Propagation · October 6, 2026

Practical HF Propagation Planning: Tennessee to Chatham Island and Bishkek

A detailed guide to VOACAP, HamClock, and the maps that help you get on the air

HF propagation tools are most useful when they help you choose what to try next. HF means high frequency, the radio spectrum from 3 to 30 megahertz; a megahertz, abbreviated MHz, is one million cycles per second. Propagation means the way a radio signal travels from one place to another. This guide combines long-term predictions, space-weather checks, reports from listening stations, and your own receiver into a practical method for making distant contacts from Tennessee.

The central lesson is a repeatable method: form a testable expectation, plan with VOACAP, read the maps correctly, check current evidence, and test on air. This is a beginner’s how-to guide to making contacts scientifically, including adapting as solar activity declines. Examples teach decisions under contrasting conditions; accurately forecasting a particular event is not their purpose. A green square deserves a listen, not a signed affidavit from the ionosphere.

VOACAP stands for Voice of America Coverage Analysis Program. It was developed for predicting HF radio coverage, originally for Voice of America. It is a propagation prediction program. HamClock is a radio information display that brings maps, clocks, propagation information, and space weather together. These tools serve different purposes, and several other tools supply evidence neither can provide by itself. [1]

This manuscript uses two deliberately demanding paths: Nashville to Chatham Island in Oceania, and Nashville to Bishkek in Kyrgyzstan. Oceania is the geographic region encompassing Australia, New Zealand, and Pacific islands. A path is the geographic route considered between the transmitting and receiving stations. A station is the combination of operator, radio equipment, antenna, and location. The examples do not disclose AF4OZ’s home coordinates or claim to represent the exact installed station.

What the numbers in this guide actually represent

Three kinds of information appear here. Documented tool behavior comes from the primary references listed at the end. Calculated quantities, including bearings, distances, and civil-time conversions, come from the explicit assumptions in this manuscript. Illustrative scenarios are invented teaching exercises showing how to interpret a possible result or respond to a possible observation. Every reliability table in the worked examples is illustrative and is not a VOACAP run.

No live solar index, current opening, actual reception report, or Tennessee-to-destination VOACAP prediction is claimed. The VOACAP website expressly prohibits automated access to its prediction services without prior agreement. Its public documentation and interface were researched; automated prediction requests were not made. A local prediction engine was not installed or run. Readers can reproduce the input recipes manually and replace the teaching values with genuine saved outputs. [1]

Documentation was reviewed for this manuscript on 6 October 2026. Links, service availability, and software versions can change. The HamClock discussion distinguishes the original application and community continuations because the name now covers more than one implementation. This is a complete operating guide and a reviewable manuscript, not a published blog article.

1 Start with an operating question

Ask where when and on which mode

DX is amateur-radio shorthand for a distant station or distant communication. A QSO is a two-way radio contact. A band is an allocated range of radio frequencies, often named by its approximate wavelength in meters. A mode is the way information is carried on the signal. Common modes in this guide are SSB, single sideband voice; CW, keyed Morse code; and FT8, a computer-generated digital mode intended for short exchanges with weak signals. WSPR, pronounced whisper, means Weak Signal Propagation Reporter and is primarily a beacon-like propagation measurement mode rather than an ordinary conversational contact mode.

“Is propagation good?” is too broad to guide the evening. “Can I make a CW contact with a station on Chatham Island between 0000 and 0300 UTC on 20 or 17 meters with 100 watts and a modest antenna?” is a question the tools can help investigate. A watt is the unit of power. UTC means Coordinated Universal Time, the common clock used for radio schedules worldwide; it does not change for daylight saving time.

A useful operating question names the destination, the available time, the band choices, the mode, and the real station limitations. It also names success. Hearing a beacon, receiving an FT8 message, completing a short CW exchange, and holding a reliable voice conversation are different successes. A prediction suitable for one does not automatically promise the others.

Write the question down before opening several colorful maps. Otherwise you can spend an hour admiring the planet and still be unsure where to tune. Maps are excellent at encouraging optimism. Your receiver is excellent at requesting evidence.

Use four layers of evidence

A prediction is a model-based estimate. A model is a mathematical representation of physical behavior with simplifying assumptions. An observation is a measured or reported event. A forecast is an estimate of future conditions. A timestamp identifies when data were measured, issued, or displayed; these can be three different times.

First, use a monthly prediction to select candidate bands and hours. Second, inspect current space weather for reasons the monthly expectation might fail. Space weather means changes in the Sun and Earth’s surrounding magnetic and ionized environment that can affect radio. The solar wind is the flow of particles and embedded magnetic field from the Sun into space. Third, inspect recent reception reports or beacons near the relevant route. Fourth, listen and test from your own station. These layers answer different questions, so agreement is helpful but disagreement is often more informative.

If a monthly chart likes 15 meters but fresh reports favor 20 meters, try 20 first and check 15 briefly. If reports look promising but your receiver is buried in electrical noise, the problem may be local. If both destinations collapse at the same time, check equipment and space weather before blaming the geography. The purpose is to reduce wasted effort while preserving the willingness to try an unexpected opening.

2 The ionosphere without the mystery fog

What bends an HF signal back toward Earth

The ionosphere is the part of the upper atmosphere containing enough free electrons and charged particles to affect radio waves. Ionization is the process that frees electrons from atoms or molecules. Refraction is the gradual bending of a wave as the medium through which it travels changes. HF skywave is radio that reaches distant places by traveling upward into the ionosphere and being refracted back toward Earth. The familiar word “bounce” is convenient, but the ionosphere is not a billiard cushion. [2]

The D region is the lower ionospheric region important for absorption, the loss of radio energy as it passes through ionized air. The E region lies above it and can support some shorter skywave paths. The F region lies higher and is particularly important for long-distance HF; its F1 and F2 subdivisions describe different parts of the daytime electron-density structure. Electron density means the number of free electrons per unit volume. Solar radiation, the time of day, season, latitude, and disturbances all affect that density. [2]

A hop is one skywave trip from the ground to the ionosphere and back. A multi-hop path uses more than one such trip. Each passage through absorbing regions and each interaction with the surface can add loss. Groundwave, by contrast, travels along or close to Earth’s surface and is not the mechanism being planned for these very long examples.

At low frequencies, absorption and noise can dominate. At high frequencies, the ionosphere may not bend the signal back toward Earth sufficiently. Between those limits is a useful operating range. That range moves. The trick is choosing a band within it that your antenna and mode can exploit at the hour you are actually available.

The three frequency limits worth understanding

Open the live VOACAP online

MUF means maximum usable frequency for a specified propagation geometry and conditions. In monthly prediction work it is a statistical quantity, not a fixed ceiling stamped onto the sky. LUF means lowest usable frequency for a specified communication requirement, where absorption and noise make lower frequencies insufficient. FOT, also called optimum working frequency in some references, is a statistical frequency selected to improve dependability below the median MUF; its precise definition depends on the model and product. None of these is a universal frequency for the entire Earth.

Suppose, as a purely conceptual exercise, a route has enough ionospheric support for 18 MHz but daytime loss makes 7 MHz poor. That gives you a reason to investigate 14 and 18 MHz. It does not tell you whether a small antenna produces enough signal, whether the receiver at the far end has high noise, or whether a station is there. Propagation support is one ingredient in communication, not the finished sandwich.

Do not interpret “below MUF” as “guaranteed contact.” Also do not treat a modeled MUF as an absolutely sharp boundary: variability and less common propagation mechanisms can produce reception above a median prediction. Use frequency limits to rank experiments and combine them with signal-to-noise estimates and direct evidence. The VOACAP references explain these distinctions and the relationship between frequency support and circuit reliability. [3]

Day night season and latitude all act together

Daylight supports ionization but also increases ordinary lower-region absorption. Darkness reduces much of that absorption, while the available upper-region support can fall. A winter night can favor a low-band path because of reduced absorption and less thunderstorm noise, yet fail on a high band because support is inadequate. A summer afternoon may favor a high band while being unkind to a weak low-band signal.

These are tendencies to investigate, not calendar commandments. The two example routes have very different geography. Chatham crosses a broad part of the Pacific and connects opposite hemispheres. Bishkek’s shortest route from Nashville passes far north. A hemisphere is half the Earth, here defined relative to the equator. Latitude measures north or south position; longitude measures east or west position. Both are expressed in degrees.

An equinox is the seasonal point when the Sun crosses the equatorial plane and day and night lengths are broadly similar. A solstice marks the Sun’s greatest seasonal north or south displacement. March and September are useful comparison months around the equinoxes, and June and December sample the solstice seasons. Actual monthly predictions should decide which season is promising for your station. “Equinoxes are worth comparing” is useful. “September always works” is an invitation to disappointment.

3 The Tennessee station and the two destinations

Explicit location assumptions

The origin is a representative Nashville location at latitude 36.1627 degrees north and longitude 86.7816 degrees west. In signed decimal coordinates this is +36.1627, −86.7816. Positive latitude means north; negative latitude means south. Positive longitude means east; negative longitude means west. Decimal coordinates avoid ambiguity when entering values into tools.

The Oceania destination is a representative point on Chatham Island at −43.9500, −176.5500. Chatham Island is part of New Zealand and is a demanding Pacific target for an inland Tennessee station. This coordinate is a planning point, not a claim about any particular operator’s antenna site. If you are chasing a specific visiting operation, substitute its actual coordinates when available. DXpedition means a temporary radio operation organized to activate a desirable or uncommon destination. DXCC is the amateur-radio awards system that recognizes contacts with defined entities, which do not always correspond one-for-one with sovereign countries.

The Central Asian destination is central Bishkek, Kyrgyzstan, at +42.8746, +74.5698. Kyrgyzstan is the country; Bishkek is the example receiving area. A station outside the city, especially in mountainous terrain, can have a very different horizon and noise environment. A country-center default in a tool is therefore useful for a quick look but is not a substitute for the operating site.

These numbers are reproducible scenario assumptions. Their displayed precision supports consistent entry; it does not imply that propagation can be predicted to the distance between neighboring houses.

Calculated path geometry

A great circle is a circle on a spherical Earth whose plane passes through Earth’s center. The shorter arc along it is the short path, abbreviated SP. The complementary arc is the long path, abbreviated LP. Azimuth or bearing is the horizontal direction, measured clockwise from north. True north means geographic north; magnetic north is the direction indicated by a compass and differs by a location-dependent magnetic declination, the angle between the two north directions.

The following values were calculated for a spherical Earth with radius 6371 km, using the coordinates above. Kilometers, abbreviated km, are units of distance. They are geographic calculations, not ionospheric predictions. Distances are rounded to the nearest 10 km and bearings to the nearest degree because that is more appropriate for practical station aiming than excessive decimal places.

Geographic calculations from the stated coordinates

The Chatham short-path midpoint is approximately 5.5 degrees south, 128.4 degrees west, in the eastern Pacific. The Bishkek short-path midpoint is approximately 78.4 degrees north, 22.5 degrees west. A midpoint is the halfway position along the selected arc. Bishkek looks like an Asian destination on a wall map, but its short path from Nashville is a high-latitude route. High latitude here means far from the equator; geomagnetic latitude, introduced later, is a related but different coordinate tied to Earth’s magnetic field.

The return bearing is computed at the destination. It is not simply Nashville’s bearing plus 180 degrees, because true-north directions change along a curved great circle. At a given starting point the LP departure bearing is 180 degrees opposite the SP departure bearing. Chatham’s 232-degree SP becomes 52-degree LP; Bishkek’s 14-degree SP becomes 194-degree LP. Do not confuse that local opposition with the far-end return bearing.

