docs(wiki): cover the recent work, with worked examples

Asked for the wiki to catch up, and to be explicit — examples rather than
descriptions. New page and six rewritten sections.

Recent QSOs and Filters, the page named in the request: the right-click menu is
now a table of what every entry does, and the filter builder has five worked
multi-criteria filters written out field by field, each ending in what you then
do with the result. Including the distinction that prompted it — "export
selected" takes the highlighted rows, "export filtered view" takes every QSO the
filter matched, and after building a filter it is the second one you want. Plus
the traps: ADIF date format, and "is empty" not being "equals N".

Connections (new): the page for a settings section that had none. What OpsLog
listens to and sends, which outbound row another logger actually wants — a
logger asking for "WSJT-X UDP" discards a bare ADIF record — the custom-message
triggers with their placeholder lists, three worked examples, and the port-clash
warning that produced this week's CAT-drop report.

Amplifiers and Switches: the generic HTTP relay in full — per-relay URLs, the
pattern form, {relay}, {relay-1} and {value} with what each one sends — relay
automatic control with a four-relay antenna-switch example, and the bench power
supply including why OpsLog writes only its output.

Rotators: the two new native types (SPID with the Rot1Prog/Rot2Prog choice,
DCU-1), two-rotor setups, and that Test only reads the heading on the types
where it does.

Awards: the US county database, and the difference between the automatic fill
(blanks only) and the right-click re-derive (replaces) — with the Connecticut
planning regions, which is the case that needs it.

QSL Management: QSL_VIA against QSL_SENT_VIA/QSL_RCVD_VIA, why an import shows E
on every Log4OM record, and that V counts as confirmed.

Troubleshooting: the three field reports of the last few days, by symptom.

Every [[link]] checked to resolve.

wiki/Amplifiers-and-Switches.md

@@ -55,6 +55,8 @@ Over TCP / GSCP — a docked **A/B antenna-switch** widget:
 
 Configure the host / password in Settings → Antenna Genius.
 
+---
+
 ## Relay boards (Station Control)
 
 Relay boards for antenna/accessory switching, with named buttons in Station
@@ -68,11 +70,139 @@ Control:
   web page — a factory board needs neither.
 - **Denkovi** USB (4/8 relays, FT245) and generic **CH340 / LCUS** USB-serial
   boards.
+- **HTTP relay (home-made / generic)** — see below.
 
