hamradiotools

Nexus 1.24.0

Released 11 November 2026. This is the current release. Download it · Every version

Nexus can now hold your audio back while an amplifier switches, so the first moment of every over reaches the antenna through a settled relay. A new Best times box tells you whether you can work a station now, on which band, and when the next good window opens. Predictions now use the P.533 model by default, with your own power and antenna.

When a spot or your callbook names a station's island group, the IOTA reference is now logged for you and sent with the contact. LoTW reports and eQSL InBox files you download by hand are read as confirmations, not as new contacts. A contact you delete stops going to every service it has not reached yet.

Most of this release's fixes concern Stop TX and the end of an over. Stop TX silences the tones at once, stays exactly where it is on screen, and works on the first click over a dialog or an open menu. A slow radio no longer holds your transmitter past the end of an over.

Added

  • Amplifier TX delay

    Nexus can wait 0 to 500 ms after keying before sending audio (Settings ▸ Radio ▸ Amplifier), so an amplifier keyed from the radio has time to switch. It is off unless you set it.

    Full details

    An amplifier TX delay. Settings ▸ Radio ▸ Amplifier ▸ Amplifier TX delay, 0 to 500 ms in steps of 10, set per radio. After Nexus keys the radio it waits this long before sending any audio, so an amplifier keyed from the radio's TX or linear jack has time to switch before RF reaches it. An over that starts later is held that much longer, never past its slot. It is 0 (off) unless you set it, and at 0 nothing changes. It works wherever Nexus makes the audio: the FT modes and the other digital modes, the computer or remote microphone on Phone, the voice keyer, soundcard CW, and Tune when Nexus makes the tune carrier. It cannot help with FM, AM, FSK RTTY, CW the radio sends itself (an external keyer, the serial key line or the radio's keyer), the radio's own microphone, VOX, Tune on a Flex or TCI radio that makes its own tune carrier, or an amplifier keyed by RF sense; for those, use the radio's own TX delay. A program transmitting through Nexus's TCI server gets no delay from Nexus; use that program's own Tx delay. FT8 already has half a second of quiet at the start of each over, so a delay changes nothing there unless the radio is slow to key; then the first tones are left silent, as WSJT-X does, and the over still starts on time. FT4 keeps WSJT-X's own timing and ignores the setting, as WSJT-X does. JS8 keeps its slot clock the same way. FT2 starts its tones later by the delay, which the other station's decoder allows for. FT1 already has 0.3 seconds of quiet at the start of each over, so a delay up to that changes nothing there; a longer one starts the tones later and trims the end of each over at the slot boundary, and on a good signal it still decodes. Stop TX during the delay unkeys at once with nothing sent. On Phone, letting go of PTT keeps the radio keyed until the speech you had already spoken has been sent. A computer driving the station from afar shows the setting as set at the station. Tested in software only; not yet tried on the air with an amplifier.

  • wfview's virtual port as your CAT port

    Type the virtual port wfview (or socat) makes into Serial Port and Nexus opens it. Every real USB or serial port still has RTS and DTR held low, as before.

    Full details

    A virtual serial port, such as wfview's, can be your CAT port. Type the port wfview makes into Serial Port, ~/Downloads/rig-pty1 on a Mac or ~/rig-pty1 on Linux (Nexus fills in the ~ itself), pick your Icom's own Rig Model, and Save. A socat link on Linux works the same way. Before, Hamlib refused to open such a port, because Nexus asked it to hold RTS and DTR low and a virtual port has neither line. Now Nexus recognises a virtual port from what the system says the file is, without opening it, and only such a port is opened without the hold. Every real USB or serial port, and anything Nexus cannot positively identify as virtual, still has RTS and DTR held low exactly as before, so an interface that keys on them cannot key your radio when the port opens. Turn on Native Icom CI-V too, and Nexus opens the port itself for rig control and the radio's own scope, if wfview passes the scope on (the Icom guide has a one-line check). It sets no baud rate on a virtual port, since there is none to set, and it never sends the scope-off command there, so that wfview's waterfall can keep running. When the CAT link drops, Nexus reopens the port by itself once it is there again. Test CAT says plainly that a virtual port has no baud rate instead of trying other rates and blaming the port, and Save no longer warns that a typed virtual port "is not connected right now". A real port typed by a name the port list doesn't show, such as a Linux /dev/serial/by-id/… link, now reconnects by itself after a CAT drop too, with its lines held low as always. Tested on Linux against a simulated radio behind a virtual port, with Hamlib 4.6.5 and 4.7.1 and with Native Icom CI-V; not yet tried with wfview itself, on a Mac or on Linux, so whether the scope comes through and how wfview's display behaves are still to be seen.

  • QO-100 in the band menu

    Phone and CW's band menu now ends with QO-100, which opens the Satellites screen on it. It tunes nothing by itself and appears only when Satellites is on.

    Full details

    QO-100 in the Phone and CW band menu. The band menu on Phone and CW now ends with QO-100 satellite. It opens the Satellites screen on QO-100 (listed there as ES'HAIL 2), where you pick its narrowband transponder as before: SSB and CW, the 13 cm and 3 cm licence check, and the contact logged on 3 cm. It tunes nothing by itself. With SDR Console, which works out the uplink itself, keep the satellite uplink mapping at its default so Nexus leaves the uplink alone. The entry is there only when Satellites is switched on in Settings.

  • Memories reopen their screen

    A memory now recalls to the screen you saved it from, so an SSTV memory brings you back to SSTV. The editor's new Opens on box shows or changes it.

    Full details

    A memory opens the screen you saved it from. Save 144.8875 FM from SSTV and Tune brings you back to SSTV on that frequency, not to Phone. It works for FT, CW, Phone, RTTY, PSK, JTTY, JS8, SSTV and APRS. To save from a screen with no memory strip, go from it to Memories and press + Save. The strip on Phone and CW does the same. Saving a frequency you already have from another screen moves it to that screen. The memory editor's new Opens on box shows the screen and lets you change it; memories you already have stay on By its mode and recall just as they did. A memory falls back to its mode if its screen is switched off in Settings, or if its mode was later changed to one that screen can't work. An older version of Nexus still loads the file, keeps these memories and ignores the screen, though saving from that version drops it. Tuning uses each screen's usual tune, so nothing new is sent to the radio. A memory recalled to SSTV, RTTY, PSK or JTTY tunes and leaves Enable TX as it was, so it stays off if it was off, and you turn it on when you mean to send; a Phone or CW memory arms as it always has, and opening any screen by its tab is unchanged. With "tune when SSTV opens" on, a recalled SSTV memory stays on its own frequency.

  • Large TX meters pane

    ⊞ Panels ▸ "TX Meters, large" on FT, Phone and CW shows power, ALC and SWR as big needle meters you can size. It starts hidden and only shows readings.

    Full details

    A large TX meters pane you can size. ⊞ Panels ▸ "TX Meters, large" on FT, Phone and CW shows power, ALC and SWR as big needle meters with a numbered scale and fine ticks, all of which grow with the pane, whichever way you draw the meters in the strip. Drag the dividers beside it (and the one under it, where another list stands below) to make it bigger; the size is remembered. It starts hidden, so your layout does not change. The small button after the meters on the Phone and CW transmit strip opens it, and on FT so does "Show as a pane" in the larger meters dialog. It only shows readings: COMP, PROC and every transmit control stay where they were.

  • One click to POTA's upload page

    After exporting an activation, Open POTA's upload page opens it in your browser and shows the file. Nexus holds no POTA login and sends nothing to POTA.

    Full details

    One click from an activation's file to POTA's upload page. After Logbook ▸ Export activation writes a POTA file, Open POTA's upload page appears beside it. It opens POTA's My Log Uploads page in your browser and shows the file in its folder, on Windows, Mac and Linux, with nothing to set up. You then log in there and drop the file on the page, or choose it. Nexus holds no POTA login and sends nothing to POTA, so POTA's page is what tells you the log arrived. With no internet the file just waits in Downloads. If the file was moved, or no browser or file manager will open, Nexus says what to do instead. The Stop checklist is unchanged.

  • Best times to work a station

    A new Best times box on Conditions, or right-click a call almost anywhere, says whether you can work that station now, on which band, and when the next good windows open.