For reproducibility, compute central angle c from cos(c) = sin(latitude1) × sin(latitude2) + cos(latitude1) × cos(latitude2) × cos(longitude2 − longitude1). Use radians inside trigonometric functions; a radian is an angular unit in which a full circle is 2π. SP distance is 6371 × c. LP distance is 6371 × (2π − c). Initial bearing is atan2(sin(Δlongitude) × cos(latitude2), cos(latitude1) × sin(latitude2) − sin(latitude1) × cos(latitude2) × cos(Δlongitude)), converted to degrees and normalized into 0–360. atan2 is the two-input inverse-tangent function that preserves the direction’s quadrant. These equations are enough to reproduce the rounded numbers.

Why a flat map can fool your antenna

A map projection is the method used to show a curved Earth on a flat display. A rectangular world map can make the shortest route look bent and stretch polar regions severely. An azimuthal equidistant map centered on Nashville shows directions from Nashville as radial bearings and preserves distance from the center. It is a good companion to a rotatable antenna. It does not preserve every area or shape.

Trace the complete route, including any portion that crosses the edge of a rectangular map. A line ending at the left edge and reappearing at the right edge is one continuous path. The International Date Line is the conventional boundary across which civil dates change; it is not a radio barrier. A radio signal does not stop at the date line to adjust its watch.

Antenna aiming starts with the calculated bearing. It can then be tested a little to either side because real propagation can deviate from ideal great-circle geometry. Do not call every off-bearing peak long path. Long path is a specific complementary route, and identifying it requires stronger evidence than a lucky turn of the knob.

4 Time conversions that do not steal your opening

Nashville does not share one time zone with all Tennessee

For this scenario Nashville uses the America/Chicago civil-time zone: Central Standard Time, CST, UTC−6 in winter, and Central Daylight Time, CDT, UTC−5 during daylight saving time. Parts of Tennessee use Eastern time. If your actual station is in eastern Tennessee, convert for that location rather than borrowing Nashville’s clock.

Bishkek uses Asia/Bishkek, UTC+6 for these example dates. Chatham uses Pacific/Chatham, with a 45-minute component: UTC+12:45 in standard time and UTC+13:45 during daylight saving time. Chatham does not use the same offset as mainland New Zealand. The dates below are specific examples calculated using these named civil-time zones, not a rule for every date forever.

Exact civil-time conversions for selected 2026 examples

Date rollover is the change to the preceding or following civil date when applying an offset. The first row puts Nashville on the previous date. The fifth puts Chatham on the next date. Write both the date and UTC hour in a schedule; “Tuesday evening” is not sufficiently precise when the other operator is already having Wednesday breakfast.

A worked window conversion

For a Chatham listening session from 0000 to 0300 UTC on 15 September 2026, Nashville time is 1900 to 2200 CDT on 14 September. Chatham time is 1245 to 1545 on 15 September. The session is Tennessee evening and Chatham afternoon. That is civil-time arithmetic. It does not establish a band opening or the position of every daylight segment along the route.

For Bishkek from 1200 to 1500 UTC on 15 December, Nashville time is 0600 to 0900 CST on 15 December and Bishkek time is 1800 to 2100 that day. The session straddles Tennessee’s morning and Bishkek’s evening. Determine sunrise, sunset, and path illumination with an astronomical tool for the exact date before treating it as a grayline opportunity.

Grayline is the twilight region around the terminator, the boundary between Earth’s day and night sides. Civil sunrise and sunset occur at the surface; the ionosphere is illuminated at altitude and does not switch state at exactly the same moment. Thus a grayline drawing is a useful timing aid, not a universal narrow corridor that automatically carries every HF band.

Local mean solar time is a clock tied to the Sun’s average position at a longitude. It is different from civil time, which follows legal time-zone rules. Desktop VOACAP’s local-mean-time option should not be mistaken for the time on the kitchen clock. Use UTC throughout prediction records and convert for convenience afterward. [4]

5 Build a station model that resembles your station

Separate the physical station from the software approximation

An antenna is the structure that converts electrical radio energy into an electromagnetic wave, and converts an arriving wave back into electrical energy. A dipole is an antenna with two conductive arms, commonly about half a wavelength overall. A vertical is an antenna with a substantial vertical radiating element; its return system and ground environment matter. A Yagi is a directional antenna with a driven element and additional elements that shape its radiation pattern. A radiation pattern describes how strongly an antenna transmits or receives in different directions and elevation angles.

Elevation angle is the angle above the horizon. Takeoff angle, often TOA, is the elevation at which transmitted energy leaves the antenna toward a modeled path. Antenna height above ground, abbreviated AGL, helps determine a horizontal antenna’s elevation pattern. Height in wavelengths matters: 10 meters AGL is approximately half a wavelength on 20 meters but about a quarter wavelength on 40 meters. The same installation can behave very differently on different bands.

An isotropic antenna is an ideal reference that radiates equally in all directions. It is useful for a controlled comparison but is not a backyard antenna. Antenna gain is the concentration of radiation in a direction relative to a reference. dBi means gain in decibels relative to an isotropic antenna. A decibel, abbreviated dB, expresses a ratio logarithmically; for power it is 10 × log10(power ratio). Zero dBi is a reference assumption, not “no signal.”

Do not select the best-looking antenna from a menu simply because the resulting chart becomes greener. That changes the question being asked. If your wire hangs low, a prediction using a tall Yagi describes someone else’s station. Someone else may enjoy it very much, but the chart will not erect a tower for you.

The controlled baseline for the worked exercises

Use 100 W at the modeled transmit antenna feedpoint, the point at which the feed line connects to the antenna. A feed line carries radio energy between radio and antenna; coaxial cable, or coax, is a common type. Feed-line loss is power dissipated along that connection. The baseline uses ideal 0 dBi isotropic antennas at both ends so that changes in time, band, and mode can be compared without changing the antenna. This is a reference calculation and is not a realistic efficiency claim for a Tennessee vertical.

Use a minimum TOA of 3 degrees for the reference sweep. Minimum TOA is the lowest elevation angle the model is allowed to use. Three degrees is a starting assumption, not a statement that your site has a clear 3-degree horizon. Run an additional case at 10 degrees to see how sensitive the answer is to low-angle access. A hill, trees, roof, surrounding soil, or nearby structures may make the reference overly optimistic.

Use a residential receive-noise category where the tool offers that label. Receive noise means unwanted energy entering the receiver along with the desired signal. The origin of this noise can be atmospheric, equipment-generated, or internal to the receiver. If reproducing the desktop numerical convention, record the residential input 145 and confirm the output noise reference corresponds to −145 dBW at 3 MHz. dBW means decibels relative to one watt. The sign and reference matter; never type −145 blindly into a field expecting the positive code 145. [4]

Use the interface’s CW preset for the first sweep, then its SSB and FT8 presets separately. A preset is a stored collection of mode-dependent assumptions. Record the required SNR if the interface exposes it. SNR means signal-to-noise ratio: desired signal power divided by noise power, usually expressed in dB. For desktop CW and SSB reproductions, 24 and 45 dB-Hz respectively are explicit demonstration thresholds consistent with the input documentation. dB-Hz is a signal-to-noise quantity referenced to a 1 Hz noise bandwidth; it is not the same number as a receiver’s SNR measured across several kilohertz. Bandwidth is the frequency width over which a receiver or decoder accepts energy. [4]

Use monthly smoothed sunspot number, SSN, sensitivity cases of 30, 80, and 140. A sunspot is a dark solar region associated with concentrated magnetic activity; sunspot number is a standardized numerical index of this activity. Smoothed SSN is a time-averaged monthly index used by the model. These three values are hypothetical low, middle, and high input cases, not a forecast for any 2026 month. Disable dynamic SSN, meaning an automatically substituted solar input, when the interface allows it, and verify the entered number appears in the output. [5]

Add a realistic antenna case after the baseline

For a modest home-station comparison, rerun with a horizontal dipole at 10 m AGL at each end if a matching pattern is available. Record the precise pattern name and orientation. Broadside means perpendicular to the wire’s length, the direction of strongest radiation for a simple dipole in its usual free-space pattern. Align the model’s principal beam toward the selected path and check the software’s convention: an azimuth field may describe beam direction or physical wire orientation, depending on the antenna file.

For a stronger station comparison, use an available three-element Yagi pattern at 15 m AGL on the bands where that is realistic. Do not assume a 20-meter Yagi also provides a valid 40-meter pattern. Keep a separate antenna entry for each band or record exactly what the tool substitutes. A vertical case can be useful if its ground and efficiency assumptions are known. Efficiency is the fraction of accepted antenna power that becomes radiated power; remaining power is lost as heat. A nominal antenna category does not capture every radial layout, soil condition, or installation problem.

An impedance is the electrical relationship between voltage and current, including resistance and reactance. Resistance describes the in-phase part; reactance describes the energy-storing part. An antenna tuner adjusts the impedance presented to the transmitter. SWR, standing wave ratio, describes mismatch on a transmission line. A low SWR does not prove high radiation efficiency or a useful low-angle pattern. A dummy load, a device designed to absorb transmitter power as heat, can have wonderful SWR and an impressively short DX log.

Use the baseline to understand the model, then the realistic pattern to decide what your hardware might do. Treat a custom antenna model as another assumption requiring validation, especially if its dimensions, height, ground treatment, or orientation are uncertain. The primary antenna-pattern reference explains how inspecting the pattern can reveal errors hidden by attractive circuit outputs. [6]

Power and noise arithmetic you can check yourself

With 1 dB feed-line loss, 100 W at the radio becomes 100 × 10^(−1/10), or 79.4 W, at the antenna. With 2 dB loss it becomes 63.1 W. If software power is explicitly the feedpoint value, enter the reduced power. If it already includes your feed-line loss elsewhere, do not subtract it twice. In desktop software expecting kilowatts, kW, divide watts by 1000: 100 W is 0.100 kW and 5 W is 0.005 kW. [5]

Changing from 100 W to 400 W gives 10 × log10(400/100) = 6.02 dB of additional transmitted power. Changing from 100 W to 5 W loses 13.01 dB. These are real mathematical changes under the stated power assumptions. They do not mean reliability rises by 6 percentage points or falls by 13 percentage points. Reliability depends on the whole distribution of modeled signal and noise.

For an illustrative receiver example, suppose desired signal power is −120 dBm and noise power in the same bandwidth is −125 dBm. dBm means decibels relative to one milliwatt, and a milliwatt is one thousandth of a watt. SNR is −120 − (−125) = +5 dB. If local noise rises to −117 dBm while the signal stays −120, SNR becomes −3 dB. The transmitter has done nothing wrong; the receiving environment has taken away 8 dB.

Because dBm = dBW + 30, a predicted signal of −150 dBW is −120 dBm. Always match bandwidths before comparing SNR values. To convert a 1 Hz reference into a bandwidth B in Hz, subtract 10 × log10(B) from the normalized SNR. For B = 2500 Hz, that subtraction is about 34 dB. This is a normalization exercise, not a universal FT8 decoding threshold. Use each decoder’s own specification and the software’s preset.

6 Set up VOACAP deliberately

What VOACAP contributes

VOACAP provides statistical HF circuit estimates for specified locations, month, solar input, antennas, power, noise, and communication requirement. A circuit is the complete modeled communication link. Its monthly behavior is useful for planning repeated opportunities, but it does not predict the exact minute of a particular day’s opening. An output is only as meaningful as the assumptions that produced it. [7]

Use the current VOACAP Online interface at https://www.voacap.com/hf/ and the associated manual. The inspected interface offers point-to-point analyses, including band-by-band, full analysis, all-year views, and frequency comparisons, plus coverage maps. Point-to-point means one specified transmitter and one specified receiver. A coverage map repeats an analysis over many receiving locations. The date on a day/night display should not be mistaken for a daily ionospheric forecast. [8, 9]

A reproducible manual input procedure

1. Enter the Nashville coordinates directly as the transmitter, TX, and the destination coordinates as the receiver, RX. TX and RX are abbreviations for transmitter and receiver. Check the displayed hemisphere signs and the resulting distance and bearing against this manuscript. A misplaced longitude sign can send your planning session on an entirely different vacation.