-A Test-connection button and detection feedback live in the setup panel.
+### HTTP relay (home-made / generic)
+
+For any box that switches on a plain HTTP GET: an ESP8266 with a web page, a
+Shelly, a Sonoff on third-party firmware, a qro.cz board, a home-brew switch.
+Pick the relay count, name the relays, and give the addresses.
+
+There is **no Host field and no connection test**. This device type has no
+address of its own — each relay carries its own full URL, and they need not
+even be on the same box. Nothing is read back either: OpsLog remembers what it
+commanded, so after a restart each managed relay is re-commanded once.
+
+**Two ways to give the addresses**, and you can mix them.
+
+**A — one URL per relay.** The reason this type exists: a hand-made switch
+often has addresses with nothing in common between channels.
+
+| Relay | ON URL | OFF URL |
+|---|---|---|
+| 1 | `http://192.168.1.9/FF0101` | `http://192.168.1.9/FF0100` |
+| 2 | `http://192.168.1.9/relay2/on` | `http://192.168.1.9/relay2/off` |
+
+**B — one pattern for the whole board**, with the relay number filled in:
+
+```
+ON  pattern:  http://192.168.1.9/relay?n={relay}&state=on
+OFF pattern:  http://192.168.1.9/relay?n={relay}&state=off
+```
+
+#### Substitutions
+
+| In a URL or a pattern | Becomes |
+|---|---|
+| `{relay}` | the relay number, 1-based |
+| `{relay-1}` | the relay number counting **from zero** — for boards whose channels are 0, 1, 2… |
+| `{value}` | **that relay's label**, from the *Relay labels* row below |
+
+`http` may be left off — it is assumed. `https` works and is used as typed.
+
+`{value}` is for a switch addressed by antenna name rather than by channel
+number. Name relay 1 **Ant1**, and:
+
+```
+Per-relay ON URL for relay 1:   http://10.10.10.100/relay?on={value}
+Sent:                           http://10.10.10.100/relay?on=Ant1
+```
+
+Rename the antenna and the address follows it — the name on the button and the
+name on the wire cannot drift apart, because they are the same text. A label
+with a space or an accent is percent-encoded automatically. A relay whose URL
+uses `{value}` **must have a label**; the editor warns you while you type.
+
+A per-relay URL always wins over the pattern, so you can write a pattern for
+the seven ordinary channels and one full URL for the odd one out.
+
+### Manual switching
+
+Every relay is a button in **Station Control** — click to toggle. The generic
+HTTP board is never greyed out (there is no connection to wait for); the other
+types are enabled once the board answers.
+
+---
+
+## Relay automatic control
+
+Settings → **Relay automatic control** turns relays by frequency or band, the
+equivalent of PstRotator's automatic control. Each relay of each configured
+board gets a rule:
+
+| Mode | Meaning |
+|---|---|
+| **Off** | OpsLog never touches this relay. |
+| **Band** | ON while the current band is one of the bands you tick. |
+| **Frequency** | ON while the frequency is inside a kHz range you type. |
+
+Example — a three-way antenna switch on a KMTronic board:
+
+| Relay | Label | Mode | Setting |
+|---|---|---|---|
+| 1 | Beam 20-10 | Band | 20m, 17m, 15m, 12m, 10m |
+| 2 | 40 m dipole | Band | 40m |
+| 3 | 80 m inverted-V | Band | 80m, 160m |
+| 4 | 6 m preamp | Frequency | 50000 – 54000 kHz |
+
+Rules to know:
+
+- A relay is only switched when it is **not already** in the wanted position.
+  On the first evaluation after launch OpsLog reads the boards' live state, so
+  a relay already correct is left alone rather than clunking.
+- An **unknown** frequency or band changes nothing. When CAT disconnects the
+  frequency reads zero, and treating that as "out of range" would switch every
+  relay off.
+- The rules follow **your radio**. Changing band in the QSO entry strip drives
+  the rig, the rig reports back, and the relays follow. **Without a CAT
+  connection** the Band selector itself is the band change, and the relays
+  follow that instead — unless the band or frequency 🔒 is on, which means the
+  entry is deliberately decoupled from the rig (logging an old contact) and no
+  hardware should move.
+
+For a switch driven by a URL or a UDP message rather than by a relay board, see
+[[Connections]].
+
+---
+
+## Bench power supply
+
+Settings → **Power supply**. A programmable supply on a serial port speaking
+**Modbus RTU** — BSIDE, Wanptek and the other supplies that use function codes
+03 and 06.
+
+Give the COM port. The other two fields are the factory settings and only need
+changing if you changed them on the supply: **9600 baud**, **Modbus address 1**
+(8 data bits, no parity, 1 stop bit).
+
+A card appears in **Station Control**: an **Output** on/off button, the measured
+volts / amps / watts, and — in small type — the voltage and current the supply
+is *set* to. That pairing is the useful part: `13.80 V set` beside `0.020 A`
+says at a glance that the supply is on and the radio is drawing nothing.
+
+> **OpsLog only ever writes the output on/off.** The supply's register map also
+> exposes the voltage and current set points and the three protection trip
+> levels, and none of them belong to a logbook — a wrong value there is 30 V
+> where a radio expected 13.8. They are read and displayed, never changed.
+> Set them on the supply's own front panel.
+
+If the protection has tripped, the card says so with the supply's status word.
+
+---
 
 ## Station Control
 
 The Station Control tab is a dashboard of fixed-width panels that wrap to fill
-the window — amplifier(s), tuner, relays, rotator, Ultrabeam… Drag a panel by
-the grip on its left edge to reorder; the order is remembered.
+the window — amplifier(s), tuner, relays, rotator, power supply, Ultrabeam…
+Drag a panel by the grip on its left edge to reorder; the order is remembered.
+A column-count selector (Auto / 1–6) caps how many sit side by side.

wiki/Awards.md

@@ -163,3 +163,40 @@ built-in award without fear of a future release overwriting it.
 