    Full details

    Best times to work a station. A new Best times box on Conditions: type a call (or a grid square), or right-click a call almost anywhere (the Call Roster, Band Activity, Spots, Needed, Chase, the chase feed, POTA/SOTA, DXpeditions, the Logbook, a band map label, the map's selected station, a roster card, a JS8 station row) and choose Best times at the top of the menu, or press Best times on the station card, and Nexus says whether you can work it now and on which band, until about which hour, the next good windows over the coming day in UTC and your own time, and which days this week a Kp forecast reaches the path. It says what the answer rests on: the station's grid and where Nexus got it, or "centre of the country, rough" (then it asks your callbook once), and the model, its month, the mode, and the power and antenna gain it assumed. It never gives a percentage, leaves out bands your licence keeps you off in that mode, and asks nothing on the air. During a flare it skips every band estimated to lose 10 dB or more right now and says which one it skipped. When the callbook answers a portable call with the home call's address, Nexus uses it only if it is in the same country, and then says it is the home address and the station may be elsewhere; in another country it uses the country's centre, marked rough. Under 300 km it gives no answer, because that close the model calls nearly every band open all day. The menu item opens a small card beside the call that never covers Stop TX, and its See full day opens Conditions in its own window, so your operating screen stays put. Your current Conditions layout gets the box once, as a tab, and keeps its name; if you remove the box, it stays removed.

  • S-meter peak hold

    On Phone the S-meter keeps its highest reading marked for a second after a signal drops. (#485)

    Full details

    The S-meter holds its peak. On Phone, the highest reading stays marked on the S-meter for a second after a signal drops, then falls back, the way the needle TX meters already hold theirs. (#485)

  • Caller's grid on a map

    Click the grid square on a caller's card to open it on OpenStreetMap. Only the square goes to the website. (#483)

    Full details

    A caller's grid square opens on a map. Click the grid square on a caller's card to open that square on OpenStreetMap in your browser. Only the square goes to the website; Nexus itself downloads nothing. (#483)

  • Volume from the top bar

    While Nexus plays receive audio, click the RX level meter in the top bar for a volume slider on every screen. (#468)

    Full details

    Listening volume in the top bar. While Nexus plays receive audio on this computer (Settings ▸ Radio ▸ Receive audio on this computer), clicking the RX level meter in the top bar opens a volume slider on every screen, so changing how loud the band is no longer means opening Settings. It is the same Listening level setting. (#468)

  • RTTY baud rate on the RTTY screen

    A picker next to Arm RX switches between 45.45, 50 and 75 baud. It can't be changed while you transmit. (#488)

    Full details

    Pick the RTTY baud rate on the RTTY screen. A picker in the receive pane, next to Arm RX, switches between 45.45, 50 and 75 baud, the same setting as in Settings ▸ RTTY. It cannot be changed while you are transmitting. (#488)

  • Build a log from a LoTW report

    After importing a LoTW report, you can add its unmatched confirmations as new contacts. Nothing is added unless you press the button, and Undo takes them out.

    Full details

    Build a log from a LoTW report. After Import of a LoTW report finds confirmations that match no contact in your log, the import window offers Add N unmatched rows as new contacts. Nothing is added unless you press it. Each contact it adds comes in as LoTW has it, confirmed on LoTW with its credit, and is marked as already on LoTW, so it is never uploaded back. Contacts already in your log, and a second press, add nothing, and the toast's Undo takes the added contacts out again. For an operator whose only log is on LoTW.

  • Clear stray eQSL "QSL sent" marks

    Earlier versions copied eQSL senders' "QSL sent" marks onto your contacts. A Logbook button lists them, you untick any you set yourself, and Clear removes the rest.

    Full details

    Check the "QSL sent" marks eQSL cards left on your contacts, and clear them. Earlier versions copied the sender's own "QSL sent, electronic" mark from each eQSL card onto your contact. While a contact still holds that mark (electronic, with no date, beside an eQSL card), the Logbook has a Check eQSL sent marks button saying how many. It lists each one with its call, date, band, mode and mark, all ticked: untick any you set yourself in another program, and Clear takes the mark off the ticked contacts and changes nothing else on them. A mark you set in Nexus carries a date and is never listed. The notice after a clear has an Undo that puts the marks back as they were. Nothing is uploaded, and a later import of a file that still holds a mark you cleared does not put it back. With a long list, the first Tab offers "Skip to Cancel and Clear".

  • eQSL messages and AG

    eQSL cards now bring the sender's message and fill in a missing grid. Authenticity Guaranteed cards show as AG in the Logbook's QSL column.

    Full details

    eQSL cards bring the sender's message, and AG shows in the Logbook. A sync, Check confirmations or an imported InBox file now keeps the message the sender wrote to you on the contact the card confirms, as a message you received, never as the message on your own card, and apart from any paper card's message another logger kept. A contact with no grid takes the sender's grid from the card. The Logbook's QSL column shows AG in place of E for an Authenticity Guaranteed eQSL, with the day the card arrived in its tooltip, and a contact's card shows the AG mark and the message. A callsign correction, or Check confirmations taking a card off a contact, takes the card's date, AG mark and message with it.

  • Keeping good time without internet

    A new guide section explains why FT8 and FT4 need a clock within about a second, and how to keep it right off the grid: syncing before you leave, a phone hotspot or GPS.

    Full details

    The guide explains how to keep good time with no internet. A new section in Scenarios & troubleshooting says why FT8 and FT4 need a clock within about a second of UTC, and what Nexus already does: it corrects its own FT timing from the clock check and keeps doing so for a while after the internet goes. It then covers syncing before you leave, a phone's hotspot, and a GPS receiver, and has a paragraph for clubs on setting every laptop before the event.

  • Large TX meters on SSTV

    The large TX meters pane now works on SSTV too, above the gallery, so you can watch the drive through a whole picture. (#479)

    Full details

    The large TX meters pane on SSTV. ⊞ Panels ▸ "TX Meters, large" now works on SSTV as it does on FT, Phone and CW: power, ALC and SWR as big needle meters, above the gallery, so you can watch the drive through a whole picture. It starts hidden, and the small button after the meters on the SSTV transmit strip opens it. It only shows readings; Stop TX, Tune and the Send/Stop bar stay where they were. (#479)

  • Only new parks and summits

    An Only new chip on the POTA/SOTA board shows just the parks and summits you have never logged. It is off at first and remembered.

    Full details

    Only new parks and summits on the POTA/SOTA board. A new Only new chip beside Hide worked today shows just the parks and summits you have never logged (the ones marked NEW PARK), on POTA, SOTA, WWFF or All, in the main view, its pop-out and the Connect box. Parks in your imported Hunted Parks list count as logged. While it is on, the chip shows how many activators it is hiding, and an empty board tells you it was the chip. It is off at first and Nexus remembers your choice. The chip is not shown when the board cannot tell which parks are new, such as a station you are only watching.

  • TX meters on RTTY, PSK and JTTY

    These screens' transmit strips now show power, ALC, SWR and COMP from a medium-sized window up, so you can set audio drive against ALC. (#489)

    Full details

    TX meters on RTTY, PSK and JTTY. Their TX strips now show the same read-only meters as SSTV and JS8 (power, ALC, SWR and COMP, whichever your radio reports), from a medium-sized window up, so you can set the audio drive against the radio's ALC. When a CAT outage has turned transmit off, the words that say so ("CAT lost: press TX On", then "CAT is back: press TX On") stand where the meters are, on these screens and on SSTV and JS8, which had shown only "TX off" there from a medium-sized window up. Stop TX and the other strip controls stay where they were. (#489)

  • Field Day bonuses per unit

    Emergency power, youth and message handling now score per transmitter, operator or message: type the count beside the box. GOTA contacts count from the club log, and two new bonus boxes are added.

    Full details

    ARRL Field Day bonuses that count per unit score per unit. Emergency power is 100 points per transmitter, up to 20 (rule 7.3.1); youth participation 20 per young operator, up to 100, or 40 for a Class B station (7.3.15); message handling, new, 10 per formal message, up to 100 (7.3.6). Type the number beside the ticked box. The club's score adds five points for each contact the GOTA station makes (7.3.13.1), counted from the club log with nothing to tick. Two new boxes: the GOTA coach (100, 7.3.13.2) and site responsibilities (50, 7.3.18). A box ticked before this update counts as one unit, so Emergency Power still scores 100. Two older ticks change: a ticked Youth box now scores 20, one young operator, until you type how many; and the old GOTA tick adds nothing, because GOTA points now come from the club log. The values come with the rules data, and older Nexus versions keep scoring each box once, as before.

  • Field Day natural power list

    Tick "On natural power now" and contacts logged meanwhile are marked; Export ▸ Natural power saves the list. Each laptop exports its own.