2. Select March, June, September, or December for the first seasonal sweep. Use the selected year for labeling and date display, but do not infer day-specific precision from it. In desktop VOACAP select the month without a day component; the desktop input guide explains that entering a day can alter the coefficient selection. Coefficients are stored numerical values used by the ionospheric model. [4]

3. Enter one of the hypothetical monthly SSN values 30, 80, or 140, disable automatic replacement where available, and record what is actually used. When preparing a real operational forecast, instead use the monthly series recommended for your engine and record its source, issue date, and scale. The daily sunspot count and the 10.7 cm solar flux value are not interchangeable with this input. [5]

4. Select the stated baseline antennas, receive-noise category, 100 W feedpoint power, 3-degree minimum TOA, and CW preset. In desktop work use the documented CCIR coefficient option, the original coefficient set named for the international radio consultative organization. If a method selector is exposed for these greater-than-7000-km paths, record Method 30, the documented smoothing method for long circuits. This method selection is separate from choosing geographic SP or LP. [4]

5. Select short path first. Examine all 24 UTC hours and all available amateur bands from 80 through 10 meters. A 24-hour sweep is safer than choosing only the hour you hope will work. Save the input screen and output with the case label, for example Nashville Chatham September SSN80 CW100 SP reference. That label is your audit trail, a record allowing someone else to repeat what you did.

6. Repeat for long path without changing the other assumptions. Then repeat the mode, power, antenna, and horizon cases one variable at a time. A sensitivity test is a rerun that changes an assumption to see how strongly the conclusion depends on it. When several inputs change simultaneously, the result may improve but you cannot tell which change helped.

7. Swap TX and RX for the reverse direction, carry the correct transmitting power and antenna to each end, and assign the correct receiver-noise environment. Do not simply assume equal reliability in both directions. Geographic reciprocity does not ensure equal communication performance when the two receivers experience different noise or the two transmitters use different power. Reciprocity means a relationship that is unchanged when its direction is reversed under appropriate physical conditions.

8. Save the numerical output when available. In a final operational record include tool and engine version, month, SSN, path choice, antennas by band, orientations, power, noise, TOA, mode, required SNR, output units, UTC hour, and generation date. A screenshot with only colored cells is incomplete evidence.

The steps above are this guide’s original reproducibility protocol. Exact menu labels can change. Check the current interface and its own help instead of treating a button location in an old screenshot as permanent.

A compact run matrix

Cases to run manually for each destination

This matrix produces many cases. Start with season and solar sweeps, narrow to plausible hours, then run detailed station comparisons. A full experiment should serve an operating choice rather than generate a folder full of colorful rectangles you never consult.

7 Read VOACAP outputs without awarding yourself a contact

Reliability is conditional

REL means modeled circuit reliability for the required SNR. A decimal result of 0.70 means 70 percent. It describes the statistical fraction of days in the modeled month on which the specified hour and communication requirement are expected to be supported. In a 30-day conceptual month, 70 percent corresponds to roughly 21 qualifying days. It does not specify which days, whether an operator will be present, or whether you will win a pileup, the crowd of stations calling the same rare station. [10]

Required reliability is an input design target. Computed REL is an output. Entering a required reliability of 90 percent does not command the atmosphere to deliver it. A median is the middle value of a distribution, with half the modeled values above and half below. A percentile locates a specified proportion of that distribution. A median SNR says less about dependable communication than the lower portion of the SNR distribution relevant to your target reliability. [7]

Read the requested mode and threshold before reading the percentage. A good FT8 reliability estimate does not automatically describe good SSB reliability. FT8 can carry a compact exchange under conditions that make comfortable voice communication difficult; that advantage is a property of the communication system, not a change in ionospheric refraction.

Signal power is not readability

S DBW or SDBW denotes modeled received signal power in dBW. N DBW is modeled received noise power in dBW. If the references and bandwidth conventions match, subtracting noise from signal gives a signal-to-noise measure. A loud signal in louder noise can be less readable than a weak signal against a quiet background. Signal strength is therefore only part of the analysis.

An S-meter is a receiver’s signal-strength indicator, commonly marked in S units. Do not treat a conversion from modeled power to an S-unit label as a precision prediction for your radio. Receiver calibration, antenna mismatch, gain settings, and noise matter. For comparisons, relative changes under controlled settings are often more useful than a claim that a distant operator will be exactly S5. The VOACAP S-meter discussion explicitly notes practical calibration limitations. [11]

MUFday and reliability answer different questions

MUFday is the modeled fraction of days with ionospheric support at a specified frequency for the relevant propagation mode. REL also considers whether the communication requirement is met. A teaching example with MUFday 0.90 and REL 0.35 would suggest that support is common but adequate signal-to-noise is less common under the station assumptions. More power, a better useful-angle pattern, a quieter receiver, or a less demanding mode might matter. Lowering frequency is not the only possible response. [3]

In VOACAP output, MODE can refer to a propagation mode, a modeled sequence of hops or ionospheric interactions, rather than the operating mode CW or SSB. TANGLE is the modeled transmit elevation angle, RANGLE the receive elevation angle, DELAY a travel-time quantity, and LOSS a propagation-loss quantity. Read their units from the engine output. These are engineering clues, and TOA estimates deserve particular caution rather than being used as a precise construction prescription. [12]

Read chart axes and legends before the colors

On an hour-by-band chart, identify which axis is UTC hour and which is band or frequency. Check whether time runs midnight to midnight or rolls forward from now. Read the legend’s numerical limits and determine whether cells show reliability, SNR, or power. Those are three different quantities. A blank cell may mean no computed value, a below-threshold result, or a display issue; consult the tool’s legend rather than inventing a meaning.

For a color-coded wheel, identify where the UTC-hour sectors begin and which rings correspond to bands. Verify SP or LP. A wheel can summarize the same type of information as a rectangular chart, but attractive radial geometry does not create additional evidence. When the display is too compressed to read accurately, use a band-by-band or numerical view.

For a coverage map, identify the selected band, hour, transmitter, receiving antenna, mode, power, and color scale. A bright patch shows the model’s value for hypothetical receivers in that area. It does not show all currently transmitting stations. A signal-strength map cannot be read using a reliability legend. A percentage legend cannot be read as an S-meter.

When a chart shows a best-frequency recommendation, inspect neighboring legal amateur bands and their reliability rather than transmitting on an arbitrary recommended frequency. A model knows physics and its inputs; your license privileges, regional band plans, interference, and the target’s actual operating frequency determine where you can make a contact. This guide’s rounded calculation frequencies are analysis frequencies, not universal operating instructions.

8 Use HamClock as a situation display

Identify which HamClock you have

The original HamClock was created by Elwood Downey, WB0OEW. Community projects now continue the software and data services. The hamclock.com continuation reports that the original Clear Sky Institute data server is offline; the Open HamClock Backend project separately publishes client code and current release information. A backend is the remote service supplying data or calculations to an application. Record your client version and configured backend before comparing screenshots or relying on a troubleshooting guide. [13, 14]

The Open HamClock Backend client status page listed version 4.32 dated 26 September 2026 when researched. That identifies one project’s release; it is not a guarantee that your device runs it or that every application called HamClock behaves the same way. A browser dashboard with a similar name, a historical classic build, and a community client can have different controls and assumptions. [14]

This guide could not retrieve the current downloadable client manual through the research tools. Therefore it does not claim to have verified every 4.32 menu or gesture. Consult the current manual linked from the project’s status page. The workflow below deliberately relies on checking visible labels, selected values, and data age rather than memorizing a particular tap position.

Establish DE and DX before studying the panel

Open the live HamClock

DE is radio shorthand for “from” and marks your home location in this context. DX identifies the selected distant location. Set DE to the Nashville scenario and DX to Chatham, then repeat with Bishkek. Confirm the geographic marker and bearing. A call sign, the identifier assigned to a licensed radio station, may help locate a station, but a call-sign lookup does not always identify its current portable or expedition site.

Find the propagation panel for the DE-to-DX circuit. Check power, operating mode, path selection, TOA where available, and the solar input used. A reliability estimate based on generic antennas is useful for comparing bands but may differ from your detailed VOACAP station model. Matching the station settings is necessary before judging whether two tools disagree. The Open HamClock Backend VOACAP service identifies its output as VOACAP-based band reliability information supplied to compatible clients. [15]

Set the time-axis display to UTC for the main comparison and keep a local-time clock visible separately. Verify whether the displayed hours are absolute hours, local DE hours, or elapsed hours from now. Read the actual cells or scale rather than assuming that the far-left column is midnight.

Make the display answer a small set of questions

Keep visible the destination location and bearing, the relevant propagation panel, UTC, and a space-weather display with timestamps. Add day/night shading when inspecting the route. If the panel rotates among several products, pause or select the view you need so that you can inspect its legend calmly. A pane is one region of the display dedicated to a product or chart.

If your map offers both reliability and TOA layers, verify which is active. Reliability answers a communication-performance question. A TOA map shows a modeled angular requirement or selected angle. It cannot be read as a probability map simply because the same band is selected.

If a HamClock build uses traffic-light reliability categories, use the legend supplied by that build. Do not transplant fixed numerical color thresholds from an old manual onto a different client. The practical response is stable even when colors change: identify the higher-ranked hours and bands, then validate them with current observations.

Check freshness before diagnosing a bad band

A live-looking clock is not proof that all its data are live. The clock can advance while a solar panel or propagation product is cached. Cached means previously saved data reused without a fresh fetch. Compare the product timestamp with current UTC and the expected update interval. If a panel has no obvious time, consult its status or help and compare with the original provider’s page.

If the application stops updating, verify connectivity and the configured service before interpreting stale values. The continuation’s recovery instructions apply to supported builds and may differ for older hardware. This manuscript does not instruct you to run installation commands from an unknown page; it directs you to the project’s own version-specific guidance. [13]

Keep the dashboard’s role modest but valuable. It can keep you oriented while you operate, alert you to changed conditions, and make you revisit a band. It cannot hear the electrical noise in your room, the operator’s schedule on Chatham, or the contest crowd calling Bishkek.

9 Space weather that matters to these paths

Separate solar support from magnetic disturbance

SFI means solar flux index, usually the Sun’s 10.7 cm radio flux expressed in solar flux units. A solar flux unit, SFU, is 10^−22 watts per square meter per hertz. A hertz, Hz, is one cycle per second. The index is a proxy, an indirectly related measurement, for aspects of solar activity that influence ionization. It is not the operating frequency of an HF contact and not the monthly SSN input used in the preceding recipes. [16]

Kp is a planetary geomagnetic activity index describing disturbance over three-hour intervals on a scale from 0 to 9. Geomagnetic means related to Earth’s magnetic field. Ap is a related linearized planetary activity measure, and an A index often describes activity summarized over a day. Linearized means transformed onto a scale that more directly represents disturbance amplitude than the quasi-logarithmic K scale. Read whether a display shows estimated, observed, or forecast Kp. NOAA’s estimated planetary product is assembled from ground magnetic observations. [17]

An illustrative combination of SSN140 in a monthly run and current Kp6 would mean “strong modeled solar-support assumption, disturbed present magnetic environment.” It would not mean “140 plus 6 equals good DX.” Keep the two pieces separate. A large solar-support index can coexist with poor HF communication during a disturbance.

Geomagnetic latitude refers to position relative to the magnetic field rather than geographic latitude. An auroral zone is the magnetic-latitude region associated with auroral activity and related particle effects. The Bishkek short path’s far-northern geography is a reason to watch high-latitude conditions, although the calculated 78.4-degree geographic midpoint is not itself an auroral-zone measurement.