 **Rescan** re-pulls the logbook and recomputes — it picks up fresh LoTW / QRZ /
 eQSL confirmations (see [[QSL Management]]).
+
+---
+
+## US counties (CQ USA-CA)
+
+A county award needs a county on every US QSO, and most logs do not have one:
+neither the callsign nor the prefix carries it, and few operators send it.
+
+**Settings → US counties** downloads the FCC **ULS** database — every US
+amateur licence with its address — and builds a local index of callsign →
+county + grid. It is a few hundred megabytes to fetch and takes a few minutes;
+once built it works offline and needs no key.
+
+With it downloaded:
+
+- A **new US QSO** gets its county (and grid, if you have none) filled in
+  automatically — **only where the field is empty**. A value you or QRZ
+  supplied is never overwritten: the ZIP-derived county is about 98 % right and
+  the one you logged is usually better.
+- **Right-click → Update US county from the ULS database** re-derives the
+  county of the selected contacts and **replaces** what is stored. That is the
+  deliberate opposite of the automatic pass, and it is the way to correct
+  counties already in the log — see the Connecticut note below. Contacts
+  outside the US, and callsigns the database does not hold, are left alone.
+  The entry only appears once the database is downloaded.
+
+### Connecticut
+
+Connecticut **abolished its eight counties** for statistical purposes in 2022;
+the Census Bureau replaced them with nine **planning regions**. CQ USA-CA still
+uses the old county names, so OpsLog maps a Connecticut address back to its
+legal county.
+
+If your database was built by an older OpsLog, the Settings panel says so and
+offers a refresh — a download date alone cannot show it, because a database
+fetched yesterday by an older version still holds the wrong Connecticut
+counties. After refreshing, use the right-click update on your CT contacts.