    Full details

    The natural power list ARRL Field Day asks for (rule 7.3.8). Tick "On natural power now" beside the Natural Power bonus, and each contact this laptop logs while it is on is marked. Export ▸ Natural power saves the list. Each laptop exports its own list; club sync does not carry the mark.

Changed

  • SQLite 3.54 security update

    The logbook's database engine moves to SQLite 3.54.1 for a security fix. Your log is unchanged and there is nothing to do.

    Full details

    The built-in SQLite is updated to 3.54 for a security fix. The logbook's database engine moves from SQLite 3.50 to 3.54.1, which fixes a flaw where a specially made pair of database files could make the program read leftover memory. Your log, its format and how it behaves are unchanged, and nothing needs doing.

  • P.533 is the default predictor

    Predictions now use P.533 with your power and antenna, so some band and hour answers change after updating. In a geomagnetic storm it is lowered by the same storm penalty as before.

    Full details

    P.533 is now the default prediction engine. It uses your station power and antenna gain, models the long path and gives per-mode answers, so some band answers change after updating: Band Outlook, the path card, the region strips and the DXpedition windows may name a different band or hour than before. It is a monthly median, so in a geomagnetic storm Nexus lowers its hours by the same storm penalty the fast heuristic always used: the current hour from today's Kp, later hours from SWPC's Kp forecast, and an hour with no forecast is left alone. The pane says when it has done this (a storm from Kp 5, unsettled conditions below), and which power it assumed: your Station power setting, the power your radio measured on your last transmission, or 100 W when neither is known. Live spots still win over any model. Band Outlook's words now follow the modes a band carries, so Excellent means SSB is usable most days at its best hour and a band no longer reads Excellent just because FT8 gets through, and the outlook to DX reads what a band reaches toward at least half the directions; the Chase pane's "call it now" follows the same words. The Band Advisor and the DXpedition cards' "work now" status still use the fast heuristic; the Band Advisor says so. Stations already on the fast heuristic move to P.533 once when they update, and a notice in the pop-up corner says so until you dismiss it. If you want the heuristic, set it in Settings ▸ Services ▸ Prediction engine and it stays. If P.533 cannot run, the pane shows the fast heuristic's answer and says so instead of going blank. Predictions run below normal priority, so a burst of them gives way to decoding.

  • Nexus's path grade stays in Nexus

    The path fields Nexus adds to a contact stay in your log and no longer go to other loggers, uploads, exports or the log.adi copy.

    Full details

    The path Nexus grades for a contact stays in Nexus. The APP_NEXUS_PATH fields Nexus adds to a new contact are kept in your log for Nexus's own use, and no longer go to DXKeeper, HRD Logbook, any upload, any file export (including activation and operator exports) or the log.adi mirror.

  • IOTA reference filled in for you

    When a cluster spot or your callbook names a station's island group, the contact is logged with it. The forms show it filled in, so you can change or clear it first.

    Full details

    The IOTA reference is now logged for you, and sent with the contact. When a DX-cluster spot comment names the station's island group, or your callbook lookup for that exact call does, the contact is logged with that IOTA reference without you pressing anything. The log strip, the Logbook's Add form and the confirm-before-log popup show it already filled in, with the group's name and where it came from, so you can change it or clear it (×) before logging; the Use button is gone. An FT contact logged with no form in front of you takes it too, when Auto-log writes it or Log QSO does with the popup off, and so does a contact logged from the contest entry strip, whose reference (and the group's name in its comment) goes into the Logbook with it when you merge the contest log. The contest strip shows no IOTA box; change the reference in the Logbook after the merge. There a spot of that exact call made within an hour of the contact on its band wins, else the callbook answer already in hand; a station spotted only on another band gets it only when your callbook answer names the same group, and nothing is written when spots disagree. For a portable call (K1ABC/P, F/K1ABC and the like) the callbook answer is the home call's, and it is used only when a spot of the portable call names the same group, never by itself. The log strip, the Logbook's Add form and the confirm-before-log popup also take a spot up to an hour old. Logging never makes a lookup of its own, so with no spot and no answer in yet the contact logs with no reference. A reference typed in an obvious way (eu-5, EU 5, EU-05) is logged as EU-005; anything that is not a reference is shown as such beside the box and the contact is not logged until you fix or clear it. Correcting the call in the popup drops a reference filled in for the call as first copied, and fills in the corrected call's own if it has one; a reference you typed or changed stays. The reference now goes with the contact to every service and logging program (QRZ, Club Log, eQSL, HRDLog, Cloudlog, the LoTW upload, DXKeeper, HRD Logbook) and counts for the IOTA award like one you typed. A reference an earlier version held back for your confirmation (one you took with Use) is no longer held: it goes out the next time that contact is sent, and the contact card's Confirm button and the Awards view's "waiting for you to confirm" count are gone. The contact card says when a reference was filled from a spot or the callbook, and Clear there takes a wrong one out; a LoTW confirmation that names the station's island group still replaces one Nexus filled. To add a reference to a contact that was logged without one, open it in the Logbook (double-click for the contact card, or the pencil for the edit form), type the reference in the IOTA box and save.

  • Island group's name in the comment

    With the IOTA island list on, the group's name is added to the contact's Comment, such as "IOTA EU-005 Great Britain", and shown in the Logbook's IOTA column.

    Full details

    The island group's name goes into the contact's comment and the Logbook. With the IOTA island list on (Settings ▸ Logbook ▸ IOTA island list), a contact logged with an IOTA reference gets the reference and the group's name added to the end of its Comment, for example "IOTA EU-005 Great Britain", after anything already there and separated by "; ". It is added once, and a comment that already names the group is left as it is. The Comment is sent with the contact as before, so the services see the name too. The Logbook's IOTA column now shows the group's name beside the reference. Changing the reference later on the contact card does not change the comment.

  • SWR tip shows your cutoff

    With the high-SWR stop on, the SWR meter's tooltip shows your own cutoff and where to change it. (#476)

    Full details

    The SWR tip names your own cutoff. With "Stop transmitting when SWR is high" switched on, hovering over the SWR meter now shows the number you set and where to change it (Settings ▸ Radio ▸ Transmit limits & sharing) instead of the fixed "keep it under 2:1". With the cutoff off, the tip is as before. (#476)

  • Green "TX on" light

    In Settings ▸ Radio the TX on light shows a green ring while transmit is on, so it no longer looks like TX off. (#486)

    Full details

    The "TX on" light in Settings ▸ Radio is green when transmit is on. It shows a green ring while transmit is switched on, so it no longer looks the same as "TX off". It still turns solid green once Nexus has seen the radio put out power during Tune or Prove TX. (#486)

  • SSTV composer at the bottom

    The picture composer now sits at the bottom of the SSTV screen above Send and Stop, leaving the middle to the gallery. Send, Stop and Stop TX have not moved.

    Full details

    SSTV's picture composer sits at the bottom, beside its options. The Transmit composer has moved from above the gallery to the bottom of the SSTV screen, just above Send and Stop, so the gallery gets the middle of the screen. The picture is on the left and the text overlays, the ident and the warnings are beside it; on a narrow window they go under the picture as before. On a short window the preview is smaller, about a third of the window's height, so the composer no longer takes most of the screen. The waterfall above is sized by dragging its divider, as before (double-click puts it back). Your SSTV panes stay shown or hidden as you had them, and Send, Stop and Stop TX have not moved. The gallery's Reveal button is now Open folder. (#479)

  • FT4 timing matches WSJT-X

    FT4 tones now start 0.2 s earlier, at the same point as WSJT-X, so your DT matches other stations. FT8 is unchanged.

    Full details

    FT4 now starts its tones at the same point in the slot as WSJT-X, so your DT matches other stations. Nexus started its FT4 tones about 0.2 s later than WSJT-X does, so stations hearing you saw your DT read later than everyone else's. Both decoded fine; now your signal lines up with theirs. Your FT4 over also ends 0.2 s sooner, and when audio stops and PTT drops is unchanged. FT8 is unchanged.

Fixed

  • FT1 decoder overflow fixed

    The FT1 decoder no longer writes past its 100-decode limit in a slot, which could have corrupted its memory. A decode past the 100th is dropped.

    Full details

    The FT1 decoder can no longer write past the end of its list of decodes. It keeps up to 100 decodes per slot and did not check that limit before adding another. More than 100 different FT1 messages in one slot is not expected on the air, but a slot that reached it could have corrupted the decoder's memory. A decode past the 100th is now dropped, as a repeat already is.