Flares and proton events need different reactions

A solar flare is a rapid release of solar electromagnetic energy. Extra X rays can increase D-region absorption on the sunlit side of Earth quickly. NOAA uses R1 through R5 to describe radio-blackout severity. X-ray flare classes A, B, C, M, and X are successively stronger categories; the number after the letter scales the level within the class. M1 corresponds to the R1 threshold and X1 to R3 in the NOAA scale. [18]

A solar energetic particle event, SEP event, involves high-energy particles that can increase polar absorption. A proton is a positively charged particle, one of the constituents of atomic nuclei. NOAA’s S scale describes solar radiation storms; its G scale describes geomagnetic storms. R, S, and G therefore label different effects. A CME, coronal mass ejection, is a large expulsion of solar plasma and magnetic field; plasma is ionized matter. Its arrival and magnetic orientation can matter later, rather than being identical to the immediate X-ray effect of a flare. [19]

If a hypothetical flare occurs while a large part of the Chatham route is sunlit, check an absorption product and actual signals before continuing a weak low-band test. If a hypothetical proton event affects polar regions, the Bishkek SP case deserves particular scrutiny. Trying LP may be an experiment if the target is heard there, but a much longer route does not automatically escape every disturbed segment.

How to read NOAA D RAP correctly

Open the live NOAA D-RAP

D-RAP means D-Region Absorption Prediction. NOAA’s global highest-affected-frequency map shows the frequency associated with 1 dB modeled absorption; its polar highest-affected-frequency views use a 10 dB criterion. Fixed-frequency maps instead display estimated absorption at the named frequency. These are different products and legends. Check which is selected. The product uses X-ray and proton measurements as inputs. [20]

If a teaching example puts a route segment in a 20 MHz region on the global highest-affected-frequency product, interpret that as the stated absorption criterion extending up to approximately that frequency in the displayed model. Do not interpret it as a 20 MHz MUF, a guaranteed 20 dB loss, or a declaration that 21 MHz is fully usable. If a fixed-frequency view says 5 dB at 15 MHz, the unit is attenuation, the amount of signal reduction, not probability.

One dB of modeled absorption is not a binary blackout. A strong signal may survive it; a marginal one may not. Real slant paths, the oblique routes through the atmosphere, and repeated traversals can differ from a simple vertical-map value. Do not add a few map colors together and claim a rigorous total multi-hop loss. Use the map to recognize affected regions and increase caution, then examine reception.

A practical storm decision

For a hypothetical Bishkek morning session, assume the monthly chart ranks 17 meters well but current Kp is elevated and northern reception reports have thinned. This is an illustrative situation, not a report for today. Begin by listening on the target’s announced band. If there is no target, test 20 and 30 meters, check recent reports from relevant regions, and shorten the session if several independent clues remain poor.

Do not mechanically drop to 40 meters whenever geomagnetic conditions worsen. Lower frequency can improve ionospheric support but can also increase absorption and atmospheric noise. Atmospheric noise is unwanted radio energy from natural processes, especially lightning. The best response may be a different hour, a quieter receiving setup, another destination, or simply a short reconnaissance rather than a prolonged pileup attempt.

10 Other tools that supply different kinds of evidence

KC2G ionosonde maps

Open the live KC2G ionosonde maps

An ionosonde is an instrument that probes the ionosphere with radio signals and measures its response. KC2G’s map combines ionosonde-derived information with interpolation, the estimation of values between measurement points. Its MUF map is labeled MUF for a 3000 km path. Its foF2 map displays the F2 critical frequency, the limiting frequency for vertical incidence through the F2 region. Vertical incidence means a signal directed straight upward. These are distinct quantities. [21]

A contour is a line connecting locations with an equal plotted value. Suppose, illustratively, the Pacific map region you are inspecting lies between contours labeled 15 and 17 MHz. Its displayed estimate is around 16 MHz. That is a reason to investigate 20 meters at 14 MHz before trusting 17 meters at 18 MHz. It is not a complete model of the 12680 km Tennessee-to-Chatham circuit.

The provider explains that actual station measurements appear as numbered dots and other areas are estimated; faded dots have lower confidence. It describes map production at 15-minute intervals using measurements that may already be several minutes old. Read the actual header timestamp. Coverage gaps and changing input stations can alter the map, particularly far from measurements. [21]

For Bishkek, do not stare only at the Nashville dot. Trace the northern route and notice sparse measurements. For Chatham, large ocean areas make unsupported precision particularly tempting. Combine the map with the route and actual signals, and resist turning an interpolated pixel into a guarantee.

PSK Reporter

Open the live PSK Reporter

PSK Reporter collects reception reports submitted by participating decoding software. A decoder is software or circuitry that extracts information from a received signal. A reception report is evidence that a receiver decoded or reported a transmitter at a time and frequency. The map is useful for investigating who hears your signal or who is being heard near your area. Read sender, receiver, mode, frequency, report time, and location together. [22]

For a practical outbound test, filter for your own call sign as transmitter, select FT8 and the band being tested, and use a short recent interval such as 15 minutes. Transmit normal permitted CQ calls on a clear frequency, then allow reporting time. CQ is a general invitation for any station to reply. The goal is not to flood the band; it is to obtain a small, interpretable sample.

For an inbound test, listen and decode, then inspect stations your receiver heard. For a regional comparison, inspect recent reports from stations near Nashville, but record that their antennas, power, and local noise are unknown. A Tennessee station with a high Yagi is encouraging evidence, not proof that a low wire at your house will do the same.

Do not assume a map line is a two-way contact. Also do not assume no dots means no propagation. There may be no receiver at the destination, no station transmitting, a reporting delay, an incorrect filter, or a software outage. Check the map’s own legend; marker colors and display options can be customized, so this guide does not assign universal colors to transmitter and receiver dots.

Reverse Beacon Network

The Reverse Beacon Network, RBN, consists of receiving stations that report signals they hear. It is particularly useful for checking whether your CQ reaches distant listeners and for comparing recent signals. A spot is a report containing a call sign and frequency, often with time and signal information. It is evidence of reception by the reporting node, a participating receiving station, rather than proof that the target operator can hear you. [23]

For CW tests, use clean CQ transmissions with several repetitions of your actual call. The RBN’s own operating guidance explains that recognition depends on the call pattern and repeated decoding. Keep speed and spacing consistent. A memory keyer is a device or program that sends stored Morse text; it can improve consistency for a controlled comparison. [24]

If you compare two antennas, use the same receiving node, band, approximate time, and transmit power, alternating in an A B A B sequence. Fading, time-varying changes in signal strength, can make a single comparison misleading. Record the reported SNR convention and do not compare numbers from unrelated receivers as though they were one calibrated instrument.

NCDXF and IARU international beacons

NCDXF is the Northern California DX Foundation and IARU is the International Amateur Radio Union. Their International Beacon Project supplies scheduled CW transmissions on 14.100, 18.110, 21.150, 24.930, and 28.200 MHz. A beacon is a station transmitting a known signal to help assess propagation. The network provides regional clues even when conversational activity is sparse. Check the current schedule and status before concluding that an unheard beacon proves a closed band. [25]

Each scheduled transmission contains a call sign followed by power steps: 100 W, 10 W, 1 W, and 0.1 W. Successive levels differ by 10 dB. If you hear the 1 W segment but not the 0.1 W segment, you have evidence of useful sensitivity on that beacon’s path at that moment, not a calibrated 10 dB margin for every station in its region. [26]

New Zealand or Australian beacon reception is a useful Pacific clue for Chatham, but Chatham is a different location with a different route. A Russian or Asian beacon is similarly an imperfect proxy for Bishkek. A proxy station is a nearby or route-related station used as indirect evidence. No single beacon represents all of Oceania or Central Asia.

WSPR and your own operating log

WSPR reports can reveal weak-signal paths and useful antenna patterns, but a WSPR reception does not establish voice readability or a two-way QSO. Its narrow transmission and long integration, the accumulation of information over time, are suited to measuring weak signals. Look for repeated receptions on the relevant route and compare identical station configurations before drawing conclusions. The primary WSJT-X documentation is linked as a mode reference; exact decoder thresholds should be taken from the version you run. [27]

Your log is an equally important tool. A log is a record of station activity and contacts. Include failed tests, bands monitored, received signals, local noise, antenna, power, mode, UTC, path choice, and weather-product timestamps. A log containing only successes makes the station look wonderfully capable but conceals how often the preferred window failed.

After a month, compare predictions with repeated observations for the same path and configuration. Do not declare a model accurate because one opening matched or useless because one did not. Accumulated evidence can show a persistent local noise problem, a seasonal opportunity, or an antenna pattern limitation that a generic prediction does not describe.

11 Worked example one Nashville to Chatham Island

Define a realistic goal and a controlled recipe

The goal is a short two-way CW contact with a hypothetical active Chatham station during a three-hour Tennessee evening session. This is a planning exercise, not an announced expedition or an assertion that any station will be available. Use Nashville +36.1627, −86.7816 and Chatham −43.9500, −176.5500. SP is approximately 12680 km at 232 degrees from Nashville. LP is approximately 27350 km at 52 degrees.

The first reproducible case is September, hypothetical monthly SSN80, CW preset, 100 W at the antenna feedpoint, 0 dBi reference antennas at both ends, residential receive-noise category, minimum TOA 3 degrees, SP. Inspect all 24 UTC hours before narrowing to 0000–0300 UTC. For the dated 15 September 2026 session, that selected window is 1900–2200 CDT on 14 September in Nashville and 1245–1545 on 15 September on Chatham. The dates support time conversion only; the prediction recipe is monthly.

The physical follow-up station case is a dipole at 10 m AGL at both ends using a verified available pattern directed toward the path, with the same mode, feedpoint power, and noise setting. If the target’s station details are unknown, leave the reference receiver as an assumption and run alternatives rather than describing an invented target antenna as fact. At a real expedition, station equipment may be published; use it and save the source.

Start with 20 meters, inspect 17 and 15 as higher-frequency candidates, and keep 30 and 40 as lower-frequency candidates at other hours. This is a sensible reconnaissance order to test with a real run and observations, not a claim about September’s computed optimum. The 10- and 12-meter bands deserve attention in a stronger solar-input sweep. Eighty meters is a demanding long-distance exercise with greater antenna and noise requirements; do not let its presence on a chart imply easy coverage with the baseline antenna.

Interpret an illustrative chart fully

Open the live VOACAP online

The following values are fabricated solely for teaching. They have not been generated by VOACAP or read from HamClock. They describe what the operator would do if a genuine saved output under the recipe produced a similar pattern. Percentages are hypothetical CW circuit reliability, not success probabilities for calling an individual station.

Illustrative Chatham interpretation exercise not a prediction

Step one is to check the header. If this is CW, SP, 100 W, September, SSN80 and the stated antenna/noise case, proceed. If it is SSB, LP, a different solar input, or an unrecorded receiving antenna, you are interpreting a different experiment. A number without its inputs is a rumor wearing a lab coat.

Step two is to identify the broader useful window. In this exercise, 20 meters is favored over 17 and 15 during the selected hours, with the highest teaching value at 0100 UTC. That suggests beginning before 0100 and listening through the surrounding hour rather than treating 0100 as a start bell. Actual openings can shift, and the prediction is not minute-resolved.

Step three is to plan a secondary experiment. At 0100, 17 meters is weaker in modeled reliability but still deserves a check, especially if current reports support it or the target is active there. The 25-percent 15-meter value does not forbid listening. It suggests a shorter speculative check rather than spending the entire session there without evidence. Ten meters gets a brief listen if genuinely relevant, not the main time budget in this teaching scenario.

Step four is to translate the chart into a practical allocation. Arrive at 2345 UTC, check current data and the station, and begin listening on 20. Around 0030, check 17 briefly. Around 0100, prioritize whichever band has actual target signals. Recheck the weaker candidates when reports or target announcements justify it. Keep a record of what was heard rather than simply writing “good propagation.”

Step five is to assess meaning. Seventy percent would be roughly 21 days out of a 30-day modeled month meeting the chosen CW requirement at that hour. It is not 70 percent of your calls getting answered. If a rare target is listening to louder stations, its availability and competition can dominate your result even while the path supports your signal.

