Summary of NI Operation
23 April - 3 May 2016
Compiled 10May 2016
Firstly I would like to point out that this operation was a “holiday activation” and not a true DX-pedition so the station was not on air all day and all night. The weeks previous to the event were marred by bad family circumstances such that I was tired before I went, more-so while I was there, mainly due to lack of sleep and things still happening at home. That does not really make for late nights working DX when you wake at maybe 2AM, 3AM or 4AM and you have to be considerate of your XYL who is similarly tired so you can’t just turn on the radio and make a lot of noise, and even just speaking into a microphone is too much at those hours.
What went right…
The gear transported well within my main suitcase but it was packaged in bubble wrap to reduce the probability / possibility of damage. The total weight of the gear plus coaxes plus antennas plus accessories plus suitcase was such that I could only add two pairs of cotton shorts, two polo shirts, one pair of joggers and a pair of rubber thongs (/scuffs /jandals) before hitting the 23Kg suitcase weight limit.
The radio gear set up on the table in the apartment more or less as I had expected, the palm tree just outside the “apartment” was within a 20 metre run of coax (two x 10m actually) and was a good mounting point for a halyard to support the 1:1 balun. The fully homebrew multiband trapped dipole was able to be tied off to the frame of the communal BBQ shelter at one end and the other was held in place with one of two tent pegs taken “just in case”. I used only part of one of the two 50m x 4mm packages of nylon rope to create both the halyard and the two end ropes. The other returned home unopened.
The care I had taken in fitting toroids and clip-on ferrites on the various interconnecting leads reduced self-generated QRM to almost nothing both to and from the radio gear from the nearby antenna arrangement, particularly from RF into the headset lead and newly-homebrewed headset adapter. The only issue I noted was that on 80m and 40m, the audio in the headset seemed “edgy” on transmit but that was solved by creating a 9-10 turn coax choke balun with the feeder’s extra and unused length. I doubt that the RF pickup would have been an issue at all if an effective RF earth (eg stake) had been available and close at hand. Otherwise, the homebrew headset adapater box worked well and provided easy setup of the transmit audio from both the headset and the computer’s headphone output for the voice keyer, the internal VOX gain and delay settings, and probably the most important facet was the receiver audio and sidetone level in the headphones. That sidetone level was intentionally high to prevent me from shouting into the microphone – something I tend to do if I can’t hear myself.
The little 10” Asus Netbook EEEPC (with Win7) worked well for the logging function, my simple voice keyer software worked well too and saved my voicebox from a lot of calling of ‘CQDX’ over the days. I used the N1MM+ logging software set up for Dx-pedition style logging so that I only had to enter the callsign and the two RS(T) values, the frequency and emission mode details being taken care of by a C-IV connection to the IC-7000. I did end up with a corrupted database on day 2 due to an error I made in my backup batch file but I had done an Export to ADIF just prior to that so I was quickly able to create a new empty database and then import the ADIF file to recreate the entire log. I then made it a point to File/Export the log data to a new filename (ie xxxx.ADI ) at least every day, mostly more than once a day, so that I need not panic if something went wrong again. The whole of the N1MM logging data was backed up to an external USB drive multiple times each day too.
The multiband antenna initially tuned up close to the desired segments on each band to suit my planned operating frequencies (eg 3610, 7145, 10120, 14195, 18115, 21250, 24950, 28450 and 50110). I did take the time to do a slight tail adjustment on some of the lower bands to get them a little closer given the physical mounting arrangement was lower at the centre and the ends versus the pre-trip setup values at home. The SWR vs frequency sweep function on the IC-7000 made that task quite easy. The LDG Z100 tuner sorted out the SWR so that the transmitter always saw < 2:1 regardless of operating away from the antenna’s actual resonant frequency. I can only recommend the “tail” concept for the fine tuning process because the overall antenna length does not change. The advantage of the trap style is multiband operation but when operating from such a physical location, the shorter overall length becomes a bonus. A full size dipole antenna on 80m would not have fitted where it needed to be placed, and the same probably goes for a full size 40m dipole.
There was no real need to run split frequency on any of the bands. The pile-ups were largely manageable on a single frequency even though some JA’s do not understand USA ONLY. I had the radio set up so that I could run split if needed but it wasn’t necessary.
