The Wireless Set No. 19 Homepage

Free translations of the 19 Set pages

Atlantic Hop III Progress Report

by Chuck Counselman W1HIS 2000 December 16 SUMMARY Inspired by the success of Operation Atlantic Hop II in 1998-1999, M0BWH in England and LA5MT in Norway have joined several U.S. and Canadian members of the Wireless Set No. 19 Group in Atlantic Hop III exercises this (2000-2001) winter. One successful exchange has already been made, and several near misses have occurred. We have learned that we need to avoid weekends, improve our no-contact / loss-of-contact procedures, be nimbleto evade QRM, work on our antennas, and be very respectful of ionospheric conditions. More exercises are planned for this winter. INTRODUCTION Reading Dave Lawrence VA3ORP's Final Report of Operation Atlantic Hop II has inspired several members of the Wireless Set No. 19 Group to conduct a more modest, but similar, "Atlantic Hop III" exercise this (2000-2001) winter. As in the 1998-1999 Atlantic Hop II, the nominal object of the present exercise is two-way communication across the Atlantic Ocean between North America (NA) and Europe (EU) using Wireless Set No. 19 transceivers on both sides. The real objects are to become better WS 19 operators, and to have fun, not necessarily in that order. Our experience, the experience of others in Atlantic Hop II, and recent developments such as Bob Cooke's (VE3BDB's) establishment of the Wireless Set No. 19 e-mail list, website The Wireless Set No. 19, and QSO (Chat) Room, dictated the following parameters for Atlantic Hop III: Mode: CW Frequencies: between about 7.015 and 7.035 MHz, avoiding integer kHz and especially avoiding integer multiples of 5 kHz. Frequency changes to be expected several times per session, to avoid QRM. Several schemes for selecting frequencies have been tried (see below). Season of year: winter in the Northern Hemisphere. Time of day: late night / early morning (entire path in darkness). Day of week: "weekdays" (i.e., avoiding Friday, Saturday and Sunday nights; or Saturday, Sunday and Monday mornings). Mode of communication for advance planning, scheduling, organization, and recruitment: e-mail -- both direct and via eGroups subscribers who have elected to receive single daily message "digests" should be reminded to change their settings to receive individual messages immediately as sent, lest they miss eleventh-hour notices. Modes of communication for eleventh-hour and real-time coordination, including go / no-go decision-making in light of changing band conditions, etc.: (1) e-mail, both direct and via; (2) telephone; (3) chat room. We tried broadcasting all relevant information in real time via the chat room, but no one ever logged into the chat room to receive it or to contribute signal reports as we had hoped. See Lessons Learned, below. Establish contact on the working frequency initially; re-establish it if lost; jawbone and wave off interfering stations; and QSY, using high power on _both_ sides of the Atlantic. Choose antennas for maximum gain, in view of ground properties, at elevation ("take-off") angles of 10 - 12 degrees, corresponding to three-hop F-layer propagation, and at the relevant azimuths, e.g., 40 degrees from W1HIS to LA5MT. NEC-4 simulations assuming poor ground show that a properly oriented (broadside) horizontal half-wave dipole 10 m above ground beats a quarter-wave vertical having four radials elevated 1 m. It also beats a quarter-wave (loaded) vertical dipole with its bottom end 1 m above ground. A wire antenna with gain, such as a wire Yagi, a single long wire, a long sloping vee, or a rhombic, could be very helpful; however, so far no Atlantic Hop III participant has used one. The problem is that more than 1.5 wavelengths (i.e., > 123 m, or 201 feet) of wire are needed to improve significantly upon a half-wave horizontal dipole. In my opinion, use of a Yagi on a tower, or a phased array such as a four-square, would be unsportsmanlike because such a mechanically elaborate and unportable antenna would not have been used with a 19 Set in WW II. Warm up 19 Sets for at least 30 minutes in advance, for frequency stability. Monitor band conditions and Internet propagation reports/forecasts between 60 and 10 minutes before sked time, and consider canceling operations for the night if conditions seem hopeless. Ideally, the decision should be made jointly by the NA and EU co-NCS ops, via direct e-mail, telephone, or chat room; and the cancellation announced via and chat room. However, see Lessons Learned, below. Propagation forecasts are not very reliable (as detailed below), and conditions can change very suddenly. In my experience the two best indicators of conditions have been: (1) scanning the band and listening to signals from the other