Add hypothetical observations and choose the next action

Assume, for this exercise only, that recent PSK Reporter entries show several receivers in New Zealand hearing Tennessee-area FT8 stations on 20 meters. Assume a New Zealand beacon is also audible. This improves the case for a Pacific opportunity, but it does not prove a Chatham CW path. The stations’ locations and equipment differ and the signals use different communication systems.

Now assume you hear the hypothetical Chatham target at a weak but readable level on 20-meter CW. Listen long enough to learn its exchange and listening behavior. If it announces split operation, meaning it transmits on one frequency and listens on another, identify the actual listening range before calling. The propagation tool chose a band to inspect; the operator’s instructions choose the transmit frequency within your privileges.

Suppose the target fades for 20 seconds and then returns. Keep the received pattern in view and call during a readable period without filling every silence. A fade does not necessarily mean the opening has ended. A sustained disappearance across several relevant signals is stronger evidence that conditions have changed. Distinguish QRM, interference from other transmissions, from QRN, noise often associated with natural sources, and from simple low signal strength.

If the target remains absent but regional FT8 evidence is good, make a small FT8 reconnaissance if that fits your goal and the target’s operating mode. If the goal is specifically CW, do not treat a digital decode elsewhere as completion. Record it as evidence of regional support and continue targeted CW listening within a reasonable time budget.

Change the season without inventing a forecast

Run the same recipe for March, June, September, and December with SSN80 fixed. This isolates the model’s seasonal effect. June gives Nashville a northern summer and Chatham a southern winter; December reverses that pairing. March and September sample different equinox-season cases. Each can differ in path illumination, modeled absorption, and available upper-region support. The actual run determines the ranking.

For June, compare the Tennessee evening window with an early-morning window such as 0900–1200 UTC, then use the time and day/night tools to understand the route. On 15 June, 0900–1200 UTC is 0400–0700 CDT in Nashville and 2145 on 15 June through 0045 on 16 June on Chatham. This civil-time comparison is genuine arithmetic, not a claim that those hours are better. It gives you a second physically different part of the daily cycle to investigate.

For December, do the same rather than preserving a summer local-clock schedule. If you insist on 2000 Nashville time, winter gives 0200 UTC the next date rather than summer’s 0100 UTC. A one-hour seasonal clock shift can move your listening away from a chart’s favored hour. Work in UTC first and preserve the dated conversion beside it.

If a monthly result looks favorable for 30 or 40 meters during a different part of the cycle, check the entire route’s darkness and the receiving conditions. Endpoints alone are insufficient. A long low-band path can lose signal through sunlit intermediate segments even if both station clocks say nighttime.

Change solar input power mode and horizon separately

Hold September fixed and compare SSN30, SSN80, and SSN140. If the higher-frequency bands improve in a real output, consider those bands more seriously during fresh favorable observations. If the improvement is small or your modeled antenna has a poor relevant-angle pattern, do not assume the solar input has solved the station problem. The run is a sensitivity study and none of these SSN values is asserted as current.

Repeat with SSB and FT8 presets. If FT8 remains promising where SSB falls sharply, plan the mode that matches your objective. For an SSB goal, use the chart to identify a stronger hour, antenna case, or power margin rather than promising that voice will work because digital messages were heard. Avoid comparing a CW result at 100 W against an SSB result at 400 W and claiming the difference is solely mode.

Repeat 100 W versus 400 W. The power ratio gives 6 dB more transmitted energy, but the computed reliability change must be read from the output. Then repeat 3-degree versus 10-degree minimum TOA. If the result collapses at 10 degrees, label the opportunity highly dependent on low-angle access. That can matter more to a Tennessee station behind a hill than buying an amplifier.

An amplifier is a device that increases signal power. It can improve your outgoing signal but cannot improve the receiving noise environment or force the far operator to be present. A receiving antenna with better noise rejection can help you hear a target, but the amount of improvement depends on the direction and origin of noise and cannot be inferred from a generic gain number alone.

Compare long path honestly

Set LP and aim a directional antenna initially toward 52 degrees. Save a separate output. The path is over twice the SP distance, traverses different regions, and has different illumination and potential loss. Greater distance alone does not rule out a real LP opening, but the mere existence of an LP map line does not establish one.

If the target is heard, compare 232 and 52 degrees with the same antenna, receiver settings, and short alternating intervals. Front-to-back ratio is the difference in antenna response between its main forward direction and the opposite direction. A directional antenna’s back response, multipath, and fading can complicate the comparison. Multipath means simultaneous signals arriving by different routes, with different phases or delays. Phase is a signal’s timing position within a wave cycle; combining signals with different phase can strengthen or weaken the received result.

With a roughly omnidirectional antenna, meaning broadly similar horizontal response in many directions, the listening comparison may not establish the route. Record “path uncertain” if that is the evidence. A confident label is less useful than an honest observation that another operator can reproduce.

End the session with a result you can learn from

Record the UTC window, mode, band, power, antenna, chosen path, data timestamps, and what was heard. Include whether the goal was a completed QSO, a one-way copy, a digital decode, regional proxy reception, or no relevant evidence. A one-way copy means you received the target without confirmed return communication. It is still a valuable result, but it is not a QSO.

For this illustrative evening the operating plan is concrete: investigate 20 meters first, compare 17 when evidence supports it, keep 15 as a shorter test, and revisit the plan if real signals contradict the teaching chart. The genuine result must come from an actual manually saved run and the on-air session. That is the part of the experiment the sky gets to write.

12 Worked example two Nashville to Bishkek

The geography changes the priorities

Open the live VOACAP online

The goal is a short CW or FT8 contact with a hypothetical active operator in the Bishkek area. Use Nashville +36.1627, −86.7816 and Bishkek +42.8746, +74.5698. The SP distance is approximately 11030 km and the Nashville departure bearing is 14 degrees. The LP distance is approximately 29000 km and departure bearing 194 degrees. The SP midpoint near 78.4 degrees north is the main geographic clue: a northward bearing matters more here than the destination’s position on a rectangular map.

The first recipe is December, hypothetical monthly SSN80, CW preset, 100 W feedpoint power, 0 dBi reference antennas, residential receive noise, 3-degree minimum TOA, SP. As before, sweep every UTC hour before choosing a session. The selected teaching window is 1200–1500 UTC on 15 December 2026, equivalent to 0600–0900 CST in Nashville and 1800–2100 in Bishkek. Those are exact civil conversions under the named zones.

The physical follow-up case uses the same verified 10 m AGL dipole pattern as the Chatham study, reoriented to the selected Bishkek route. Rerun with the receiving antenna and noise changed if published destination station information supports it. Mountains near a real Central Asian operating site can obstruct useful elevation angles; do not model the city coordinate as a guarantee of an open horizon.

Interpret a second illustrative chart

Open the live VOACAP online

Every value in this table is invented for instruction. These are not real December predictions, current conditions, or reported results. They demonstrate the reasoning to apply to a genuine output with a similar shape.

Illustrative Bishkek interpretation exercise not a prediction

First verify the month, SSN, station settings, operating mode, and SP selection. Under this fabricated pattern, prioritize 20 meters around 1200–1400 UTC. Keep 17 as a serious secondary check and 30 as a lower-frequency alternative. The chart does not justify choosing 15 merely because you prefer that band. It does justify listening briefly if the target is announced there or new observations point toward it.

The 1300 teaching value on 20 meters would correspond to roughly 22 or 23 qualifying days in a 30-day conceptual month at the chosen requirement. It does not identify which calendar days are good. December has 31 days, and model percentages should not be mistaken for an exact count of successful sessions; the 30-day explanation is a convenient way to understand the fraction.

In the example, 30 meters has less modeled reliability than 20 but may offer a useful weak-signal alternative if the target operates there. Thirty meters is roughly 10 MHz. Its operating rules differ by jurisdiction, so check your own privileges and the mode in use. This guide compares bands as propagation choices and does not reproduce a worldwide licensing chart.

Now introduce a hypothetical quiet day

Assume a current check shows low geomagnetic activity, no material polar-absorption indication on the selected product, and recent reports from northern-route proxy receivers. These are hypothetical observations. Begin at 1130 UTC so you can listen before the chosen window. Aim the directional antenna toward approximately 14 degrees true, check the target’s actual frequency, and monitor 20 meters.

If the target is readable on 17 meters while 20 is crowded, use the actual evidence and move to 17. QSY means changing operating frequency. Before transmitting, identify whether the target is operating simplex, using the same frequency for transmitting and receiving, or split. A better propagation prediction is not permission to ignore the listening instructions.

If you hear the target but receive no reply, consider unequal power, different receiver noise, a poor return antenna angle, or competition. Run the reverse-direction case later using the correct destination transmitting assumptions and Nashville receive-noise model. Hearing someone is necessary for many contacts but not proof that they hear you at the required quality.

Introduce a hypothetical disturbed day

Now suppose the same monthly chart remains on the screen but a current product shows Kp6 and a polar absorption concern, with few recent reports on comparable northern paths. This scenario is fabricated. The chart has not become a live storm forecast just because it is displayed beside live weather. Its month-based expectation still needs present-condition qualification.

Begin with a short listening check on the target’s announced band. If absent, inspect 20 and 30 meters and recent target-region reports. If the station is genuinely heard LP, investigate that direction. If not, resist spending the whole morning calling toward an invisible target simply because a green cell from the monthly model still looks cheerful.

A practical allocation might be ten minutes of target listening, several minutes checking another band and current reports, then a decision to continue or stop. These are operator time-budget suggestions, not a propagation law. If actual signals appear, extend the session. If everything stays poor and the relevant disturbance persists, preserve the observation and try another day or route.

Do not multiply the teaching reliability by an invented “Kp correction factor.” A statement such as “Kp6 halves reliability” would require a validated model and supporting data, which this guide does not have. Use the disturbance as a reason to lower confidence and seek current evidence, not as a numerical patch pasted onto the monthly result.

Compare summer with winter and equinox cases

Repeat December, March, June, and September with the same SSN80 and station settings. For June, the 1200–1500 UTC window is 0700–1000 CDT in Nashville and 1800–2100 in Bishkek. The civil hours resemble the December plan, but the path’s daylight pattern differs substantially at high northern latitude. Identical destination clock time does not mean identical ionospheric conditions.

A summer northern route may retain more high-latitude illumination and offer different high-band opportunities. It can also retain absorbing sunlit segments and experience disturbances. Winter’s longer darkness can change low-band possibilities but may reduce upper-region support. Let the real all-hour sweep decide which candidates deserve time; do not elevate these physical tendencies into specific predicted percentages.

For March and September, compare the model’s favored hours and use an exact-date sunrise/sunset tool to inspect the endpoint and midpoint geometry. A grayline label deserves a full-route check. If a low-band result looks better than a high-band result, inspect the noise and antenna assumptions too. A low dipole can radiate at angles favorable for shorter communication while being less effective for the long route being studied.

Compare the long path and recognize its limits

For LP, aim initially at 194 degrees from Nashville and save a separate monthly result. The complementary route is much longer and samples a different part of the globe. It may offer an alternative under some conditions, but the result must be calculated and then checked on air. “The north is disturbed, so south must work” is not a defensible conclusion.

If a directional comparison produces a stronger target toward 194 degrees, repeat the test during a relatively stable interval. Keep power and receiver settings fixed, note fading, and be aware of rear antenna response. If no directional evidence is available, leave the route uncertain. A straight line drawn from Nashville to Bishkek on a website often indicates geometry, not the measured trajectory of that signal.

For a real Bishkek session, the most useful final plan contains a preferred UTC window and two backup bands, an SP/LP comparison where possible, a condition that would prompt you to shorten the attempt, and the actual destination operating information. That plan remains usable even when the model is imperfect.

13 Turn competing information into an operating decision

A repeatable ten minute preparation routine

1. State the target, goal, time available, and station configuration. Confirm the target is expected to be active; a great path to an empty frequency is a very quiet accomplishment.

2. Load the saved monthly prediction case and identify two or three candidate bands. Confirm coordinates, SP or LP, mode, antennas, power, noise, and UTC interpretation.

