAIL PROTECTED] wrote: at 1730Z calling CQ on 10140 1000X32. heard MFSK down a bit but band condx too bad to copy more than 50% at best. John VE5MU Hi - I've just received your mail and am now listening/calling on 10140 with Olivia 1000/32 Maybe my MFSK you saw - tried calling a few times over last hour to no avail. Cheers, John GM4LSV signature.asc Description: PGP signature
Re: [digitalradio] Clarification : Establishing digital calling/beacon frequencies
On Thu, 21 Dec 2006 17:52:37 -0500 Danny Douglas [EMAIL PROTECTED] wrote: As you say, everyone is a bit different as to where their waterfall pointer lies on a digital signal. It is much better to just spot the actual freq where the audio signal comes out on the waterfall. I.E. 14.0731. No matter who clicks on a spot like that, their trace comes out on the waterfall as that freq. We dont care where each others dial freq is - but the location of the signal. In your case its dial plus 1500 - In mine its dial plus 1000, other will have similar setting according to where their sound card best passes a signal (sweet spot), and there is no qestion where the transmitting signal is found. Double click a spot, and bang- your reciever/transmitter freqs are set properly, and no one has to look around to see where in the world the spotter found it. I'm not sure I follow this argument...not the part about deriving RF frequencies from dial audio...about using DX cluster spots to auto tune If I click on a spot on a cluster (as above...say 14.0731) then my rig QSYs to 14.0731...ie carrier/dial frequency. If the actula RF frequency of the signal I want is on 140731 then I'm not going to see it, as it'll be a zero Hz on the waterfall. All the signals in the passband are therefore above 14.0731 (assuming USB)...with the middle of the passband corresponding to roughly 14.0746). If I read you right you're suggesting reporting spots with the actual RF frequency of the TX (derived from your software adding AF frequency to the reported RF frequency from the rig - taking into account USB/LSB). This is fair enough since that's all you can know, you can't know what combination of dial and audio was configured by the sending station, nor the infinite combinations used by everyone else. My problem is that in my experience clicking a spot on a DXCluster will put the rig's dial to that QRGyou still need to offtune down a kHz or so to put that specific frequency into the passband... I agree that the actual QRG should be the one reported - but AFAICT we'll still need to do our own maths to put the required RF frequency into the right area of baseband on the waterfall. BTW I find most Olivia sigs in Europe around 14106.5 or 14107.5 (centre)...with 1k/32 that puts the bottom carrier @ 14106 or 14107. I see MFSK, as already reported, around 14073 and up with very little below the PSK31 region. Never knowingly seen Throb. Cheers from Shetland John GM4SLV signature.asc Description: PGP signature