side of the Atlantic; and (2) the Millstone Hill (Massachusetts, USA) ionosonde . In an ionogram you do _not_ want to see sporadic-E, ragged or "spread" F, much vertical motion of the F layer, or foF2 below 4 MHz. Late at night, foF2 is lower on the eastern side of the Atlantic and at higher latitudes, where F-layer ions have had more time to recombine. This is a problem when solar activity is low. Operations: Exchange and confirm accurate receipt of: call signs, signal reports, 19 Set descriptions, and 19 Set serial numbers. Recommended procedure: Send at 10 to 15 words per minute. Begin each transmission with callsigns; acknowledge receipt by echoing just-received information, or request retransmission; send just one item of information, e.g., signal report; end with callsigns and KN. Repeat _everything_, including callsigns and prosigns(!), unless conditions are impossibly good. The sender should honour the receiver's wishes regarding the number of repetitions. The tendency is for a sending station to be impatient and not repeat as many times as requested, and/or to end a transmission with just one "BK", which can get lost in a fade. Stations on both sides (EU or NA) having high power may act as co-NCSs. On 40 m late at night foF2 is usually substantially less than 7 MHz, so there is a sizeable skip zone. Either co-NCS may be unable to hear either QRM or WS 19 Group participants from his own continent. However, VA3ORP has argued against having _any_ NCS: "Using an NCS requires that all/some/most of the stations can hear the NCS. If this doesn't happen (as seemed to be the norm during AH-II), then it is very hard to QSY or to turn the frequency over to two stations for them to try the exchange. I am now tending towards a simpler procedure where eastern stations would transmit on [say] 7017 (+/-5) kHz from hh:00 to hh:10. During this period western stations would listen and respond to any strong signals they heard. This would be followed by western stations transmitting on 7017 (+/- 5) kHz from hh:10 to hh:20 while eastern stations listened/responded. This rota would continue until two stations linked up, at which time they would pass their exchange. This appeals from a couple of points-of-view. To start with, it is simple to implement. Secondly, several stations can be 'having a bash' at the same time, increasing the likelihood of success. Finally, and most importantly from the point of view of vintage operations, it depends on a good loss of contact drill and operator skill rather that a lot of technology that was not available 50 years ago." I personally favour having a high-power NCS launch the session, maintain control of the frequency, and facilitate frequency changes. Fifty years ago, I suspect, there wasn't so much QRM. See Lessons Learned, below. An announcement of a scheduled operation must include UTC date(s), UTC time(s), initial frequency and frequency range (for avoidance of QRM), and unambiguous instructions regarding the Net Control Station(s) and who calls whom, when. The initial call should begin on time and be fairly long, e.g., LA5MT LA5MT LA5MT LA5MT LA5MT DE DE DE W1HIS W1HIS W1HIS W1HIS W1HIS KN KN KN. Instructions regarding mode(s) of communicating updated information, e.g., via <>, should be included in the initial e-mail announcement. It is helpful also to announce the power that will be transmitted initially, so that listeners can estimate, from the initial signals, what strength to expect for the WS 19 signals to follow. CHRONOLOGY OF EVENTS 2000 November 17 On 2000 November 17 (Friday) at 0430 UTC, W1HIS in Massachusetts called M0BWH (Steve Jones) in England on 7037 kHz, within the pre-announced range of 7030 to 7040 kHz. M0BWH replied on the same frequency. This exchange was made with high-power modern equipment. W1HIS transmitted 1.5 kW -- 20 dB, or about 3.3 S-units above his (actually John Forster's) WS19 (Canadian) Mk III. M0BWH transmitted 100 W -- 4 dB or about 0.7 S-units above his WS 19 (American) Mk III and British 19 HPA. It was planned that if/when contact was made, a QSO between 19 Sets would be attempted on the same frequency. The plan had been announced via As had been predicted for days, propagation was good, in fact excellent, on the day of the sked. The Boulder K index had been equal to 0 (zero) for six hours, and was equal to 1 for the last six hours before the sked. NOAA/WWV reported at 0604 UTC that the Boulder A-index was = 1, which is exceptionally low (good). But minutes before our 0430 UTC sked time, the Millstone Hill ionosonde showed that the F layer was messed up and sporadic E was intense, with foEs exceeding 3 MHz. Steve M0BWH telephoned Chuck W1HIS to report that he could barely hear Chuck's 1.5-kW signal. Neither Chuck nor any other North-American