  • Serial CAT port never opened unprotected

    If Hamlib's tools could not start for a moment, Nexus could open a serial port without holding RTS and DTR low, which can key some interfaces. It now refuses and asks you to try again.

    Full details

    A serial CAT port is no longer opened without RTS and DTR held when Hamlib's tools would not start for a moment. Before opening a serial port, Nexus asks Hamlib whether it will hold RTS and DTR low. If neither rigctl nor rigctld could be started for that question, for example while an antivirus scan held the file, Nexus went on and opened the port anyway once rigctld did start, with both lines left as the driver raises them, which can key an interface wired to RTS or DTR. Now it does not open the port and says to try again. A missing Hamlib still gets the message saying how to install it, and a virtual port or a port with every line set to Untouched is opened as before.

  • Stop TX silences the tones first

    On a radio slow to answer, Stop TX now stops the audio at once and unkeys straight after, as WSJT-X does, instead of sending tones for a second or more.

    Full details

    Stop TX silences the tones at once on a radio slow to answer. Stop TX, and a logger's Halt Tx, used to unkey the radio first and stop the audio only after the radio answered, or Nexus stopped waiting. On a network radio or a slow serial link that wait is about a second, longer if the link has to reconnect, and the over's tones went on being sent for all of it. Now the audio stops first and the radio is unkeyed straight after, as at the end of every over and as WSJT-X does, so any wait for the radio is a silent carrier instead of tones. The unkey is sent no later than before. The same order now applies when a computer driving the station from afar loses its permission to transmit, when you switch radios in the middle of an over, and when the radio connection stops on an error. Tested in software only; not yet tried on the air.

  • FT1 overs start 0.3 s into the slot

    FT1 overs were meant to start after 0.3 s of quiet but began the moment the radio keyed. They now start 0.3 s in, so an over that arrives slightly early still decodes.

    Full details

    FT1 overs now start 0.3 seconds into their slot, as intended. FT1's decoder cannot hear a signal that starts before its slot does, so each FT1 over was meant to begin with 0.3 seconds of quiet, leaving room for a station whose clock runs a little fast. That fix never reached the radio: FT1 overs went out with their tones from the moment the radio keyed, so the other station lost any over that reached it even slightly early. They now start 0.3 seconds after the key. They end before the slot does when the radio keys promptly; a radio slow to key loses the last moment of the over at the slot's end, and on a good signal it still decodes. The other station needs no update to hear them. The cost: the other station's decoder searches only so far into the slot, and the quiet start uses 0.3 seconds of that. An FT1 over now still decodes when it arrives up to about 0.6 seconds late, counting both the time your radio takes to key and any error in your clock, where before it decoded up to about 0.9 seconds late. This has three limits. A radio that takes more than about 0.6 seconds to key, such as some remote or network setups, now loses its FT1 overs, which it used to get through. With weak signals, around -12 dB, some overs fail to decode once the radio takes more than about 0.3 seconds to key. And a Call CQ or double-click during your own FT1 slot now goes out at once only in the first 0.16 seconds or so of the slot; after that it waits for your next slot, at most 4 seconds. Tested in software only; not yet tried on the air.

  • First over on time with older Hamlib

    On the FT-991A, FT-710, FTDX10, FT-450 and FT-450D with a Hamlib older than 4.6, the first over after connecting no longer stalls Stop TX and the unkey for a second or more.

    Full details

    Stop TX and the end of an over are on time in the first over after connecting, on the FT-991A, FT-710, FTDX10, FT-450 and FT-450D with an older Hamlib. With a Hamlib older than 4.6, such as the one Ubuntu 24.04 installs or your own rigctld, Nexus's way of reading these radios' ALC on their own scale got no answer, and each try held everything up for a second or more, three times in the first over after connecting. Stop TX, and the radio unkeying at the end of the over, waited for it. Nexus now asks that question once each time it connects to the radio, while you are receiving, never while you transmit, and on an older Hamlib reads the ALC the way Hamlib gives it from then on. With an older Hamlib, a transmission that starts during that one check, one you start yourself, the first one after you turn TX on, or one Nexus sends by itself in a session still running after a short CAT dropout, can start up to about two seconds late; once it is on the air nothing is delayed. With the Hamlib that comes with Nexus the check takes a moment and the ALC is still read on the radio's own scale, on every mode; on FT2 with TX left on through a CAT recovery, the ALC goes back to the radio's own scale only once TX goes off. Checked against simulated radios, not yet on the air.

  • Slow meter reads no longer hold PTT

    A radio that stalled on a meter reading near the end of an FT8, FT4 or other timed over could keep you keyed past the over. PTT now drops on time.

    Full details

    A radio slow to answer a meter reading no longer holds your transmitter past the end of an over. While you transmit, Nexus reads Po, SWR, ALC and COMP from the radio a few times a second. If the radio stalled on one of those reads near the end of an FT8, FT4 or other timed over, Nexus waited the link's whole timeout (about 1 s on a local link, 2.5 s on a network radio link or a slow serial link) before dropping PTT, so the carrier could run into the other station's period, and a Stop TX pressed then waited too. Near the end of an over each reading now has only the time left before the over ends, and none is taken when too little is left for the radio to answer, so PTT drops on time. A Stop TX pressed earlier in an over, while the radio is stalled on a reading, can still wait up to the link's timeout.

  • Band change before your slot

    With Split Operation = Rig, changing band just before your own slot could start that over late by up to about 4 seconds. The over now goes first.

    Full details

    A band change just before your own slot no longer makes that over start late. With Split Operation set to Rig, Nexus puts the new band's TX frequency on the radio while you receive. If the band changed in the last 4 seconds or so before your own slot with TX on, those writes ran at the start of the slot, ahead of the over, and the over started late by however long they took, up to about 4 seconds on a slow network link, losing its start. Now the over goes first and writes the TX frequency itself, as it did before Nexus set it ahead. As then, on a radio too slow for those writes that over is not sent and TX turns off with a message saying the radio did not take the split TX frequency, rather than the over going out late; Nexus then sets the TX frequency while you receive, so the next over keys normally once you turn TX back on.

  • Blind calls keep your period

    Calling a station you haven't decoded used to switch your Tx 1st / Tx 2nd to match whoever was decoded last. A blind call now keeps your period, as WSJT-X does.

    Full details

    Calling a station you have not decoded keeps your Tx 1st / Tx 2nd period. With other stations decoding on the band, typing a callsign to call it blind (or sending a chat message to a station not yet heard) switched your transmit period to match whoever was decoded last. That station's period says nothing about the one you are calling, so half the time you transmitted on top of them, and it overrode a Tx 1st / Tx 2nd you had picked. A blind call now keeps your period, as WSJT-X does. Clicking a decoded line, a roster entry or a station you have heard still answers on the opposite of that station's period. A line you click after Nexus has stopped holding its decode (about two minutes back on a busy band), from a station not heard since, is no longer answered on a guessed period, in a contest too: Nexus asks you to wait for the station's next transmission and click again, as it already did with nothing decoded.

  • Top bar Stop TX stays put

    On screens without a transmit strip, a status message no longer slides the top bar's TX On, Tune and Stop TX sideways.

    Full details

    The top bar's Stop TX stays put when a status message comes up. On the Logbook, Field Day and the other screens without a transmit strip, where the top bar's TX On, Tune and Stop TX are the only ones on screen, any status message, a recording starting or the TX watchdog appearing in the bar slid that whole group sideways, by as much as about 490 px on a 1920×1080 window, so a press meant for Stop TX could land elsewhere. The messages now sit in the free middle of the bar, after your callsign, so nothing else in the bar moves when one comes or goes. On those screens the bar now always keeps room for an urgent message's warning sign, so one arriving never pushes the bar onto another line either.

  • Stop TX holds when the amplifier answers

    Amplifier controls are now shown from the start, greyed out until the amplifier answers, so their appearing no longer pushes Stop TX to a new spot.

    Full details

    Stop TX no longer moves when the amplifier first answers. With an amplifier set up, its controls in the screen's header appeared only once the amplifier first answered after Nexus started or after you switched radios, which could push the header onto another line and move Stop TX under it. They are there from the start now, greyed out and showing "—" until the amplifier answers, and the Amp pane says "Connecting…" until then. The amplifier's AMP temperature line beside the transmit meters is always there too, with "—" while it has no reading, so neither its first reading nor a dropped link moves Stop TX any more. Setting an amplifier up in Settings, or switching between a radio with an amplifier and one without, still changes what the header holds, so Stop TX can move then.

