Short answer: The week before a contest is the best time to find problems, because you can still fix them. Plan one short test per day: antennas and feed lines, receiver and transmitter settings, logger and macros, audio and power. Make real contacts where you can, change one thing at a time, write down what you changed and stop making changes two days before the start.
Why a test week works
Most contest disasters are not dramatic. A loose connector, a cable that falls out of a socket, a logger that opens with the wrong contest or a microphone setting that changed after a software update: these are small things, but they appear exactly when the pressure is highest. Every one of them can be found in a calm week. The method is simple. Instead of one big station check, run small tests on several days and let each one answer one question. When something fails, you have time to think, order a part or ask a friend.
The idea also protects you from a common mistake: improving the station right before the contest. A new antenna, a new software version or a new filter introduced on Friday evening has not been tested under real conditions, and it can reduce your score. Fix what is broken, and leave what works.
Monday: antennas and feed lines
Begin with the outdoor part. Walk around the antennas if it is safe. Look for loose guy lines, corroded connections, cracked insulators and cables that rub against metal. Measure the SWR on every band you plan to use and compare it with your notes from the last contest. A change of more than a little is a signal to investigate. If you use remote antenna switches or a band-pass filter, switch through all positions and confirm that each one selects the right antenna.
Check also the antenna direction display if you have a rotator, and compare the displayed heading with a known direction. A wrong zero point costs QSOs in the very contests where beam heading matters. The article on beam heading and distance statistics shows what a clean heading log can tell you after the contest, and it is much easier to read when the heading was right in the first place.
Tuesday: receiver and noise
Listen on each band at the time of day when the contest will be busy. Check the noise level and write it down. If a new noise source has appeared since last time, such as a switching power supply, a charger or a neighbour’s lighting, you now have time to find it. Try the receiver settings that you plan to use in the contest: filter widths, noise blanker, AGC and preamplifier. In a contest the receiver is under much heavier load than in an ordinary evening, so test with strong signals if you can, for example on a weekend with other activity.
If you have a second receiver or a receive-only antenna, test it as well. The article on using a second radio for search and pounce describes how a second receiver is used in practice, and the test week is the time to check that your two receivers behave sensibly together.
Wednesday: transmitter, audio and keying
Test the transmit side. On CW, check the keyer, the weighting and the speed range and listen to your own signal for clicks. On SSB, check the microphone gain, compression and the processor. On RTTY or other data modes, check the audio drive and the clean signal. Ask a friend for a report or use an online receiver nearby if you have one. A report that tells you your signal is wide or distorted is worth more than any manual.
Look at the power and the cooling. A transmitter that runs at full power for many hours needs good ventilation. Hold a hand near the heat sink after a long test transmission, and check the fans. A drop in power after two hours of continuous operation is a typical sign of a heat problem.
Thursday: logger, macros and clock
Now the computer. Start the logger as you will on the contest morning. Open the contest, check the call, the category, the exchange and every macro. Compare the exchange text with the current official rules. Log ten test QSOs and delete them afterwards, or use a separate test database. Export a Cabrillo file and read the header. A trial export shows the problems that normally appear only after the contest, when it is too late.
Check the computer clock against a time service and look at the time zone setting. The article on logging QSO time correctly explains why UTC is the only sensible choice. Update the country file or the call history if your logger needs it, but do it early in the week, so that a new version has time to prove itself. See country file updates for contests for the reasons.
Friday: power, backup and comfort
Look at the power supply and the cables. Is there a spare fuse? Is the extension lead rated for the load? Test the backup: start the logger, make a test QSO and check that the backup copy appears where you expect it. If a laptop is part of your station, check the battery and the power plan, so that the computer does not go to sleep in the middle of the night. A sleeping computer is a common cause of gaps in the log, and the article on time gaps in your log describes the symptoms.
Finally, the human part. Check your chair, the lighting, the water, the food and the plan for rest. A comfortable operator makes fewer mistakes. If you are going to use a headset for many hours, wear it for an hour of the test week and see where it presses.
If there is another contest or a busy weekend before yours, use it as a rehearsal. Operate for a few hours with the real settings, the real macros and the real log file. Observe what annoys you: a key that is hard to reach, a screen that is too small, a sound that distracts. Write it down and fix what is quick to fix. If the rehearsal ends with the whole chain working, you can relax.
Afterwards, put the notes in one place. The article on a station change notebook describes a simple format, and the notes will be useful when you compare your results with last year’s logs.
What to write in the test notes
A test is only useful if the result is recorded. For each day write the date, what you tested, what you found and what you changed. Add the numbers: SWR on each band, noise level at a set time, power output after one hour, the clock offset in seconds. Short lines are enough, and a paper notebook beside the radio works as well as a file. Two weeks later, when a band behaves strangely during the contest, you can look at the notes and see whether the same symptom was there in the test week. The notes also give you an honest list for next year: the things that were repaired this year, the things that were postponed and the things that never gave any trouble.
It helps to rate each item as green, yellow or red. Green means tested and fine, yellow means works but is weak or untested under load, red means broken. Before the contest there should be no red items, and every yellow item should have a plan, such as a spare part or a workaround.
Two days before: stop changing things
Set a firm point in time after which nothing new is installed or updated. Two days before the contest is a good choice. After that point only repairs are allowed, for things that are clearly broken. Everything else waits for the next contest. The rule feels strict, but it prevents the typical pre-contest accident in which an improvement breaks something that worked yesterday.
Make a printed list of the things you have checked and the settings you use. If something goes wrong during the contest, the list helps you return to a known state quickly.
When the contest ends, upload your Cabrillo file to Contest Analyzer to see the statistics and the short list of QSOs worth a second look. The service checks every callsign and every zone against official license databases and several independent reference sources, and it supports year-on-year comparison with your own logs. Compare the test week notes with the report: did the changes you made lead to a better rate, fewer doubtful QSOs or fewer breaks? See the contests page for the list of supported contests and log deadlines.
Related reading
- First CW Contest Tips: What Helps Most
- Contest Log Checklist for the Final Hour
- Setting Up N1MM Logger+ for CQ WW
The bottom line
A test week turns surprises into tasks. Test one area per day, change one thing at a time, rehearse on the air if you can and stop changing things two days before the start. Write everything down, and use the notes after the contest to see what helped.
FAQ
How long should the test week be?
Five to seven days is enough for most stations, with one small test per day. A short week can be compressed, but do not skip the logger and clock tests.
Should I update software in the test week?
Only early in the week and only if you need to. Avoid updates in the last two days before the start.
What if I find a problem on Friday?
Fix it if the repair is simple. If not, choose a workaround you already know and leave the real repair for after the contest.
Do I need a rehearsal for a casual contest?
Even a short evening of real operation helps. It shows whether the logger, the exchange and your comfort are in order.