participant heard Steve's 100-W signal. By the time Steve called, Chuck had given up listening for him and had QSOs with WS 19 Group members Paul Beckwith K2LMQ/7 (transmitting 500 W through a 160-m inverted-V G5RV up 50 feet in Arizona) and Vic W1NU (transmitting 10 W with a 19 Set and standard-size inverted-V G5RV up 50 feet in Connecticut). Steve, guided by Chuck via telephone, tried but was unable to hear either K2LMQ/7 or W1NU. Reed Park VE1NU, in New Brunswick, heard W1HIS "loud and clear" on both his Icom with a dipole, and his 19 Set with a vertical antenna. Reed also had a nice chat with Vic W1NU, and "finally got his serial number logged correctly". Bruce MacMillan VE7MT, on the west coast of NA, monitored with an Icom IC-735 and a N-S inverted V. He reported that "Chuck was of course loud but conditions to Europe were [rotten]. However I could copy about 50 % of Vic W1NU in his QSO with Paul, K2LMQ/7. Vic's chirps were fading in and out of the noise about 0440Z but seem[ed] to improve as 0500Z approached. Local QRM at 0505Z obliterated any further copy. Still, Vic's 10 watts made it 2400 miles to the west." Paul K2LMQ/7 also "had a neat 25 minute QSO with Vic W1NU. His 19 Set was 579 in Northwestern Arizona." Mike Ursin LA5MT (successful in Atlantic Hop II), in Norway, monitored "from 0430Z - 0500Z using my [Icom] IC-781 and delta loop. Could copy Chuck quite well at about 0445Z in QSO with K2LMQ/7. RST was 539; no QRM, but moderate QRN." 2000 November 18 A similar exercise was scheduled for 0430 UTC on the next day, 2000 November 18 (Saturday morning). At 0400 UTC Meir Ben-Dror WF2U, in New York, reported via e-mail that he could "hear a lot of Russian stations on 40 m." Propagation was indeed better than on the previous night, but not good enough. At 0300 UTC (1.5 hrs before sked time), solar flux was 163; the Boulder A-index was 2; and the Boulder K-index was 0 (zero). The Boulder K-index was still 0 (zero) at 0600 UTC. All pretty wonderful numbers. Sporadic E, which had been deadly the night before, was absent tonight. However, the F layer still looked ragged in the Millstone Hill ionograms. QRM was more of a problem because of the weekend. However, no big CW contest was under way, and the ARRL SSB Sweepstakes may have diverted stations away from the CW sub-band. At 0423 UTC, W1HIS, transmitting 1.5 kW from a modern rig, began calling M0BWH on 7034 kHz. M0BWH answered on the same frequency with 100 W from a modern rig. W1HIS' signal was RST 579 with QRM at M0BWH. M0BWH was RST 229 to 339 with QSB but initially no QRM at W1HIS. M0BWH then switched to his WS 19 and British HPA, transmitting 40 W, and W1HIS reported, still via 1.5 kW, very difficult copy. However, during favorable "fades" Steve's WS19/HP signal was copyable and easily recognizable as a 19 Set by its chirp and its dyno-modulated note. With much repeating by M0BWH, W1HIS was able to copy M0BWH's name (Steve), QTH (Ely), rig (WS19/HP), and serial number (96818). It helped for M0BWH to repeat each item five (yes, five) times and to break for confirmation before sending the next item. At about 0530 UTC W1HIS switched to transmitting with John Forster's WS 19 Mk III. M0BWH was able to hear it, but just barely; he could recognize the signal as that of a 19 Set but it was too weak to copy, so no attempt was made to complete a two-way exchange with 19 Sets. At 0552 UTC M0BWH began sending strings of Vs and his callsign, to enable W1HIS to comparereception on the 19 Set with reception on his Icom IC-775DSP. Surprisingly, in view of the bandwidth difference, reception on the 19 Set was not much worse than on the Icom -- a testimony to the ability of the human ear & brain to pick a 19 Set's chirpy signal out of a mess. Others have reported similar observations (see below). Propagation, which hadn't been great to start with, seemed to be deteriorating, so we pulled the big switches around 0600 UTC. Throughout the operation the participants kept having to pause, find new frequencies, and QSY due to QRM. These frequency changes were coordinated via e-mail for a while, and then, when the QRM got really thick, by telephone. The QRM and resulting frequency changes made it difficult for monitoring stations to follow. As of this night we had not learned how to get the WS 19 chat room to work, so there was no way to broadcast frequency changes to the WS 19 Group in real time. (See Lessons Learned, below.) Meir WF2U, in New York, monitored with an Icom IC-740. At 0430Z he "heard W1HIS (R 5, S 3 to 4, T9, with noise level S3 to S5) calling M0BWH on 7034 kHz. At 0440Z heard M0BWH answering HIS. M0BWH was R 4, S 3-4, T9. BWH then transmitted with the typical 19 Set tone; heard him with the same signal strength. I copied M0WBH giving the signal report to HIS, his