  • Stop TX holds when CAT drops

    On FT and Phone, CAT going down or coming back no longer moves Stop TX or the transmit strip up or down.

    Full details

    Stop TX no longer jumps when CAT drops. On FT, Stop TX and the whole transmit strip moved up or down by as much as 56 px whenever CAT to the radio went down or came back, so a press meant for Stop TX could land on something else. Now Split, and BW on CW, stay in the header while CAT is down, greyed out, with "No CAT link" when you point at them, and the CAT ✗ pill beside them says the rest. On Phone, "⚠ no rig control", the rig-mode warning and the mic-off-for-DAX note sit at the end of the header line and shorten with "…" when the line is full, with the whole words when you point at them; on a full 1920×1080 header they used to push the line down and move Stop TX 32 px. A transmitter alarm now shows on the transmit strip itself, in amber beside the TX state, instead of as a row of the header above it, which moved Stop TX 26 to 38 px on every screen when an alarm came or went. It takes one line, keeps at least its warning sign, the start of its words and its ✕ (Dismiss) even on the smallest windows at the largest zoom, shortens the rest with "…" and shows the whole alarm when you point at it; "+2" means two more are waiting, and pointing at it shows them. It stays until you dismiss it, as before. The space for it is kept on the strip whether or not an alarm is up, so nothing moves when one arrives. On the operating screens the top bar no longer grows a line for a status message, a recording or the TX watchdog: on a full bar a message may now be cut short there, and every urgent one (PTT off failed, the radio engine or sound card stopped, the TX watchdog halt and the like) is also shown in that same place on the transmit strip, next to Stop TX. On screens with no transmit strip, such as the Logbook and Field Day, an urgent message always keeps its warning sign in the top bar, where room for it is always kept, and a screen reader announces it at once; its words may be shortened there, and pointing at it shows them whole. The TX watchdog's halt is now one of those messages rather than a chip of its own, and it still says to press TX On to transmit again. The amplifier's Operate/Standby button keeps one width, so switching it no longer moves the header either.

  • Stop TX visible on small windows

    On small windows at 150 or 175 % text, Stop TX is now on screen as soon as a screen opens, and SSTV, RTTY and JS8's send bar no longer covers it.

    Full details

    Stop TX is on screen the moment a screen opens, on small windows at large text sizes too. On FT at 150 or 175 % on a small window, Stop TX sat below the bottom of the window until you scrolled down to it. Now the strip's first line, with TX On/Off, Tune and Stop TX, waits at the foot of the window, and the rest of the strip comes up as you scroll. On SSTV, RTTY and JS8 at 900×600 at 175 %, an alarm, a status message or SSTV's dropped-picture message could make the send bar taller than the room left and cover Stop TX. Now the send bar takes only the room under the strip and scrolls inside itself where it has to, so Stop TX is never covered. On the smallest of these windows that room is a single line: what does not fit, Send and the bar's own Stop among them, is still there when you scroll the bar or Tab to it, and the bar scrolls to its own Stop when it squeezes and when you start sending, wherever that one button fits. Where everything fits, nothing changes. A waterfall height you dragged on FT now counts against the whole screen, as it does on the other screens, so it may open a little taller once.

  • Call CQ keys after a band change

    With Split Operation = Rig, a radio slow to answer split commands could refuse the first over after Call CQ on a new band. That over now keys on time.

    Full details

    FT8 and FT4 key after Call CQ on a band change, on a radio slow to answer split commands. With Split Operation set to Rig, Nexus already puts the new band's TX frequency and mode on the radio while you receive, but pressing Call CQ made the next over send them again, just before keying. A radio that took about 170 ms or more per command could not answer all three inside the half second Nexus allows, so the over was refused ("FT8 not keyed: the radio did not take the split TX frequency") and transmit halted. Now that over sends only the split command before it keys, so a radio that takes up to half a second per command keys on time: in testing against a simulated radio, one taking 258 ms per command (775 ms for the three) now keys, on FT8 in either period and on FT4. Nexus still sends the TX frequency and mode again whenever they may have changed: a different mode, a frequency change, a reconnection, or a command the radio did not take. A radio slower than half a second per command is still refused, as before. Fake It and None are unchanged.

  • Saving a CW macro interrupts nothing

    Saving a CW macro key no longer ends a Remote browser's control of the station or drops an FT message waiting to go out.

    Full details

    Saving a CW macro key no longer interrupts anything. It used to save the way Settings does, which ends a Remote browser's control of the station and drops an FT message waiting to go out. It now saves only your CW macro profiles, as RTTY and PSK already did. What you save is still what Settings ▸ CW ▸ Macros shows.

  • Contest CW keys kept separate

    Editing a CW key during a contest now saves to a profile named for that contest. Your everyday keys stay as they were.

    Full details

    Editing a CW key during a contest no longer turns your everyday keys into the contest's. On a profile still on the built-in keys, the first change you saved in a contest made that profile the contest's set, CQ TEST and all, for good. The change now goes into a profile of its own named for the contest (ILQP, Field Day), which the CW screen shows in that contest only; your everyday keys stay as they were.

  • Send once keeps your text

    In RTTY and PSK, if a message doesn't go out, the editor now stays open with your text instead of losing it. CW does the same when the send never reaches the station.

    Full details

    Send once keeps your text when the message does not go out. In RTTY and PSK, when the station turned the message down after Nexus handed it over (TX off, a clock correction holding TX), the editor closed and what you typed was gone. It now stays open with your text, as it already did when the screen itself refused the send, and closes once the message is taken. In CW the same holds when the send does not reach the station; a message the CW keyer takes in and then drops (during a clock correction, say) still closes the editor, as before. While Nexus waits for the answer, pressing Send once again sends nothing more, and Esc stops the transmission rather than only closing the editor.

  • CAT Auto-test waits longer

    Auto-test now gives each radio up to 2.5 seconds to open, so slower ones like the FT-847 and FT-817 are found, and with no model chosen it also tries the FT-847.

    Full details

    CAT Auto-test waits for each radio it tries, and tries the Yaesu FT-847. Auto-test used to ask each radio for its frequency after a fixed 0.7 seconds, before Hamlib had finished opening some of them. Radios whose opening takes longer, the FT-847 and FT-817 among them, could answer and still be reported as not found, even with the rig model already set. It now waits up to 2.5 seconds for each one. With no rig model chosen, it also tries the FT-847 at 4800, 9600 and 57600 baud, and the FT-817 family at 9600 as well as 4800, before the FT-847. The FT-847 test switches CAT on in a way an FT-817, FT-818, FT-857 or FT-897 reads as "lock the dial", so those radios are tried first and answer before it is sent. With no model chosen, an FT-817-family radio may be named FT-857, since they answer alike; as before, you confirm the rig model after Auto-test finds the port. The FT-847 is now in the Rig Model list without turning on "Show all models". The waiting has a cost: a port where nothing answers can now take a minute or more, so Settings and the setup wizard show how long Auto-test has been running, and it does not look stuck. This was checked against emulated radios, not yet on the air.

  • Stop TX works over dialogs

    With a dialog or menu open, Stop TX and the stop keys now work on the first click, and the dialog stays open with your typing in it.

    Full details

    Stop TX stays clickable on top of every dialog and menu. With a dialog open (a confirmation, the Getting started guide, the prompt to log a contact, a Logbook question, the radio picker at launch) or a menu drawn over it, Stop TX, the Esc Stop or Stop key on the RTTY, PSK, JTTY and SSTV screens, and RTTY's Abort now stay on top and work on the first click. The dialog or menu stays open, with your typing still in it, and Stop TX stays exactly where it was. Esc stops as before. Tune, PTT and TX On/Off are not lifted, so they still wait until the dialog closes.

  • Stop TX on the first click past a menu

    With a menu open, one click on Stop TX now stops at once instead of only closing the menu.

    Full details

    Stop TX works on the first click even while a menu is open. With a menu open (Help, the band pill, the ATU button's menu, a decode or station row's menu, a JS8 or Connect ⋯ menu), one click on Stop TX used to close the menu and stop nothing, so you had to click again. Now that click stops at once and closes the menu (a menu drawn over Stop TX stays open, see the entry above), and a click anywhere else outside a menu does what it would do with no menu open. Arrow keys, typing a letter to jump to an item, and Esc work in menus as before.