3. Inspect current geomagnetic, flare, and absorption information appropriate to the route. Read timestamps and distinguish forecast from observation. If data are stale or unavailable, mark that uncertainty rather than assuming benign conditions.

4. Inspect recent reports for the target, your own signal, or carefully chosen proxy stations. Use short intervals so yesterday’s excellent opening does not masquerade as this morning’s evidence.

5. Listen on the candidate bands. Check local noise and confirm the station is functioning. Choose the band with actual target evidence when available, using the prediction to guide backup checks.

This is a preparation routine, not a requirement to stare at screens before every casual contact. Once you have a useful signal, operate. The tabs will survive without your supervision for a few minutes.

A decision matrix for common disagreements

How to respond when evidence disagrees

The matrix deliberately avoids numerical thresholds that would pretend to be universal. A comfortable conversation and a rare-country digital exchange have different time budgets and signal requirements. Your objective decides how long a marginal opportunity is worth pursuing.

A realistic confidence statement

Instead of “Bishkek will be open at 1300,” write “The saved December SSN80 CW reference case favors 20 meters near 1200–1400 UTC. Current high-latitude conditions need checking, and the receiving antenna is assumed. I will listen before the window and compare 17 and 30 meters.” That is actionable and reviewable.

Instead of “The Pacific is green,” write “The selected 20-meter reliability map ranks parts of the Pacific favorably under its stated assumptions. Recent New Zealand reception is supporting evidence, but the Chatham target and exact station conditions remain unverified.” This sentence keeps geography, modeled performance, and observation separate.

Instead of “VOACAP was wrong,” write “The 0100 UTC Chatham session did not produce a readable target under this station configuration; regional beacons were weak and the noise floor was higher than usual.” Noise floor means the baseline level of unwanted received energy. That statement contains clues for the next attempt. Blaming the model alone does not.

14 Bands and modes as practical choices

A frequency reference for planning

Wavelength is the distance a wave travels during one cycle. In free space, approximate wavelength in meters is 300 divided by frequency in MHz. Band names therefore describe approximate wavelength, not an exact single operating frequency. The table below uses rounded frequencies for calculation and discussion. It is not a frequency-privilege table, a mode band plan, or an instruction to transmit on those exact values.

Rounded HF planning frequencies

The physical questions differ across the table. At 3.5 MHz, a useful efficient antenna is larger, atmospheric and local noise can be prominent, and a long path needs favorable absorption conditions. At 28 MHz, antenna size is easier but sufficient ionospheric support can be the limiting factor. A multiband antenna, one intended for multiple bands, does not have equal gain, efficiency, or useful-angle response on every band.

Why mode changes matter

SSB carries voice in a bandwidth wider than a narrow CW receiver filter and generally requires more signal-to-noise for comfortable intelligibility. Intelligibility means the ability to understand the information, particularly speech. CW lets an operator detect keyed tones in a narrower receiving bandwidth. FT8 uses synchronized transmissions and a decoder designed for compact weak-signal exchanges. Each system defines success differently.

Do not compare a bare negative FT8 SNR report with VOACAP’s positive normalized SNR and conclude that one tool contradicts the other. Noise bandwidth and reference conventions differ. A decoder can recover a structured message with a reported negative SNR in its reference bandwidth while an unstructured voice conversation remains difficult. The correct operational lesson is to select the communication system suited to the required exchange.

Digital operation also requires accurate time and clean audio. Overdriving means applying too much signal to a stage, causing distortion or unwanted spectral products. ALC, automatic level control, is transmitter circuitry that regulates drive or output. Follow the radio and software instructions rather than assuming every radio’s meter has an identical ideal reading. Poor transmit audio is a station problem that no propagation model can repair.

The purpose of mentioning modes is practical, not philosophical. A difficult path can be a satisfying CW contact, a short digital exchange, or a voice conversation on a better day. A mode that fits the goal is a tool, just like a better time window.

15 Troubleshoot the station before condemning the sky

Check what changed

If every band suddenly sounds empty, verify antenna selection, feed-line connections, receiver gain, and any active attenuator, a device or setting that deliberately reduces signal level. RF means radio frequency, the range of electromagnetic frequencies used for radio. AGC, automatic gain control, adjusts receiver gain to manage changing signal strength; gain is amplification or signal increase within a circuit. These settings can alter what you hear without any change in propagation.

If only one direction is poor, check antenna orientation, the horizon, and relevant route conditions. If only one band is poor, check matching and antenna behavior on that band. If all distant weak signals vanish while local noise grows, inspect electrical devices, chargers, solar-power electronics, and other possible noise sources using safe ordinary operating practices. Do not modify mains wiring or defeat protective grounding to conduct a propagation experiment.

An antenna analyzer is an instrument that measures antenna-system electrical behavior such as impedance or SWR. It can help identify an installation fault but does not measure the entire skywave circuit. A spectrum display shows received energy versus frequency. A waterfall is a time history of that spectrum shown with color or intensity. It can reveal interference and signal patterns, but its color scale depends on receiver settings and does not directly equal modeled reliability.

Recognize skip and silence

Skip distance is the distance from a transmitter to the nearest return of a particular skywave mode under given conditions. A skip zone is an area between useful groundwave coverage and the first skywave return where that signal may be weak. A band can support distant stations while providing poor coverage to nearby ones. Conversely, strong nearby stations do not prove a long intercontinental route.

NVIS means near vertical incidence skywave: high-angle HF propagation used for relatively nearby coverage. A low horizontal antenna can be useful for NVIS on an appropriate band while being less effective at the low angles needed for a long path. An antenna is not generally “good” or “bad” without naming the job.

Silence has several explanations: inadequate propagation, inadequate antenna response, high noise, no activity, the wrong frequency, stale information, or a technical fault. The practical way to distinguish them is to test one thing at a time and preserve evidence. A single empty dial sweep is a small observation, not a global scientific conclusion.

16 A record that makes the next attempt better

Save a prediction record

Use the following complete record format for each genuine manual run. The teaching tables cannot fill the result fields. Leave them blank until an actual tool output exists.

Prediction record fields: case name; tool URL or installed engine; software version; generation date and UTC; TX coordinates; RX coordinates; selected month; solar input value, scale, and source; automatic solar substitution status; geographic SP or LP; computational method if exposed; antenna file or preset on each band at each end; antenna height and orientation; feedpoint power; feed-line loss treatment; receive-noise label or numerical convention; minimum TOA; operating mode and required SNR; required reliability target; output variable and units; favored UTC hours; favored bands; sensitivity findings; saved output location.

That looks like a long list because it is one. Most fields can be copied from a baseline record and changed only when needed. The first complete record costs a few minutes; subsequent ones prevent hours of trying to remember why a saved chart looked better.

Save an operating record

Operating record fields: target and confirmed activity source; date and UTC start/end; band and actual frequency; mode; antenna; feedpoint or radio-output power with distinction noted; intended SP/LP and evidence for route identification; receiver settings for comparisons; local noise description; observed targets and proxy stations; report or beacon timestamps; space-weather product names and timestamps; outcome; next change to test.

An outcome can be no relevant reception, proxy reception, target copied, outbound reception confirmed by a reporting receiver, incomplete exchange, completed QSO, or completed QSO with route uncertain. Be precise. “Almost worked” is a feeling; “copied the target for six minutes but no return acknowledgment” is useful data.

An illustrative completed entry

Teaching entry only: September Chatham test; 0000–0200 UTC; 20-meter CW; 100 W at modeled feedpoint; dipole 10 m AGL; SP intended, route not confirmed; New Zealand proxy beacon audible; hypothetical target copied intermittently; no confirmed two-way exchange. Planned next test: repeat at the same UTC window after checking noise, and compare 17 meters if the target is active there.

This invented entry demonstrates record quality. It is not AF4OZ’s log, a real beacon reception, or a historical report. The value is that a future operator can distinguish what was attempted, what was observed, and what remains unknown.

Build station-specific confidence over time

Group repeated sessions by destination, band, mode, antenna, and UTC interval. Count listening sessions as well as successful contacts. A success rate based only on days you heard the target is conditional on target reception; a success rate based on all scheduled sessions answers a different question. Define the denominator, the total set against which a fraction is calculated, before interpreting a percentage.

Look for repeatable patterns rather than perfection. If your station often finds the useful Chatham window earlier than a generic chart, adjust the listening start while preserving the original records. If Bishkek is consistently difficult during northern disturbances, schedule a shorter initial check and spend the remaining time elsewhere. These are empirical station decisions, meaning decisions based on observation, and can coexist with a useful imperfect model.

17 Common mistakes and their repairs

Using the wrong solar number

A daily sunspot number, monthly smoothed SSN, and SFI are related indicators but different quantities. Repair the record by identifying the exact required input and source. Preserve sensitivity values as hypothetical. Do not relabel an illustrative SSN140 run as a forecast simply because a dashboard shows a large SFI.

Treating a percentage as the chance of winning a pileup

Repair the statement by naming the communication requirement, hour, month, and station assumptions. Consider competition and station activity separately. A reliable circuit can still produce no contact if the operator does not hear your call among many others.

Reading the wrong map legend

Repair the interpretation by naming the plotted quantity and units before describing a color. Reliability is a percentage. MUF is a frequency. Absorption is attenuation. Signal power is power on a logarithmic scale. Day/night shading is illumination. The same red color can mean a different thing on every map.

Confusing geographic LP with a computational method

Repair the setup by recording both the geographic path choice and the model method separately. A smoothing method used for long circuits does not automatically switch a Nashville-to-Chatham prediction to the 52-degree departure route.

Choosing an impossible antenna or horizon

Repair the model by inspecting antenna height, pattern, orientation, efficiency assumptions, and obstruction. Repeat a pessimistic case. A result that depends entirely on a clear 3-degree horizon should be presented as such rather than rounded up to “excellent conditions.”

Losing the date in the time conversion

Repair the schedule by recording the full UTC date and the corresponding local date at each endpoint. Chatham’s 45-minute offset and Tennessee’s split time zones are reasons to use named time zones, not mental shortcuts.

Mistaking no report for no path

Repair the experiment by checking participation, report age, filters, and software status. Combine self-reports, proxy receivers, beacons, and your own listening. An absent volunteer receiver cannot publish a negative measurement.

Changing too many variables

Repair the experiment by returning to a baseline and changing one input at a time. Mode, power, antenna, month, and noise can all alter the result. A greener chart after five simultaneous changes tells you less than a controlled comparison after one.

18 A compact operating plan for each destination

Chatham plan

Use the stated coordinates and begin with a full-hour sweep for the month and appropriate solar input. Compare SP 232 degrees and LP 52 degrees. Test 20 and 17 meters as initial reconnaissance candidates, then let the saved model and actual target activity expand or change that selection. Compare opposite-hemisphere seasons explicitly and keep 30 and 40 meters in the all-hour analysis. For the September evening teaching window, use 0000–0300 UTC with the previous-date Nashville conversion.

Before operating, check current absorption and fresh Pacific reception evidence. Treat New Zealand and Australian observations as proxies, not proof of Chatham. If the target is heard, follow its actual frequency and listening instructions. Record route uncertainty when the antenna cannot distinguish SP from LP. The goal is a real contact, not a collection of favorable map screenshots.

Bishkek plan

Use the stated coordinates and compare SP 14 degrees with LP 194 degrees. Pay particular attention to high-latitude conditions because the SP midpoint is near 78.4 degrees north geographically. Include winter, summer, March, and September in the seasonal comparison. Use the December 1200–1500 UTC window as a teaching session to investigate, not a published best-time prediction.

Compare 20 and 17 meters and keep 30 meters as a candidate where actual results and operating privileges support it. Check current geomagnetic and polar absorption information and the target’s activity. During disturbed conditions, make a shorter evidence-seeking attempt instead of mechanically correcting the monthly percentage. If actual LP signals appear, investigate them; do not assume a southern bearing cures a northern problem.