wiki/Connections.md

@@ -0,0 +1,156 @@
+# Connections
+
+Settings → **Connections** is where OpsLog talks to other programs and to
+hardware over the network: what it **listens** to, what it **sends**, and
+messages you write yourself.
+
+It was called *UDP* until it grew a second transport — a row can now send an
+**HTTP GET** instead of a datagram, which is what most home-made antenna
+switches understand.
+
+Each row has a name, a direction, a service type, a port, and an on/off switch.
+**Everything sent and received is written to the application log**, so a row
+that does not work can be diagnosed rather than guessed at — see
+[[Troubleshooting]] for where the log lives.
+
+---
+
+## Inbound — what OpsLog listens to
+
+| Service | Sends it | Use |
+|---|---|---|
+| **WSJT-X / JTDX / MSHV** | those programs, port 2237 | Log their QSOs automatically, follow the DX call being worked, and show their decodes. Multicast is normal here. |
+| **ADIF over UDP** | JTAlert, GridTracker, FLDIGI | A text ADIF record per logged QSO. |
+| **N1MM Logger+** | N1MM | Its XML contact record. |
+| **Remote callsign** | DXHunter and similar | A callsign — and optionally a frequency and mode — to load into the entry form and tune to. |
+
+### Multicast or unicast?
+
+WSJT-X and MSHV normally broadcast to a **multicast group** (usually
+`239.255.0.1`) so several programs can hear them at once. Tick *Multicast* and
+give the group. A program sending to one address only needs unicast — leave it
+unticked.
+
+> **Do not put an inbound row and an outbound row on the same port.** OpsLog
+> then receives its own messages, and something it publishes can come back as a
+> command. It says so in the log when it spots the arrangement:
+> *"X sends on port 2241 and Y listens on it — OpsLog will receive its own
+> messages there; give one of the two another port."*
+
+---
+
+## Outbound — what OpsLog sends
+
+| Service | Format | Fires on |
+|---|---|---|
+| **ADIF message** | a plain ADIF record | each QSO logged |
+| **WSJT-X logged QSO** | the same ADIF wrapped in a WSJT-X datagram | each QSO logged |
+| **PstRotator frequency** | `<PST><FREQUENCY>` | frequency change |
+| **N1MM RadioInfo** | N1MM's RadioInfo XML | frequency or mode change |
+| **Custom message** | whatever you write | a trigger you pick |
+
+### Which one for another logger?
+
+If a logger says it accepts "WSJT-X UDP", it wants the **WSJT-X logged QSO**
+row — it listens on the WSJT-X interface and silently discards a bare ADIF
+record. Logger32 is the usual case. A logger that documents a plain ADIF
+listener wants the **ADIF message** row instead. They are two rows because they
+are two different things on the wire; enable the one your logger asks for, not
+both, or the QSO arrives twice.
+
+---
+
+## Custom messages
+
+The general case: **you choose when it fires, what it says, and how it leaves.**
+This is how a home-made antenna switch, a relay box or a home-automation server
+gets told what the station is doing.
+
+### 1. Pick the trigger
+
+| Trigger | Fires when |
+|---|---|
+| **QSO logged** | a contact is saved |
+| **Band change** | the radio changes band — or, on a station with no CAT, the Band selector in the entry strip |
+| **Rotator command** | the antenna is told to turn (compass, SP/LP buttons, a spot click) |
+| **Lookup done** | a callbook lookup returns |
+
+### 2. Write the message
+
+Anything in `{braces}` is replaced. What is available depends on the trigger:
+
+| Trigger | Placeholders |
+|---|---|
+| QSO logged | `{call}` `{band}` `{band_m}` `{mode}` `{grid}` `{name}` `{country}` `{rst_s}` `{rst_r}` `{comment}` `{date}` `{time}` `{freq_hz}` `{freq_mhz}` `{dxcc}` |
+| Band change | `{band}` `{band_m}` `{mode}` `{freq_hz}` `{freq_mhz}` |
+| Rotator command | `{az}` `{el}` `{path}` (`SP`, `LP` or empty) |
+| Lookup done | `{call}` `{name}` `{grid}` `{country}` `{qth}` `{state}` `{dxcc}` |
+
+`{band}` is `20m`; `{band_m}` is just `20`, because an antenna switch usually
+wants the number and nothing else.
+
+### 3. Choose the transport
+
+- **UDP** — a datagram to an address and port. You choose the line ending.
+- **URL** — an HTTP GET. Values are URL-encoded automatically.
+
+### Examples
+
+**Turn an antenna switch on every band change, over HTTP:**
+
+```
+Trigger:    Band change
+Transport:  URL
+URL:        http://192.168.1.50/set?band={band_m}
+```
+20 m → `http://192.168.1.50/set?band=20`
+
+**Tell a rotator display where the antenna is going:**
+
+```
+Trigger:    Rotator command
+Transport:  UDP  →  192.168.1.77:8100
+Message:    <AZIMUT>{az}</AZIMUT><PATH>{path}</PATH>
+```
+
+**Announce each QSO to a shack dashboard:**
+
+```
+Trigger:    QSO logged
+Transport:  URL
+URL:        http://homeassistant.local:8123/api/webhook/qso?call={call}&band={band}&mode={mode}
+```
+
+> Credentials in a URL are sent as typed — this is meant for a LAN. Passwords
+> are **redacted in the log** so a log file can be shared safely.
+
+### When a band change is a band change
+
+The Band-change trigger follows **the radio**, because an antenna switch should
+follow the radio and not what is being typed. Changing band in the entry strip
+drives the rig, the rig reports the new band, and the trigger fires from that.
+
+On a station whose rig OpsLog does not control there is nothing to report back,
+so the **Band selector itself** counts as the band change. The 🔒 lock on Band
+or Frequency suppresses it: a lock means the entry is deliberately decoupled
+from the rig, and nothing should move for a contact logged from last year.
+
+---
+
+## Diagnosing a row that does nothing
+
+The log names every message. In order, check:
+
+1. **Is the row enabled?** The list shows it.
+2. **`udp: Reload done — N server(s) running`** at startup, and one
+   `cfg id=… name=… dir=… service=… port=…` line per row. A row that failed to
+   start is listed with its error.
+3. **Outbound:** `udp: [NAME] sent N bytes to ADDRESS (service)` for a datagram,
+   or the URL for an HTTP row.
+4. **A message that renders empty is skipped** and the log says so — usually a
+   placeholder that the trigger does not provide.
+5. **`a QSO was logged but no outbound "ADIF message" row is enabled`** — the
+   answer to "why does my other logger get nothing".
+
+See also: [[Amplifiers and Switches]] for relay boards and their own automatic
+control, and [[DX Cluster and Spots]] for spot sources.