name and QTH. I called HIS with the modern rig to draw his attention that I'd like to join. I think he answered "AS" but at that point his signal was weaker than M0BWH's signal. I copied M0WBH on my WS 19 Mk III with the same perceived signal strength as on the Icom except it seemed like there was less QRN heard on the 19 set. Then the QRN increased for a while and signals faded. I kept searching for the stations [and] heard W1HIS periodically, between barely discernible to RST 229 - 549. At approximately 0508Z I heard M0BWH repeating 19 Set serial number a number of times, responding to W1HIS's repeat requests. I heard the whole exchange on my 19 set, while the actual S-readings on the modern rig were between 3 and 5 depending on QSB, with QRN between S 2 and 4. At this point HIS and BWH had approximately the same signal strength. I didn't attempt any further transmissions after my initial call to W1HIS because I didn't want to interfere with the ongoing efforts between him and M0BWH to communicate. The antenna I used was the same 1/4-wave 40-m ground-plane that I used during the previous (AH II) operation, when I had a successful 19 Set to 19 Set contact with LA5MT. Propagation as heard was not too great,regardless of what the numbers said." 2000 November 26 A third attempt was made at 0430 UTC on 2000 November 26 (Sunday morning). We expected QRM to be worse than usual, even for a Saturday night / Sunday morning, due to the CQ World Wide DX Contest. The Australian IPS was also predicting that high-latitude propagation would be "Poor". Despite the DX contest we found a quiet spot at 7085.40 kHz, and despite the dire predictions, propagation was not totally bad. At 0439 UTC, Steve M0BWH heard W1HIS' QRO sig RST 579, and immediately switched to his WS19/HP. W1HIS heard Steve's 19 Set well for a while, copied his QTH and rig description without difficulty, and got part of the serial number before propagation died and S9 QRM buried Steve's signal. The operation was scrubbed at about 0515 UTC. For the duration of this exercise John Forster was logged into the WS 19 Chat Room from the W1HIS shack, but no one else entered except Steve, just to test the chat room, after we'd shut down the radios. 2000 November 29 On 2000 November 29 (Wednesday) at 0500 UTC an attempt was made by Chuck W1HIS and Mike LA5MT. A slight complication was that Mike's computer, telephone, and high-power modern rig are not in the same room as his WS 19, so he has no real-time "supervisory" channel. However, at his WS 19 position he has a GRC-165 (100 watts output), and he can switch rapidly between these two rigs. LA5MT called W1HIS on time, transmitting 100 W on 7020.1 kHz. W1HIS heard him immediately because he was so close to the planned center frequency of 7020 kHz, and because the band was so dead that there was no QRM. Conditions were dreadful, due to a solar-geomagnetic storm. We learned later that the Boulder K index and the planetary-average index Kp had both been equal to _7_ (seven)! NOAA's web page showed that the entire path between W1HIS and LA5MT was within the northern auroral zone. Mike's signals were RST 229 fading to 119. If Chuck hadn't known his and Mike's call signs, he could not have copied either one accurately. During one of the better moments he was able to copy his signal report as RST 329. This surprisingly good report probably reflected the fact that Chuckwas running 12 dB more power than Mike (1500 W vs. 100 W). At about 0510 UTC Chuck told Mike that he would not even try to use a 19 Set, and invited other 19 Set Group members to check in. The only one heard was Vic W1NU, who was using a 19 Set. Vic's signal was not much easier to copy than Mike's. Chuck asked Mike whether he heard Vic, and copied Mike's reply "nil copy for W1NU". We called it a night at 0528 UTC. 2000 November 30 Again on 2000 November 30 (Thursday) at 0500 UTC Mike LA5MT called Chuck W1HIS on the previously agreed frequency of 7020 kHz, and two-way contact was established using high power. Severe QRM from a relatively wide-bandwidth data-modulated signal forced a move to 7021 and then to 7025 kHz. Then both operators switched to 19 Sets; and exchanged and confirmedcallsigns, signal reports, 19 Set descriptions, and serial numbers. W1HIS then sent his name and QTH but, due to QRM, LA5MT was unable to copy. W1HIS repeated his info but again LA5MT was unable to copy, and at 0609 UTC Mike had to QRT to go to work. Vic W1NU copied both sides of the exchange and worked W1HIS at 0609 UTC but did not have an opportunity to work LA5MT. This was a shame; Mike reported later by e-mail that he "could copy Vic W1NU using his 19 Set just before I had to leave for work" and that at this later time W1HIS' "signals then were quite strong - RST 579." 