What to do when the sky surprises you

Try the readable signal, keep the exchange clean, and record what happened. The best use of a model is to put you in the chair at a plausible time with a sensible band choice. Its best use after the contact is to help explain and improve the next attempt. The ionosphere has never read your operating plan, but it occasionally cooperates anyway.

19 A guided beginner laboratory from map to contact

Learn the method rather than memorize a successful date

The purpose of these exercises is to teach a repeatable process. A date, percentage, or band in a worked scenario is a teaching device. You succeed at the exercise when you can explain what the tool is showing, choose a reasonable test, distinguish the result from your expectation, and improve the next attempt. Completing a contact is the eventual operational goal, but an unsuccessful test can still answer a useful question.

A hypothesis is a specific explanation or expectation that can be tested. An independent variable is the factor you deliberately change. A dependent variable is the outcome you measure. A control is a comparison condition that helps you interpret a change. In radio, perfectly controlled experiments are difficult because propagation changes while you work. You can still practice good science by recording assumptions, limiting changes, alternating comparisons, and resisting explanations that fit every possible result.

Your first hypothesis might be: “Under the same station settings and near the same UTC time, 20 meters will produce more relevant Chatham-region reception than 15 meters in the lower-SSN scenario.” The independent variable is band. The outcomes include target copy, reception reports, and whether a two-way exchange completes. Keep antenna, power, mode, and test duration as consistent as practical. If antenna response differs across bands, record it as a limitation rather than pretending the comparison measures only the ionosphere.

The goal is not laboratory perfection in a backyard. It is evidence better than “I turned some knobs and the map looked cheerful.”

Laboratory one read a propagation map aloud

Open one map before opening five. Write six answers: who produced it; what quantity it displays; its units; its timestamp; its selected band or frequency; and whether it is a prediction, measured report, or interpolated estimate. If a field is absent, write unknown and consult the help page. A confident guess is not an additional data source.

For a VOACAP reliability map, a good description is: “This is modeled percentage reliability from Nashville for the selected month, hour, mode, power, antenna, noise, and SP assumption.” For KC2G, say: “This is the displayed MUF estimate for the product’s 3000 km reference geometry; the numbered station dots identify measurement inputs and the space between them is estimated.” For D-RAP, name whether the legend is frequency at an absorption criterion or absorption at a fixed frequency. These sentences distinguish three visually similar maps with three different meanings.

Now locate Nashville and the destination. Trace the complete selected great-circle route. Identify the day/night shading, any map-edge wrap, and poorly measured regions. Describe a possible consequence: “The northern Bishkek route gives me a reason to inspect geomagnetic and polar absorption information.” Avoid jumping to “Bishkek is closed,” because the map-reading exercise has not supplied that observation.

Finally select one color or contour and state its numerical meaning. If the legend does not let you assign a precise value, use its actual interval. “Between 15 and 17 MHz” is better than inventing 16.37. If the legend shows reliability, never substitute a frequency or signal-strength interpretation. This exercise is complete when you can explain the map to another beginner without using its color as the explanation.

Laboratory two make your first useful VOACAP comparison

Use the Nashville and Chatham coordinates and the reference recipe in Chapter 6. Begin with a single month, hypothetical SSN80, CW, 100 W feedpoint power, SP, and the stated antenna, noise, and minimum-angle assumptions. Save the input record. Run the manual point-to-point analysis and identify the two best candidate bands and a useful hour range in the genuine output.

Then change only SSN to 30 and repeat. These are sensitivity cases, not claims about current or future solar conditions. Write what actually changed: the highest-ranked band, the width of a useful window, the relative reliability, or perhaps very little. If the result surprises you, first check that the input really changed and that dynamic solar substitution did not override it.

Change back to SSN80 and change only the mode to SSB. Ask whether the preferred hours change and whether the communication requirement reduces usable opportunities. A beginner should understand that “the path exists” and “this voice system can communicate reliably” are separate questions. Restore CW before making the next comparison so the baseline remains clear.

Repeat the exercise for Bishkek. Keep location and path records beside the charts; two saved files named good map and better map will not help you next month. The exercise is complete when you can say which one-variable change produced which result and why that result changes your next listening choice.

Laboratory three distinguish a model change from an observation

Imagine two contrasting situations. In Situation A, your prediction favors 20 meters and you hear the target there. The next action is to understand the target’s exchange and listening arrangement, then attempt the contact. You do not need to finish every map exercise before answering a readable station.

In Situation B, the same prediction favors 20 but the target is absent, while a fresh regional reception report exists on 17. First verify the report time, actual target identity, and receiving location. Listen on 17. If you hear the target, use that evidence. If only a nearby proxy is heard, record regional support and keep the exact target uncertain. The prediction remains a planning aid; the observation guides the immediate action.

In Situation C, neither band yields a target and your local noise is higher than usual. Check station settings and receiving conditions. Do not attribute all three situations to the solar cycle simply because that explanation is available. A hypothesis should discriminate among possible causes, not become a decorative label for uncertainty.

Laboratory four test your own signal with reception reports

Choose a clear, permitted frequency on one candidate band and use your actual call sign. For FT8, configure the receiving reports map to show your transmitted signal, the selected mode, the band, and a short recent interval. Make a small number of ordinary CQ transmissions. Note transmission UTC and allow upload time before checking which receivers reported you. For CW, use clean CQ calls with repeated call signs and inspect RBN reports as described earlier.

Record each useful receiver’s location, report time, band, and signal metric. A digital signal metric is meaningful only with its decoder’s reference convention. Do not compare the raw value with a CW or VOACAP SNR from a different bandwidth. Your beginner’s question is first “Where was my signal reported?” and then “Is that evidence relevant to the intended route?”

Repeat on a second band with a comparable short interval. Ideally return to the first band afterward so a general opening change does not masquerade as a band advantage. This is an alternating comparison. If the first band improves on return, you have evidence that time variation affected the experiment.

No reports can mean no propagation, no reporting receiver, a filter error, an upload delay, poor transmit setup, or insufficient signal. An exercise that records those alternatives is more scientific than a conclusion that silence proves the model wrong. Do not leave the station calling continuously just to populate the map.

Laboratory five complete a contact with an ordinary station first

A beginner can practice the entire contact procedure with a less demanding station before chasing Chatham or Bishkek. Find a readable station in a region appropriate to the current band. Listen to a completed exchange. Identify the call sign, operating mode, and whether the station is using simplex or split. Confirm that your transmit frequency and mode are appropriate before calling.

For CW or voice, send your call clearly at a pace suited to the exchange. Respond to what the operator asks rather than repeating a memorized script through every pause. For FT8, follow the software’s standard exchange and verify the call signs and acknowledgments. A reception report alone is not the finished exchange. Keep uncertain contacts marked as uncertain until the exchange is actually complete.

Log both calls, date, UTC, band, mode, and exchanged information. Make a separate note of the model and observations that led you there. Propagation tools help choose an opportunity, while operating skill completes it. Practice both. A beautifully researched path still benefits from remembering to release the transmit button.

Laboratory six write a conclusion that survives scrutiny

Use three sentences. First, state the hypothesis and fixed station settings. Second, state the observations with UTC and source. Third, state the next test and what remains uncertain. For example: “I compared 20 and 17 meters for the Chatham route using the same power and mode. During the short test, two relevant regional receivers reported me on 20 and none reported me on 17; no Chatham target was copied. I will repeat during another modeled window because receiver participation and changing propagation limit this comparison.”

That conclusion is hypothetical instructional wording, not a real test record. It shows how to avoid claiming a measured Chatham opening from regional reports. A later completed QSO can strengthen the evidence, but one contact still does not establish the best band for every day of the month.

For an antenna experiment, record A B A B observations from the same receiver rather than comparing your best A result with your worst B result. For a seasonal experiment, keep similar station configurations and compare multiple sessions. For a solar-cycle experiment, preserve the modeled monthly solar inputs and do not confuse one day’s storm with the multi-year trend.

20 Adapt the method as solar activity declines

What now means in this guide

A solar cycle is the roughly eleven-year rise and fall of solar magnetic activity, often tracked with sunspot numbers. Solar maximum is the high-activity phase; solar minimum is the low-activity phase. The declining phase is the interval after a maximum as the longer-term activity trend falls. The precise timing and strength vary, and daily conditions can move in the opposite direction from the long-term trend.

As background reviewed on 6 October 2026, SIDC and SILSO report that Solar Cycle 25’s smoothed sunspot-number maximum occurred in October 2024 and describe subsequent declining activity. Their discussion also emphasizes that strong flare and geomagnetic episodes can still occur during the decline. Thus this guide’s transition is from a relatively active, already declining cycle toward progressively lower-activity conditions, not from a guaranteed future peak on today’s date. This context is not a forecast of an individual HF opening. [28, 29]

NOAA’s progression page provides a monthly updated prediction with uncertainty shading. Use it for the broad trajectory and save its issue context when planning future months. Do not assign the hypothetical SSN140, SSN80, and SSN30 cases to specific future years without an appropriate forecast source. Those cases deliberately teach what to do as the solar input changes. [30]

Understand which limit is moving

With lower typical ionization, the high-frequency portion of an HF opportunity can become less available or shorter in duration on a given route. The practical question is which band still has adequate support and signal-to-noise at the hour you can operate. That is why the all-band, all-hour comparison should be repeated instead of keeping a favorite 10-meter schedule indefinitely.

Two limiting situations need different responses. In a support-limited situation, the ionosphere is not supporting the selected frequency well enough for the intended geometry. Moving down in frequency or changing the hour can help; extra power does not restore the missing ordinary refraction mechanism. In a signal-to-noise-limited situation, support exists but the signal does not meet the communication requirement. Better antenna response, more appropriate mode, reduced losses, lower receive noise, or additional power can help. These are diagnostic categories for planning, not perfect binary descriptions of every real path.

Use both MUF-related information and SNR or reliability outputs to distinguish the cases. If high-frequency support deteriorates while a lower band retains good relative performance, change the band search. If support remains plausible but reliability is weak, inspect the station and noise assumptions before moving. “Solar activity is lower” is context; it does not replace this diagnosis.

Repeat a controlled solar input ladder

Open the live NOAA Solar Cycle Progression

Open the live VOACAP online

For each destination, choose one month and hold coordinates, antennas, path, noise, mode, power, and minimum angle fixed. Run SSN140, SSN80, and SSN30 manually. A ladder here means a sequence of comparison cases with decreasing solar input, not a time forecast. Inspect all 24 UTC hours and save each output under a distinct case name.

From each run, extract four findings: the highest-frequency band worth serious investigation, the strongest candidate hour range, the most useful backup band, and whether the result depends heavily on low-angle access or mode. If the choices stay similar, record that too. Do not manufacture a dramatic deterioration merely because the chapter is about decline.

Then repeat for another season. A change in solar input can alter the best search strategy differently in winter and summer. Keep the season comparison separate from the solar-input comparison so you can explain the cause of each modeled change. Finally repeat with realistic antennas; the useful planning range may narrow more for a modest station than for the ideal reference.

Contrasting Chatham situations

Situation A is a hypothetical strong-support case. Your genuine run, if it produces this behavior, ranks 15 and 17 meters well during part of the Tennessee evening and retains 20 as a backup. Fresh Pacific reports and your own listening support the higher bands. Begin with the strongest actual target evidence, compare 15 and 17, and keep 20 available. Do not drop to a lower band solely because you have heard that the cycle is declining.

Situation B is a hypothetical intermediate case. The controlled lower-SSN run reduces the higher-band window while 20 stays more useful. Organize a narrower high-band check around the modeled opportunity, then give 20 more time. Compare 17 whenever fresh reports or target activity support it. The adjustment is a change in search priority and timing, not the retirement of an entire band.