wiki/DX-Cluster-and-Spots.md

@@ -56,3 +56,20 @@ A **DXHunter** remote-call UDP packet fills the callsign, and when it carries
 [[Settings and Data]] for the UDP configuration.
 
 See also: [[Maps and Antennas]] for turning the beam to a spot.
+
+## When the list looks empty
+
+If spots are arriving and none are shown, the panel says so, names the filters
+responsible and offers a button to clear them all. The two easiest to forget are
+the **band and mode locks** (🔒), because they follow the radio rather than a
+click — with the rig on 20 m SSB and both locked, a hundred spots on other bands
+show nothing.
+
+Filter selections are remembered between sessions; a fresh installation starts
+with none set.
+
+## Spotting
+
+When you send a spot, the comment carries the mode and then the **award
+references you have assigned** to that contact — the ones on the QSO, not the
+ones OpsLog would compute for it.

wiki/Maps-and-Antennas.md

@@ -23,14 +23,41 @@ portable contacts. No API key.
 ## Rotators
 
 An azimuthal-equidistant **click-to-turn** compass; the current heading and
-target are shown. Three rotator types (Settings → Rotator):
+target are shown. The needle follows a turning antenna smoothly — OpsLog polls
+faster while the position is changing and backs off once it is parked.
 
-- **PstRotator** — via its UDP interface.
-- **microHAM ARCO** — native, over LAN or USB: set the ARCO's CONTROL PROTOCOL
-  to *Yaesu GS-232A*.
-- **GS-232A (generic)** — any GS-232A controller, ERC (Easy Rotor Control)
-  included; serial speed selectable. Set an ERC to GS-232 emulation, not
-  Hy-Gain DCU-1.
+**Two rotors** can be configured (two towers, or a Rotator Genius driving two).
+A Rotor 1 / 2 selector appears on the compass; OpsLog turns and displays the
+one you pick.
+
+### Types (Settings → Rotator)
+
+| Type | Connection | Notes |
+|---|---|---|
+| **PstRotator** | UDP | Enable PstRotator's UDP listener (Setup → Communication → UDP). |
+| **Rotator Genius** (4O3A) | TCP, port 9006 | Native. *Rotator #* picks which of the two the box drives; *Two rotors* adds the second as its own rotor. |
+| **microHAM ARCO / GS-232A** | LAN or USB | Set the controller's CONTROL PROTOCOL to *Yaesu GS-232A*. An **ERC** must be in GS-232 emulation, **not** Hy-Gain DCU-1. |
+| **Hy-Gain DCU-1** | COM port or serial-over-IP | RotorCard DXA, Idiom Press Rotor-EZ, Green Heron. Azimuth only. A DCU-1 is 4800 baud; others may differ — match the controller. |
+| **SPID / AlfaSpid** | COM port | Native, so PstRotator is not needed in between. See below. |
+
+### SPID / AlfaSpid
+
+Pick the dialect to match the controller — they are not interchangeable:
+
+- **Rot2Prog** — RAS, BIG-RAS/HR, MD-01, MD-02. Azimuth **and** elevation.
+  600 baud.
+- **Rot1Prog** — the older azimuth-only controllers (RAK, RAU). 1200 baud.
+
+The two use different frame lengths, so a wrong choice shows up at once: the
+**Test** button reports a reply of the wrong length rather than leaving you to
+find out when the antenna turns oddly.
+
+Two towers means two controllers, each on its own COM port and each a separate
+rotor in OpsLog.
+
+> **The Test button on a SPID, an ARCO or a DCU-1 only READS the heading** — it
+> confirms the port, the speed and the dialect without moving anything, and
+> says so. (PstRotator and a Rotator Genius are instead sent a move to 0°.)
 