20:20 hindsight: Given a "back channel" such as the telephone, e-mail, or the chat room, or if we had reverted to high power, we could have arranged to move from 7025 kHz to a new clear frequency, quickly. Propagation was fine. We needed only to get out from under the QRM. 2000 December 1 Again on 2000 December 1 (Friday) at 0500 UTC Mike LA5MT called Chuck W1HIS on 7020 kHz. Vic W1NU was on frequency again with his WS 19. Unfortunately, and contrary to all the forecasts, propagation was worse than the previous night/morning. Mike could copy W1HIS' QRO signal with little difficulty, but he could not copy Vic's 19 Set. We all hung around and waited for conditions to improve, but they did not, and at 0600 UTC we QRT'd until another night/morning TBD. More Atlantic Hop III sessions are planned for the Christmas holidays, and in January 2001. LESSONS LEARNED E-mail, and the list and in particular, is the preferred mode of arranging Atlantic Hop exercises. The WS 19 Chat Room, and _real_time_ e-mail, are _not_ effective. For a variety of reasons, including RFI and inability to type and work a Morse key simultaneously, too many operators are unable to use their radios and their computers simultaneously. In addition, it is arguable that the use of a real-time communication link to back up the 19 Sets is unsporting. It has been argued that using high power rigs to back up the 19 Sets is likewise unsporting. In my opinion the jury is still out on this charge. On 2000 November 30 a very strong, rather wide-bandwidth, data-modulated signal was centered on, and blanketing, our sked frequency of 7020 kHz. Without high power, Mike LA5MT and I might not have found one another, and a night/morning of excellent propagation might have been lost. OTOH, if Mike and I had prepared and executed a good plan of calling on prearranged alternate frequencies in the event of QRM, we could have done fine without high power. In the event, (1) our initial use of high power facilitated our QSYing twice, which enabled our successful WS19-to-WS19 exchange; and (2) our failure torevert to high power when new QRM arose wasted so much time that W1NU was unable to make a WS19-to-WS19 exchange. In any case, it seems that attention to a no-contact / loss-of-contact drill is needed. Be prepared to QSY often, quickly, and radically. We learned this the hard way. However, our willingness to move radically yielded a clear frequency even during a CW DX contest. Avoid weekends and avoid grief. Both Atlantic Hop II and A H III have taught this lesson. Don't put much stock in short-term propagation predictors such as NOAA / WWV bulletins. The Millstone ionosonde is quite informative, but this is more of a real-time indicator than a predictor. Long-range predictors such as monthly-average patterns, solar activity trends, and solar-rotation, 28-day "recurrence" based predictions are certainly useful for planning. (Thus, we know not to attempt an Atlantic Hop on 80 meters in mid-day in June.) When solar activity is low, the MUF for the path from NA to Northern Europe can drop below 7 MHz very late at night, i.e., in the late pre-dawn hours at the eastern end of the path. This problem is more likely to occur around the winter solstice, when northern nights are longest. I believe that I have observed this problem more than once this winter, even near solar-maximum! Ionospheric maps should be consulted when scheduling AH exercises in midwinter. Three useful resources are,, and When using ionospheric propagation prediction software, beware of so-called "3000-km" MUFs, which are based on unrealistically low angles of wave departure/arrival. Consider, for example, that propagation from Massachusetts to England probably involves three F-layer hops of about 1750 km, and wave angles around 13 degrees. Most of our antennas have very little gain below this angle (and precious little here!). Most of us have paid too little attention to our antennas, IMO. I am as guilty as anyone. My station antenna, a horizontal doublet 30 m long, is precisely (within two degrees of azimuth) end-on to M0BWH in England. I believe that erecting a properly oriented (broadside) dipole would make abig difference. NEC-4 says that it'll give me 10 dB -- and that's for receiving as well as transmitting! FUTURE More exercises are planned this winter. Watch for announcements and updates. Meanwhile, get to work on your antenna! 73 de Chuck W1HIS

Return to The Wireless Set No. 19 Home Page

(Copyright ©1999, 2000, 2001 by Bob Cooke, VE3BDB. All rights reserved. Nothing on this page is to be reproduced without written permission.)

Chat directly with the Site Master!
Click below and if I am online,
we can have "real time" contact.
Otherwise, you have the option
of sending me a message if I'm busy or offline.