Situation C is a hypothetical low-support case. Ten and 12 meters show little useful ordinary HF support for the selected path, and the stronger candidates move toward 20, 30, or 40 at different hours. Inspect full-route darkness, noise, the target’s mode, and your antenna on those bands. Tennessee evening may cease to fit your goal; investigate an early-morning session using UTC and exact local conversions. Lower bands require their own evidence rather than inheriting the old high-band schedule.

Situation D is an exception worth recognizing. A declining cycle still produces an unexpectedly useful 15-meter opening and you hear the Chatham target. Try it and record the conditions. Your long-term plan should make higher-band monitoring more selective when appropriate, not prevent you from using evidence in front of you.

Contrasting Bishkek situations

In a hypothetical stronger solar-input case with quiet current geomagnetic conditions, a real model output may support serious 17- or 15-meter checks alongside 20. Use the northern route geometry to choose relevant observations and keep the actual target’s activity in view. Strong monthly solar support does not eliminate polar-path vulnerability.

In a hypothetical lower-SSN case, the model may favor lower bands or a narrower hour range. Begin with the newly favored band, compare 20 and 30 where useful, and investigate 40 only with suitable darkness, antenna performance, noise, and legal operating conditions. Change the listening schedule if the full-hour analysis justifies it. Do not keep the December teaching window simply because it is familiar.

In a hypothetical lower-SSN case with elevated geomagnetic disturbance, you face two different problems: diminished background support and current disturbance on a high-latitude route. Check them separately. A weak-signal mode may help with margin; a different hour or quieter day may help with the disturbed path. LP is an independent experiment requiring its own model and on-air evidence, not an automatic escape route.

The teaching lesson is that low SSN with quiet magnetic conditions can be more useful for a particular contact than higher SSN during a severe disturbance. That is a qualitative possibility, not a universal ranking or an arithmetic rule combining SSN and Kp.

A diminishing cycle does not guarantee magnetic quiet

A coronal hole is a region of the solar corona with magnetic structure allowing relatively fast solar wind to escape. The corona is the Sun’s outer atmosphere. A high-speed stream is a faster flow of solar wind that can disturb Earth’s magnetic environment. Persistent coronal holes can produce recurring influences as the Sun rotates, approximately a 27-day interval as viewed from Earth. NOAA explains these sources and their possible geomagnetic effects. [31]

Use recurrence as a reason to check fresh forecasts near a later rotation, not as an appointment for a guaranteed storm. A structure can evolve, its alignment can change, and the magnetic conditions of the arriving flow matter. For the Bishkek route, keep a note of disturbed sessions and compare current NOAA guidance before the next planned attempt. A smaller sunspot number does not remove the need for those checks.

You do not need to become a solar physicist to apply this. Read the current forecast’s explanation, check the observed magnetic activity and absorption products, and look for route-relevant signals. The solar-cycle graph chooses your long-term expectations; today’s measurements qualify today’s attempt.

Change how you use each tool

In VOACAP, update the monthly solar assumption as appropriate and rerun the full daily sweep. Keep an older saved baseline so you can compare how the preferred bands and hours change. A saved peak-era chart is a historical comparison, not a current operational forecast.

In HamClock, inspect the solar input actually used and confirm data freshness. Use the dashboard to revisit changing band opportunities, but compare it with the detailed station case when antenna and noise assumptions differ. As upper-band openings become less frequent, a quick repeated check can be more useful than leaving the display on an encouraging historical screenshot.

On a MUF map, trace the route and identify whether the selected high band is plausible under the displayed reference geometry. Watch the timestamp and sparse measurement regions. If one part of a long path looks unfavorable, investigate a lower band rather than trusting a favorable value above Nashville alone. The map remains a clue with geometry and interpolation limitations.

With PSK Reporter and RBN, use short recent intervals and relevant receiving locations. If activity migrates to lower bands, configure those bands in your searches. Check whether the lack of reports reflects reduced participation before concluding the path has vanished. With beacons, compare available bands using the current schedule and status rather than assuming an unheard high-band beacon proves every HF band is closed.

In your log, track the number of attempts, relevant receptions, and completed contacts under comparable settings. If your station’s best window shifts, change your schedule. If noise becomes the limiting factor on the newly favored lower bands, receiving improvements can become more valuable than additional transmitter power.

A decision table for the transition

Illustrative operating adjustments as solar support changes

The situations and actions are teaching examples. They do not assign a future year, SSN threshold, or guaranteed opening to any band. A reasonable scientific method should survive the forecast being imperfect.

A monthly review and a nightly routine

Once a month, refresh the solar-input source, rerun important destination cases, compare favored bands and UTC windows, and review recent station records. Keep source date, actual input, and station assumptions together. Update the plan when the evidence warrants it; avoid changing everything after one unusual session.

Before each planned session, check current disturbance, recent reports, and direct reception. That nightly routine remains the same from a strong cycle to a weak one. What changes is the mix of bands, timing, useful station improvements, and how often a speculative high-band check deserves a long attempt.

The durable skill is choosing the next test for a stated reason, making the test carefully, and learning from its outcome. A contact confirms that the complete communication system worked then. A failed contact tells you something only when you record enough context to distinguish the alternatives. That method will still work when the Sun is less generous with the higher bands.

21 Primary references and useful tool links

References support documented behavior and technical context. The Tennessee operating protocols, hypothetical reliability exercises, illustrative observations, and mathematical station examples are original instructional material. No source is presented as evidence that a specific Tennessee path was predicted or observed on 6 October 2026.

[1] VOACAP project home. Program identity, origins, service links, and the prohibition on unapproved automated service access. https://www.voacap.com/

[2] NOAA Space Weather Prediction Center. Ionosphere. Background on ionization and ionospheric regions. https://www.spaceweather.gov/phenomena/ionosphere

[3] VOACAP. Understanding predictions. Technical reference hub for MUF, SNR, MUFday, and signal/noise distinctions. https://www.voacap.com/2023/voacap-understanding-prediction.html

[4] VOACAP. Setting up input values. Desktop parameter conventions, UTC, monthly inputs, noise, antennas, power, and methods. Desktop labels do not necessarily match the web interface. https://www.voacap.com/2023/voacap-windows-input.html

[5] VOACAP. Ten common mistakes. Practical input pitfalls including daily solar counts, power units, and input assumptions. https://www.voacap.com/2023/understanding/10mistakes.html

[6] VOACAP. Method 15 and antenna patterns. Inspecting and validating the selected antenna response. https://www.voacap.com/2023/understanding/method15.html

[7] George Lane. Signal to Noise Predictions Using VOACAP Including VOAAREA. Technical user guide hosted by Greg Hand; an older engineering reference, not a current live forecast. https://www.greg-hand.com/manuals/VOACAP%20Users%20Guide.pdf

[8] VOACAP Online for Ham Radio. Current public prediction interface and tool entry point. Interface inspected; no automated prediction requests made. https://www.voacap.com/hf/

[9] Jari Perkiömäki OH6BG. VOACAP Online User Manual. Revision identified in the inspected document as 25 March 2025. Online controls, monthly scope, and chart interpretation. https://www.voacap.com/2023/documents/VOACAP_Manual.pdf

[10] VOACAP. Maintaining the required grade of service. Meaning of reliability and required SNR. https://www.voacap.com/2023/understanding/rel.html

[11] VOACAP. Signal power to S meter conversion. Practical limitations of interpreting modeled power as S units. https://www.voacap.com/2023/understanding/s-meter.html

[12] VOACAP. Running predictions. Text-output fields and example format. The source’s published examples are not the Tennessee examples in this guide. https://www.voacap.com/2023/voacap-windows-output.html

[13] HamClock continuation at hamclock.com. Provider’s account of original-service availability and build-specific continuation guidance. https://hamclock.com/

[14] Open HamClock Backend. Client project status and manual link. Version 4.32 release entry dated 26 September 2026 was visible during research. The linked current manual was not retrievable through the research tools, so its full UI details are not claimed as verified. https://ohb.works/hamclock-client-project/

[15] Open HamClock Backend. VOACAP service repository. Describes VOACAP-based band-reliability data supplied to compatible clients. https://github.com/openhamclock/voacap-service

[16] NOAA. F10.7 cm radio emissions. Solar-flux meaning and units. https://www.spaceweather.gov/phenomena/f107-cm-radio-emissions

[17] NOAA. Planetary K index. Estimated planetary geomagnetic index and measurement context. https://www.spaceweather.gov/products/planetary-k-index

[18] NOAA. Solar flares and radio blackouts. Flare mechanisms and R-scale thresholds. https://www.spaceweather.gov/phenomena/solar-flares-radio-blackouts

[19] NOAA. HF radio communications. Daylit and high-latitude effects relevant to HF operators. https://www.spaceweather.gov/impacts/hf-radio-communications

[20] NOAA. D Region Absorption Predictions. Distinction among the global 1 dB frequency criterion, polar 10 dB criterion, and fixed-frequency absorption products. https://www.spaceweather.gov/products/d-region-absorption-predictions-d-rap

[21] KC2G. About the ionospheric maps. MUF for 3000 km, foF2, measurement dots, interpolation, timing, and limitations. https://prop.kc2g.com/about/

[22] PSK Reporter. Developer information. Reception-record meaning and reporting context. Map entry point https://pskreporter.info/pskmap.html and technical reference https://www.pskreporter.info/pskdev.html

[23] Reverse Beacon Network. Project introduction and map entry point. https://www.reversebeacon.net/

[24] Pete Smith N4ZR for the Reverse Beacon Network. How to get spotted. CW recognition and repeated-call guidance. https://www.reversebeacon.net/pages/How%2Bto%2Bget%2Bspotted%2Bby%2Bthe%2BRBN%2B44

[25] NCDXF and IARU. International Beacon Project introduction. Frequencies and operational purpose. https://www.ncdxf.org/beacon/intro.html

[26] NCDXF and IARU. Beacon transmission schedule. Power-step sequence, schedule, and status; follow its link to the updated schedule when using it. https://www.ncdxf.org/beacon/beaconschedule.html

[27] WSJT-X project. Mode documentation and software entry point. Use the manual matching your installed version for exact decoder conventions. The full 2.7 manual could not be fetched by the research browser, and this guide does not assert its exact thresholds. https://wsjt.sourceforge.io/wsjtx.html

[28] SIDC and SILSO. Solar Cycle 25 reached its maximum in October 2024. Background on the identified smoothed cycle maximum. https://www.sidc.be/index.php/article/solar-cycle-25-reached-its-maximum-october-2024

[29] SIDC. Progress of Solar Cycle 25. Declining solar activity with continuing possibilities for substantial disturbances. https://www.sidc.be/article/progress-sc25-0

[30] NOAA. Solar Cycle Progression. Monthly updated trajectory and uncertainty display; broad-cycle context rather than individual HF openings. https://www.spaceweather.gov/products/solar-cycle-progression

[31] NOAA. Coronal Holes. High-speed solar wind sources, persistence, and possible geomagnetic effects. https://www.spaceweather.gov/phenomena/coronal-holes

Useful pages to keep beside the radio

VOACAP inputs and predictions: https://www.voacap.com/hf/

HamClock continuation and service guidance: https://hamclock.com/ and https://ohb.works/hamclock-client-project/

NOAA current space-weather products: https://www.spaceweather.gov/

KC2G MUF and critical-frequency maps: https://prop.kc2g.com/

PSK Reporter reception map: https://pskreporter.info/pskmap.html

Reverse Beacon Network: https://www.reversebeacon.net/

International Beacon Project: https://www.ncdxf.org/beacon/

Final reproducibility note

Coordinates are declared planning assumptions. Geographic distances and bearings use a spherical 6371 km Earth. Civil-time examples use America/Chicago, Pacific/Chatham, and Asia/Bishkek rules for the stated 2026 dates. The monthly SSN30, SSN80, and SSN140 cases are hypothetical sensitivity inputs. The two reliability tables, all hypothetical reception situations, and the sample operating log are original teaching material. No live solar data or actual end-to-end prediction is represented as a result. Save genuine manual outputs before using a modeled percentage as evidence in a future published article.