 ## Motorized antennas (Ultrabeam / SteppIR)
 

wiki/Profiles-and-Databases.md

@@ -51,3 +51,13 @@ This is also what powers [[Multi-Operator Live Status]].
 Optional **database + ADIF backup at shutdown** (Settings → Backup). Regardless,
 copying the `data/` folder backs up your settings and local logbook. See
 [[Settings and Data]].
+
+## What follows the active profile
+
+Every setting is per profile, station hardware included. Switching profile
+reconnects the **CAT link**, the **amplifiers**, the **motorized antenna**, the
+**Antenna Genius** and the **tuner** using the new profile's settings — so an
+operator with, say, an SPE on COM9 for HF and another on COM10 for 6 m gets the
+right one by switching profile, with no visit to Settings.
+
+The Winkeyer is deliberately left alone: you connect it explicitly.

wiki/QSL-Management.md

@@ -34,6 +34,38 @@ auto-refreshes the **award** stats.
 - **LoTW upload** goes through **TQSL** (ARRL's signer) — set the TQSL path and
   station location in Settings.
 
+## Paper QSL: sent, received, and how
+
+The QSL Manager's paper form sets four things on a batch at once — the sent
+status and date, the received status and date — plus **how the card travelled**,
+in each direction.
+
+Two ADIF fields are involved and they are **not** the same thing:
+
+| Field | Holds | Example |
+|---|---|---|
+| `QSL_VIA` | the **manager's callsign** | `M0OXO` |
+| `QSL_SENT_VIA` / `QSL_RCVD_VIA` | **how the card travelled** — an enumeration | `B` bureau · `D` direct · `E` electronic |
+
+Some loggers write the routing into the manager field, and OpsLog's own older
+versions wrote the words "Bureau" / "Direct" / "Electronic" there too. If your
+log has a mixture, OpsLog offers a **repair** — it tells you how many QSOs are
+affected **before** it changes anything, and it only moves values that are
+unmistakably a routing method (a manager is a callsign, never one of those six
+words).
+
+On **import**, the two fields are kept apart and nothing is invented: a file
+without a routing gives an empty routing. If you see `E` on every imported QSO
+it came from the file — Log4OM writes `QSL_SENT_VIA:E` on every record whether
+a card was ever sent. To clear them, filter the imported QSOs and use
+**Bulk edit field** with an empty value ([[Recent QSOs and Filters]]).
+
+## Confirmed means Y or V
+
+A LoTW confirmation can come back as **V** (verified) rather than **Y**. Both
+count as confirmed — in the awards, the band/mode matrix, the slot statistics,
+the row colours and the QSL Info tab alike.
+
 ## E-mail eQSL
 
 Right-click a QSO → **Send eQSL by e-mail** via the configured SMTP account. The

wiki/Recent-QSOs-and-Filters.md

@@ -2,29 +2,64 @@
 
 ## The Recent QSOs grid
 
-The main log table. Double-click a row to edit; right-click for bulk actions
-(see [[Logging Basics]]). Columns are configurable — click **Columns** to choose
-which are visible; widths, order and hidden columns are saved (per profile),
-with a separate layout for the narrower Main-tab pane.
+The main log table. Double-click a row to edit; **right-click** for everything
+else (see [The right-click menu](#the-right-click-menu) below). Columns are
+configurable — click **Columns** to choose which are visible; widths, order and
+hidden columns are saved (per profile), with a separate layout for the narrower
+Main-tab pane.
 
 - The selected-row count is always visible at the top, and a **Select all /
   Unselect all** button takes every displayed row.
 - QSL and upload status columns are coloured: **Y** green, **N** red, **R**
-  blue.
+  blue, **V** (verified) green — a LoTW contact marked *V* counts as confirmed
+  everywhere, like *Y*.
 - A **Distance (km)** column (also in Worked-before) is computed from the QSO's
   own locator pair.
 - There are **no per-column header filters** — they only ever searched the rows
   on screen, which made a QSO further back in the log look missing. Use the
   advanced filter, which queries the whole logbook.
 