  • FT-450D power and ALC

    Radios whose Hamlib driver gives power only as a share of the meter now show an estimated PO≈ against your rated power. The FT-450 and FT-450D ALC no longer sits at 100 %. (#489)

    Full details

    The FT-450D shows its output power and its real ALC (#489). On a direct serial link the power meter stayed empty: Hamlib gives this radio's power only as a share of its meter, not in watts, and Nexus asked only for watts. Nexus now reads that share and shows it as that share of the rated power set for the radio in Settings ▸ Radio, marked PO≈ because it is an estimate; it is not written to your log, which keeps using Station power. The same goes for every radio whose Hamlib driver has no watts reading: the FT-450, TS-2000, TS-480, QCX/QDX, (tr)uSDX, SDRuno and SDR Console. The FT-450 and FT-450D ALC meter no longer sits at 100 %: it is read on the radio's own scale, as on the FT-991A, FT-710 and FTDX10.

  • Hamlib 4.5 hangs fixed

    With Hamlib 4.5 from Linux (Ubuntu 24.04's, for one), opening a serial rig no longer hangs, and asking Hamlib about the radio no longer opens your first serial port unprotected.

    Full details

    CAT no longer hangs with Hamlib 4.5, and asking Hamlib about the radio no longer opens your first serial port. With Hamlib 4.5 installed from your Linux distribution (Ubuntu 24.04's, for one) and Nexus using it instead of its own copy, opening a serial rig hung for good: before starting Hamlib, Nexus asks it which control lines it can hold, and Hamlib 4.5 never finished answering. Worse, asking that question made Hamlib 4.5 open the computer's first serial port, /dev/ttyS0 (COM1), with nothing held low, every time CAT opened, so a keying interface on that port could key. Nexus now asks in a way that opens no port on any Hamlib, and the rig opens with RTS and DTR held low as before, about 2 seconds later on Hamlib 4.5. If Hamlib can't give the answer at all, Nexus doesn't open the port, and the CAT status says to try again and then to update Hamlib or use the copy that comes with Nexus, because opening it without the hold could key a transmitter wired to RTS or DTR. A virtual port has no such lines and opens either way.

  • POTA/SOTA board keeps your band

    The band you picked now always keeps its chip, even when nobody is on that band, so one click brings the other bands back.

    Full details

    The POTA/SOTA board always shows the band you picked. The board remembers your band filter, but its band chips listed only bands with activators on them right now. If nobody was on your band, its chip was missing and the list was empty with nothing to say why, and with no spots at all the band row was gone too. The band you picked now always has its lit chip, so one click on it, or on All, brings the other bands back. The same goes for the board's box on Connect.

  • "⊘ not reported" fits

    A meter your radio doesn't report, such as COMP on FT, now shows the full "⊘ not reported" note instead of a cut-off one.

    Full details

    FT's TX meters no longer cut off "⊘ not reported". When your radio reports no COMP reading (FT does not use the speech processor), the meter bars on FT's QSO strip showed the note cut short at every window size and zoom. A meter the radio does not report now shows the note in place of its empty bar and dash, and it fits. The meters take the same space as before, and Stop TX stays where it was. On Phone, COMP keeps its level bar beside the note, so you can still set the processor.

  • MSK144 judged where it transmits

    The licence check now judges MSK144 at its real signal, 1000 to 2000 Hz above the dial, so it can't slip into a CW-only segment. US classes also get Q65 and MSK144 limits.

    Full details

    The licence check judges MSK144 where it transmits, and holds US data to its width and speed limits. MSK144 always transmits 1000 to 2000 Hz above the dial, whatever the TX marker says, but the check judged only the marker, so with the marker dragged up a dial just under 50.1 or 144.1 MHz could send its whole signal into the CW-only segment. It now also judges the signal itself, about 105 to 2895 Hz above the dial, so the lowest 6 m dial MSK144 can use is about 50.0999 MHz and the 6 m top is judged the same way. For US licence classes, Q65-30E and Q65-15D (about 3.5 kHz wide) are refused below 28.3 MHz, where Nexus holds data to 2.8 kHz, and MSK144 (2000 baud) is refused on 2200 m and 630 m, where data may not exceed 300 baud. When the mode is the whole reason, hovering the red frequency readout says which rule and names the mode. The MSK144 and Q65 calling frequencies, every other mode and the Open class are not affected, and nothing is allowed that was refused before.

  • 3 cm no longer opens on the band edge

    Phone and CW on 3 cm used to tune 10000.000 MHz. They now open on the QO-100 narrowband transponder with the Open class, or 10368.100 with a US class.

    Full details

    Phone and CW on 3 cm no longer tune the bottom of the band. Opening Phone or CW on 3 cm with no frequency saved there sent the radio to 10000.000 MHz, the very edge of the band. If SDR Console is set to derive the QO-100 uplink from the downlink, that could have meant an uplink near 1910.5 MHz, which is not an amateur band. Now 3 cm opens on the QO-100 narrowband transponder with the Open licence class (SSB 10489.700, below the 10489.750 middle beacon, and CW 10489.520), and on the 10368.100 calling frequency with a US class. A radio that transmits on 3 cm itself, the IC-905, opens on 10368.100 with the Open class too, so it does not key straight into the satellite's downlink; SDR Console and other SDR programs, and any radio Nexus does not know, keep the transponder. With two radios it is the radio that 3 cm goes to that counts, and a 3 cm frequency remembered on one kind of radio is not brought back on the other kind. The same mistake is fixed wherever a band's first frequency was its own edge. 13 cm opens on the matching QO-100 uplink, 2400.200 SSB and 2400.020 CW, with the Open class, and on 2304.100 with a US class. US classes on 1.25 m, 70 cm, 33 cm, 23 cm and 6 cm now open on the calling frequency (222.100, 432.100, 903.100, 1296.100, 5760.100) instead of 222.000, 420.000, 902.000, 1240.000 and 5650.000, 1.25 cm opens on the 24048.200 narrowband frequency instead of 24000.000, and 9 cm opens on 3400.100. 160 m Phone opens at 1.8458 LSB, with its signal above 1.843 as the ARRL band plan has it, instead of reaching down to the 1.800 band edge. 160 m CW with the Open class opens on 1.836, not on 1.810, the Region 1 band edge. A Novice's 23 cm CW opens inside their 1270–1295 instead of on 1296.100, which they may not use. On 2200 m and 630 m, Phone and CW keep the frequency you are on. 2200 m Phone used to tune to 138.5 kHz, off the band. Your licence privileges and what you may transmit on are unchanged, and a frequency you have tuned on a band is still remembered.

  • HRD Logbook catches up

    Contacts logged while HRD Logbook was closed now wait, up to 256, and go across in order when it opens, instead of all but one being dropped.

    Full details

    HRD Logbook gets the contacts logged while it was closed. With Ham Radio Deluxe's logbook closed and several contacts waiting for it, only one of them stayed queued each time Nexus tried, and the rest were dropped. Now up to 256 contacts wait, and when HRD Logbook opens each one goes once, in the order you logged them. Past 256, the oldest waiting contact is dropped to make room, and nothing in the Connections log says so; you can add it to HRD with Logbook ▸ Export ADIF. A contact logged straight after clearing the log no longer loses its HRD Logbook forward either.

  • PstRotator elevation (#480)

    Since 1.20.0 PstRotator turned only the azimuth on satellite passes. Nexus now reads and sends its elevation again, so satellite and Moon tracks move both axes.

    Full details

    PstRotator moves the elevation again (#480). Since 1.20.0, Nexus treated every program on the "PstRotatorAz / PstRotator (UDP)" rotator entry as the azimuth-only PstRotatorAz, so with PstRotator a satellite pass turned the azimuth and left the elevation where it was. Nexus now asks the program for its elevation. PstRotator answers, so the Rotor pane shows its elevation, satellite and Moon tracks move both axes, and turning the beam keeps the elevation where it is. Elevation goes out in the same message as the azimuth, from 0 to 90°, and Nexus refuses anything higher. If PstRotator stops reporting its elevation, or reports one above 90° (its 180° elevation setup), the compass shows "—" and a beam turn moves nothing, rather than the antenna being lowered or sent to the horizon. A PstRotator that is slow to answer when Nexus starts is not mistaken for one without elevation. PstRotatorAz stays azimuth only, exactly as before. Nothing to change in Settings. This has not been tried on a real PstRotator yet.

  • High passes keep the antenna up

    On an az/el rotator that refuses elevations above 90°, a high pass no longer drops the antenna to the horizon for the rest of the pass.