What went wrong…
Propagation was relatively poor and coupled with the 100W PEP from the IC-7000, some contacts were difficult to finish. QRM in/from USA and Europe would not have helped either way. The other known factor was that the dipole/V antenna was only about 5 metres above ground at the centre and one end was only about one metre above ground. That, coupled with the use of traps which decrease bandwidth and overall efficiency, made the effective transmitted power less than optimal. The physical dipole positioning initially gave lobes SE/NW but was shifted to make it more E/W and give better signals into/from the USA.
The short “longwire” (7m long) and toroidal matching transformer was tried early in the piece but the local noise pickup was worse than on the dipole so it was abandoned and repacked in the luggage straight away. A longer “longwire” plus the pre-arranging for more wire(s) to drape along the ground in lieu of an earth stake would be a good idea.
The one bedroom / combination lounge/dining/kitchen apartment we occupied was the end one in a set of four comprising a stand-alone building. The building also had two large banks of solar panels mounted on the roof (not seen in the Google Earth images) and the two Aurora Inverters were mounted on the outside wall adjacent to the palm tree supporting the centre of the dipole. The antenna placement was thus by no means ideal and a certain amount of hash was heard across 80m and 10m that seems to have come from this array. I initially thought the hash across the whole of the 80, 40, 30 and 20m bands was from the solar power system but it stayed there at night. It only disappeared when the power to the LCD TV and the DVD player was turned off at the wall. I should have tried the longwire again after solving the noise but had no real inclination to do so.
The headset I used was a combination of two different makes of headset, in that I took the microphone stalk off one headset and fitted it to another comfortable set of headphones. The original headset that had the stalk was labelled ‘noise cancelling’ and destined for computer use. In practice, the microphone picked up noises from anywhere in the apartment and even loud noises from outside – often enough to trigger the VOX. I will need to find a better truly “noise cancelling mic insert “ arrangement to improve the rejection of background noises. I also need to revamp the adapter so that I can automatically drop the audio level in the headphones when the voice keyer is running (I feel a MK2 version might be coming along soon…).
Internet access was slow and relatively costly. The wireless hotspot was supposed to be 2MB/s download but it was more like 100KB/s (or less) as web pages were really slow to open. I uploaded the VK9NU logbook to the VK9NU.NET web site at least once a day, usually more often, and it was always quite slow even with only a small file size. The cost of just connecting to the hotspot (let alone logging in) was $35 for 10 hours in minute increments. If you forgot to actually disconnect the wireless connection (like I did once for 3+ hours even though I “logged out”), the minutes meter kept rolling onward.
I made a mistake in letting Windows update run on the Netbook computer and not making sure it was set to OFF well before the trip and re-starting Windows a number of times. The PC took about 3 hours on Monday 25th to finish installing Windows updates (and was therefore unusable) when I went to do a system restart. That meant I couldn’t log - so couldn’t operate.
The apartments were relatively close together and with the almost complete quiet at night/early morning (feral roosters crowing excepted), making almost any noise (like calling CQDX or exchanging reports) would have carried quite a distance. A more suitable arrangement would have been in a building separated further from other tenants.
My XYL Christine was attending the Quilt Norfolk retreat classes for four of the days and I had to drive her there around 8AM and pick her up about 3.30PM on those days and this later time was right in the middle of the best propagation on 20m, 17m or 15m to Europe so I was obliged to QRT for at least half an hour and usually more. We went out to dinner most nights, sometimes tourist-ing as well, so that reduced actual operating time.
What to do better….
..Maybe work harder to avoid the missed QSOs – the partial contacts where a two-way exchange was not completed so the callsign was then wiped from the log. This mostly happened due to the patchy propagation but sometimes from QRM at the other end. Most of the time I could hear signals right down to the low local noise level on 20, 17 and 15m so was able to copy quite a few stations who had issues clearly copying my transmissions.
..More transmitter power and better antennas would improve the contact rate.
..I might take an extra antenna in the form of a squid pole plus a ground mounting method so that I could get a wire truly vertical, again with a nest of wires as a ground plane. (I often heard T32TV working European, US , JA and SA DX and he was using just a vertical wire antenna.) That would give omnidirectional coverage and solve the nulls issue off the ends of the wire dipole.
..Find a DX QTH that can allow larger / full-size dipoles or wire antennas. The better the antenna then the better the DX operation. Couple that with a low-noise environment to get a true winner.