+A QSO logged from **WSJT-X, MSHV, JTDX or a net** appears here immediately, not
+when its auto-upload eventually goes out. That matters if you use a *delayed*
+upload to leave yourself time to correct a contact before it is sent.
+
 ### Award columns
 
 Each defined award can show a column with the reference the QSO counts for.
 These columns are **hidden by default** and opt-in from the Columns picker.
 
+---
+
+## The right-click menu
+
+Select one or more rows, right-click, and the menu acts on **the whole
+selection**. Right-clicking a row that is not selected selects it first, so a
+single right-click on one QSO always does what it looks like it will.
+
+| Entry | What it does |
+|---|---|
+| **Fix country & zones from cty.dat** | Re-derives DXCC, continent, CQ and ITU zones from the prefix database. Use after a cty.dat update, or on QSOs imported without zones. |
+| **Update from the callsign databases** | Re-queries QRZ.com / HamQTH per QSO. A network round trip each — a progress bar appears. |
+| **Update from ClubLog (exceptions)** | Applies Club Log's exception list: callsigns whose real entity is not what the prefix says. |
+| **Update US county from the ULS database** | Re-derives the county of US contacts and **replaces** what is stored. Only appears once the county database is downloaded — see [[Awards]]. |
+| **Send OpsLog QSL by e-mail** | Renders the [[QSL Card Designer]] card for that QSO and mails it. |
+| **Bulk edit field…** | One field, one value, on every selected QSO. Leaving the value **empty clears** the field. |
+| **Export selected to ADIF / Cabrillo** | Just the rows you picked. |
+| **Export selected — choose fields…** | Same, but you pick which ADIF fields go in the file. |
+| **Export filtered view to ADIF / Cabrillo** | **Every QSO the current filter matches** — not only the rows on screen, and with no row limit. |
+| **Send to QRZ / Club Log / HRDLog / eQSL / LoTW** | Uploads the selection now, whatever the automatic upload mode is. |
+| **Delete…** | With a confirmation naming the count. |
+
+> **Exporting a filtered search is two different entries.** *Export selected*
+> takes the rows you highlighted. *Export filtered view* takes the whole result
+> of the filter, including the QSOs scrolled off the bottom. When you have just
+> built a filter and want the lot, it is the second one.
+
+---
+
 ## Advanced filter builder
 
-Click the filter button to open the **QSO filter builder** (Log4OM-style):
+Click the filter button to open the **QSO filter builder** (Log4OM-style).
 
 - Add one or more **conditions**: *field · operator · value*.
 - Operators: equals, not-equal, contains, starts/ends with, greater/less
@@ -32,12 +67,84 @@ Click the filter button to open the **QSO filter builder** (Log4OM-style):
 - Fields include every confirmation **status and date** (sent / received,
   before / after) for paper QSL, LoTW, eQSL, Club Log, HRDLog and QRZ.com.
   Field names stay in English — they are ADIF names, a standard vocabulary.
-- Join conditions with **ALL (AND)** or **ANY (OR)**.
+- Join the conditions with **ALL (AND)** or **ANY (OR)**.
 - Save named **presets** (they travel with the `data/` folder).
 