    Full details

    A satellite pass no longer lays an az/el antenna on the horizon when the rotator refuses an elevation. With Allow flip on, a high pass asks for more than 90° of elevation, and a rotator that only reaches 90° refuses it. Nexus then turned the azimuth alone with a command that also sets the elevation to 0, so the antenna went down to the horizon and stayed there for the rest of the pass. This happened with any az/el rotator that reaches 90°, not just PstRotator. Now, on a rotator that has an elevation axis, those azimuth-only moves keep the elevation where the rotator reports it, and a park that falls back to azimuth does the same. A rotator with no elevation axis is sent exactly what it was before.

  • Spots board remembers your filters

    Your chosen modes and band on the Spots board now survive closing Nexus, instead of every mode coming back on.

    Full details

    The Spots board remembers your modes and band. Choose CW and FT8 and a band, close Nexus, and they are still set next time, instead of every mode coming back on. A mode you never hid and that first turns up later still shows. The Pop out keeps its own: with no mode or band chosen in the main Spots view, it comes back on the ones you last chose in it; otherwise it opens on the view's, as before. A chosen band or hidden mode always has its chip, even with no spots on it right now, so an empty board always shows why. The search, sort and other chips still last only until you quit, and the Spots boxes and panes on other screens keep all their filters only until you quit.

  • DXKeeper gets each contact once

    A failed upload elsewhere no longer sends the contact to DXKeeper again on every retry. Copies already sent stay in your master log until you delete them.

    Full details

    DXKeeper gets each contact once. When an upload to another service failed, for example with the internet down, every retry sent the contact to DXKeeper again, so your master log could collect a dozen or more copies of it. A catch-up of older contacts sent them to DXKeeper again too. Now a contact goes to DXKeeper once, when it is first logged or, for one held during a contest, when it is released. Copies already sent to DXKeeper are not removed, so any extras in your master log stay there until you delete them. Correcting a callsign still sends the corrected contact to DXKeeper once.

  • Contacts wait for DXKeeper

    When DXKeeper isn't running, contacts now wait, across a restart too, and go once each, oldest first, when it answers.

    Full details

    Contacts wait for DXKeeper instead of being lost. If Nexus cannot connect to DXKeeper, nothing was delivered, so the contact waits, along with any you log after it. Nexus looks for DXKeeper a few times in the first minute and then once a minute, without sending anything, and when it answers it sends each waiting contact once, oldest first, one every two seconds. If the connection drops part-way, what was sent is not sent again and the rest keep waiting. They also survive closing Nexus and go after the next launch. The Connections log keeps one line saying how many are waiting, and says when they have all been delivered. A send that connected is never repeated, since DXKeeper does not say whether it arrived: if the connection breaks during a send, a line names that contact so you can check it in DXKeeper. Nexus keeps up to 4096 waiting contacts, shared with those held for a contest; past that, the oldest one waiting is dropped and named, and you can add it with Logbook ▸ Export ADIF.

  • Deleted contacts stop uploading

    Deleting a contact now cancels every upload not yet sent, on every service, including retries and contacts held in a contest. Nexus can't take back one a service already has.

    Full details

    A contact you delete stops going anywhere it has not gone yet. Deleting a contact, clearing the log, or undoing something that took a contact out now cancels every upload not yet sent, on every service: QRZ, Club Log, eQSL, HRDLog, N3FJP, Cloudlog or Wavelog, World Radio League, HRD Logbook and DXKeeper. That covers retries waiting after a failure, contacts held during a contest, and contacts waiting for DXKeeper, and it stays that way after a restart; other contacts are not affected. If Nexus is part-way through sending the contact, the service it is talking to at that moment may still get it, and the rest do not. Nexus cannot take back a contact a service or program already has, and it does not try: it stays there until you delete it there, and the Connections log's earlier line is all that shows it went. The same goes for a LoTW upload already signed and handed to TQSL. Undoing the delete brings the contact back with its upload marks and does not send the cancelled uploads by itself: push it again from the Logbook, or use Logbook ▸ Export ADIF. Undoing a callsign correction puts aside only the corrected copy on the wait for DXKeeper, and redoing it brings that copy back in place of the original, so DXKeeper is owed the call the Logbook shows. A waiting contact a contest's held list pushes out at its limit is now named in the Connections log too.

  • Mid-contest update keeps DXKeeper sends

    Contacts still held from 1.23.0 now reach DXKeeper when released, and ones already sent don't go again.

    Full details

    Updating mid-contest keeps your DXKeeper sends. Contacts still held from 1.23.0 that had not gone out would never have reached DXKeeper after the update. They now go when released, and ones that had already been sent do not go again.

  • Needle TX meters you can read

    The needle TX meters are larger, about twice as wide on a 1080p window, with the name and value beside the dial. Stop TX stays where it was.

    Full details

    The needle TX meters are big enough to read. They were a fixed, tiny size whatever room the window had, with the meter's name and value covering the lower half of the dial. Now the name and value stand beside the dial, in larger type, the dial takes the full height, and the three meters spread into the free space on the TX strip and Operate's strip: on a 1080p window each dial is about twice as wide as it was. On a phone each meter is drawn taller as well as wider. The strips stay the same height and Stop TX stays where it was. On Operate, clicking them still opens them larger.

  • "No decodes yet" fits short panes

    On FT, a short Band Activity or Rx Frequency pane now shows that message from its first line.

    Full details

    "No decodes yet" no longer loses its heading in a short pane. On FT, a Band Activity or Rx Frequency pane too short for that message showed it scrolled past its first line. It now starts at the top, and scrolls for the rest.

  • SDR Console error at every transmit fixed

    Nexus asked SDR Console for a data mode it lacks, raising an error window. It now uses plain USB or LSB for sound card modes when the radio won't take the data mode.

    Full details

    FT8 through SDR Console no longer brings up an error at every transmit. Nexus asked SDR Console for a data mode it does not have, just before keying, and SDR Console answered with an error window. Now, after Nexus itself sets the data mode, it reads back what the radio says while you receive. If the radio refuses it and takes plain USB instead, or stays in plain USB each time Nexus asks (Nexus asks a second time to be sure, while you receive), Nexus uses plain USB (or LSB) for FT8, FT4, RTTY, PSK and the other sound card modes for as long as that connection lasts, with nothing to set up. SDR Console may show its error two or three times when Nexus connects, while you receive, instead of at every transmit. Once a radio has shown it takes the data mode, nothing you do on the radio's own panel changes that: pressing USB there, or a band change that recalls USB, never makes Nexus give up the data mode. Radios whose data mode works under another name (Icom USB-D, DIGU on piHPSDR, the QCX/QDX and SmartSDR or PowerSDR set up as a TS-2000, DATA-LSB on the FT-1000MP) keep it, and the TS-590S/SG and TS-950 are not checked at all yet. CW reverse is not affected. Data modes use plain SSB still works as before on every radio. Nothing extra is sent to the radio for this, and the mode is set at the same moments as before; only the mode Nexus asks for can change.

  • NET rigctl with a COM port explains itself

    A radio set up as NET rigctl with a COM port now gets a plain message saying what to change, instead of "No such file or directory".

    Full details

    NET rigctl with a COM port now says what is wrong. NET rigctl reaches a radio through a rigctld on the network and cannot open a serial port itself, so a radio set up that way failed with "COM4 No such file or directory" unless another program's rigctld happened to be running. Test CAT and the CAT status now say so and what to do: set Rig Model to your radio, or start the program that runs the rigctld, or use Connection: Network with its address. Test CAT also no longer spends half a minute sweeping baud rates for it. Settings shows the same advice as soon as the two are picked, and on Save. Nothing is changed for you: the setup still saves, and still shares a rigctld that is running. A NET rigctl radio with a real network address is not affected.

  • TempoFast copies far more

    In simulation, strong TempoFast frames now go missing under 1% of the time instead of about 7%, and at -10 dB about 9% instead of 43%. It helps any station you receive.

    Full details

    TempoFast copies far more of what it hears. In simulation on white noise, a frame at +15 dB used to go missing about 7% of the time, more at some time offsets, and strong signals did no better. That is now under 1%. At -10 dB the missed share falls from about 43% to about 9%, the level where half of frames are copied moves from about -11 dB to about -13.5 dB, and false decodes do not rise. The transmitted signal has not changed, so this helps any station you receive, whichever version it runs. Decode time is about the same up to 20 signals, roughly 0.1 to 0.45 s on the bench PC. A very crowded band takes about a fifth longer: 30 signals 80 Hz apart take about 4.7 s, was 3.8 s, which overruns the 4 s slot on either version. All of these are simulation figures; nothing here has been checked on the air yet.