-A filter shows **every** match (the on-screen row limit only applies to the
-unfiltered log; safety cap 10 000). The filtered view can be exported:
-**selected rows** or the **whole filtered view** → ADIF or Cabrillo.
+A filter shows **every** match — the on-screen row limit only applies to the
+unfiltered log (safety cap 10 000).
+
+### Worked examples
+
+Each of these is a few clicks in the builder. The point of the last column is
+what you then do with the result.
+
+**1. Everything worked on 20 m in FT8 that is still unconfirmed on LoTW**
+
+| Join | Field | Operator | Value |
+|---|---|---|---|
+| ALL | `band` | equals | `20m` |
+| | `mode` | equals | `FT8` |
+| | `lotw_qsl_rcvd` | is empty | |
+
+→ right-click → **Export filtered view to ADIF**, and you have the file to
+re-upload or to check against LoTW.
+
+**2. Paper cards sent but never answered, older than a year**
+
+| Join | Field | Operator | Value |
+|---|---|---|---|
+| ALL | `qsl_sent` | equals | `Y` |
+| | `qsl_rcvd` | is empty | |
+| | `qso_date` | less than | `20250816` |
+
+→ select all → **Bulk edit field** to mark them for a bureau chase, or export
+them for a follow-up list.
+
+**3. Every contact with one entity, on any band, in a date range**
+
+| Join | Field | Operator | Value |
+|---|---|---|---|
+| ALL | `dxcc` | equals | `291` |
+| | `qso_date` | greater or equal | `20260101` |
+| | `qso_date` | less or equal | `20261231` |
+
+→ this is the shape for an award submission. Save it as a **preset** so next
+year is one click and a date change.
+
+**4. Anything that might be a duplicate of tonight's run**
+
+| Join | Field | Operator | Value |
+|---|---|---|---|
+| ALL | `call` | contains | `DL` |
+| | `band` | equals | `40m` |
+| ANY | `qsl_sent` | equals | `Y` |
+| | `lotw_qsl_sent` | equals | `Y` |
+
+Note the mix: the **ALL** block narrows to the contacts you care about, the
+**ANY** block accepts a QSO confirmed by either route.
+
+**5. US contacts with no county recorded**
+
+| Join | Field | Operator | Value |
+|---|---|---|---|
+| ALL | `dxcc` | equals | `291` |
+| | `cnty` | is empty | |
+
+→ select all → right-click → **Update US county from the ULS database**.
+
+### Tips
+
+- **Dates are ADIF dates**: `YYYYMMDD`, no separators. `20260816`, not
+  `16/08/2026`.
+- **`is empty` is not the same as `equals N`.** A QSL that was never sent is
+  empty; one explicitly marked *not sent* holds `N`. Filters on confirmation
+  usually want *is empty*.
+- To find what a field is called, show its column in the grid: the Columns
+  picker and the filter builder use the same ADIF names.
+- A preset saves the conditions, not the results — it re-runs against the log
+  as it is today.
+
+---
 
 ## Worked-before matrix
 

wiki/Troubleshooting.md

@@ -76,3 +76,40 @@ selectable, with a copy button.
 
 Open an issue (or *Send log to F4BPO*) with: what you did, what happened, your
 radio/model, and the relevant log lines.
+
+## The rig disconnects during digital transmissions
+
+Symptom: with WSJT-X or JTDX set to **Fake It** split, the CAT link drops on
+every over. Fake It shifts the dial for each transmission instead of using
+split, so it changes the radio state twice per cycle.
+
+Check that **no inbound Connections row shares a port with an outbound one**.
+If it does, OpsLog receives what it publishes, and its own frequency broadcast
+can come back as a command to re-tune — a loop that fires on every dial shift.
+The log names the clash at startup. Give one of the two rows another port.
+
+## A motorized antenna does not follow the rig
+
+Look for this line in the log:
+
+    antenna: steppir connected, but TRACKING IS OFF in Settings — it will not
+    follow the rig, and only moves when you tune it by hand
+
+That is not a fault. Tracking is a setting (Settings → Antenna), and with it off
+the antenna only moves when you tune it yourself.
+
+If the antenna genuinely stops answering, the log says so and reconnects on its
+own:
+
+    steppir: status query failed, reconnecting: read status: timed out …
+    steppir: reconnected after 2 failed attempt(s)
+
+## A rotator turns the wrong way
+
+On a **SPID**, check the dialect first: **Rot1Prog** for the azimuth-only
+controllers (RAK, RAU) and **Rot2Prog** for the RAS / BIG-RAS / MD-01 / MD-02.
+They use different frame lengths, and the wrong one is the usual cause of a
+heading that reads correctly while every commanded move goes the long way round.
+
+Test with a **small** move first — from 45° to 50°, not to 180° — so a wrong
+result cannot wrap the coax at the top of the tower.

wiki/_Sidebar.md

@@ -27,6 +27,7 @@
 - [[Contest Logging]]
 - [[Net Control]]
 - [[Multi-Operator Live Status]]
+- [[Connections]]
 
 **QSL & Awards**
 - [[Awards]]