  • Silent CAT no longer cuts your over

    While the radio isn't answering CAT, Nexus now waits for your over to end, turns transmit off once, and says "CAT lost: press TX On".

    Full details

    A silent CAT link no longer cuts your transmission or keeps turning transmit off. While the radio was not answering CAT, Nexus tried its port again every minute and a half, and each try cut anything it was sending through VOX or a serial line and turned transmit off again. Now it waits for your over, tune or PTT to end, turns transmit off once when the link goes silent, and leaves it as you set it after that. Every try still releases the key first. When it turns transmit off, the strip says "CAT lost: press TX On", and the CAT status and Test CAT give the whole sentence, until you press TX On. When CAT comes back it says "CAT is back: press TX On", because transmit stays off until you turn it on. On FT the words show on the strip's TX row. On the other screens they show beside the TX state when there is room. Where there is no room, including Phone and CW at auto-fit, where the meters sit, hover the TX state or open the CAT status to read them. Nothing on the strip moves for these words; where the room is short they end in "…". It also tries less often the longer the link stays silent, up to every five minutes; a radio whose USB port comes back is still picked up at once.

  • "TX off" looks the same everywhere

    FT painted "TX off" green, the receiving colour. It now matches every other screen. (#482)

    Full details

    "TX off" is the same colour on every screen. On FT it was painted green, the receiving colour, while every other screen shows it plainly. (#482)

  • Scopes clear when the radio goes away

    Power the radio down and the Phone and CW scopes now blank and say "Scope data unavailable" instead of freezing the last picture. (#463)

    Full details

    A scope clears when the radio goes away. Power the radio down and the Phone and CW scopes, and the audio view under them, no longer leave the last picture frozen on screen. They now blank and say "Scope data unavailable", and the picture comes back with the first new sweep. (#463)

  • Closing a satellite alarm stops it

    Closing the pass alarm with ✕ now stops the sound, and Open in Satellites stops it and takes you to the pass. (#478)

    Full details

    Closing a satellite alarm stops it. Closing the pass alarm's notice with ✕ hid it but left the alarm sounding for up to a minute. It now stops the alarm, and the notice's button, Open in Satellites, stops it too and takes you to that pass on the Satellites screen. (#478)

  • Stop TX never waits on an error log

    An internal error during Stop TX could delay the unkey by up to two seconds while it was logged. The radio now unkeys at once.

    Full details

    An internal error during Stop TX no longer holds the transmitter on. If something inside Nexus failed while a Stop TX was being carried out, Nexus could wait up to two seconds to write the error to its diagnostic log before the radio unkeyed, and a remote computer's next Stop could wait that long too. It now writes the error without waiting, so the unkey goes ahead at once. The error is still recorded in the log.

  • Settings search finds the Tune timeout

    Searching Settings for "tuner" now finds the Tune timeout. (#487)

    Full details

    Settings search finds the Tune timeout under "tuner". Searching Settings for "tuner" or "tuner timeout" now lands on Settings ▸ Digital ▸ Digital (FT8/FT4), where the Tune timeout is. (#487)

  • ATU press says what happened

    When an ATU press it took comes to nothing, the status line now says why: the radio refused it, did not answer, or the press was dropped before it was sent.

    Full details

    An ATU press that comes to nothing now says why. The ATU button already answered a press it could not take, but once it had taken one, three things could still happen with nothing on screen: the radio refused the tune-up (on a Yaesu, typically because its tuner selection is not the internal tuner), the radio did not answer it, or the press was dropped before it was sent (it went stale, transmit was turned off, stood down or left your privileges in between, a band change, mode switch or radio switch came first, Test CAT was holding the port, or Split Operation could not put the radio's TX VFO back on the dial). The status line now says which: "The radio refused the tune-up (check its tuner is set to the internal tuner)", "No answer from the radio to the tune-up", with "TX stood down in case it is tuning" on FT, or "ATU press dropped" with the reason and "press again". The next press clears it. A press you cancel yourself with Stop TX says nothing. When and whether the radio keys is unchanged, and nothing new is sent to the radio. This was checked against an emulated radio, not yet on the air.

  • LoTW report imports as confirmations

    A LoTW report opened with Import now confirms the contacts it matches on LoTW and adds none. Undo takes the confirmations back off.

    Full details

    Importing a LoTW report adds no contacts and confirms on LoTW. A report downloaded by hand at lotw.arrl.org and opened with Import was read as a log: each confirmation came in as a paper card on the contact it matched, and a confirmation of a contact missing from your log came in as a QSO you never logged. Nexus now knows the file by the LoTW banner it opens with and reads it as Sync confirmations does: the import window says it is a LoTW report and how many confirmations match no contact, each confirmation lands on LoTW for the contact it matches, and none is added. Undo takes the confirmations back off. An ADIF file from any other program imports as before.

  • eQSL InBox imports as confirmations

    An eQSL InBox file opened with Import now confirms the contacts it matches and adds no duplicate or phantom contacts.

    Full details

    Importing your eQSL InBox file adds no contacts. An InBox downloaded by hand at eqsl.cc and opened with Import was read as a log. Each card, timed by the station that sent it, came in as a second copy of your contact carrying that station's "QSL sent" mark, and a card for a contact missing from your log came in as a QSO you never logged. Nexus now knows the file by the program its header names, eQSL's DownloadInBox, and reads it as Sync eQSL does: each card confirms the contact it matches, and nothing is added. An ADIF file from any other program imports as before.

  • Moved logs aren't re-sent to eQSL

    A contact another logger marked as sent to eQSL now counts as uploaded, so saving your eQSL password doesn't send it again.

    Full details

    Moving your log from another program does not send it to eQSL again. A contact that another logger's ADIF marks as sent to eQSL (EQSL_QSL_SENT Y) now counts as uploaded, as LOTW_QSL_SENT already did for LoTW, so saving your eQSL password does not send it again and Awards ▸ Confirmations does not list it as never uploaded. Nexus's own exports now say the same of each contact eQSL accepted or already held, with the day of the upload, so the next program reads it too. A contact eQSL refused, or that was never sent, says nothing.

  • QSL column shows every mark

    The Logbook's QSL column is wide enough for "L·C·E·Q", and the Check lists keep Cancel and their apply button on screen.

    Full details

    The Logbook's QSL column shows every mark. The column was narrower than the mark of a contact confirmed by LoTW, card, eQSL and QRZ together, so "L·C·E·Q" could read "L·C…". It now holds the longest mark. In Check park states, Check confirmations and Check eQSL sent marks, Cancel and the button that applies the list now stay at the bottom of the list on screen, where a long list used to put them below the window until you scrolled to its end.

  • Contest over? restores Band category

    A band category set for one contest is now listed and put back at Contest over?, so the next Cabrillo doesn't claim it.

    Full details

    Contest over? puts back the Band category. A band category declared for one contest (20M for CQ WW SSB, say) was left in place when the contest ended, so the next contest's Cabrillo file claimed it. The checklist now lists the Band category beside the other settings the contest changed, as it is now and as it was before, and puts it back when you tick it.

  • Dismiss the old-contacts merge offer

    The offer to merge contest contacts missing from your logbook can now be dismissed for good. Nothing is deleted.

    Full details

    The offer to merge old contest contacts can be dismissed. Contacts the contest journal keeps that are not in your logbook (because you deleted their logbook copies, for example) were offered every time the Contest screen opened. Dismiss puts that offer away for good on this computer, across a restart. Nothing is deleted: the contacts stay in the journal, and a contact that was not in the offer before brings it back.

  • Legal contact after a banned-mode one counts

    At the Illinois QSO Party, an RTTY contact after an FT8 one with the same station is no longer a dupe, and the same at Winter Field Day after a WSJT contact.

    Full details

    A legal contact after one in a banned mode with the same station now counts. At the Illinois QSO Party, an RTTY contact with a station you had already worked on FT8, on the same band and between the same counties, was marked a dupe and scored nothing, although the sponsor credits it (the FT8 one earns nothing). It now counts on your screen, the club line, every laptop's county board and the claimed score, and is not marked a dupe; the same at Winter Field Day after a WSJT contact. A second RTTY contact with that station is still a dupe. The FT8 and FT4 sequencers' "worked before" is unchanged.

  • Field Day host contacts after Contest over?

    A contact the host laptop logs under your own call after Contest over? now stays in its own log, not the Club Cabrillo.

    Full details

    At ARRL Field Day, the host's own contacts after Contest over? stay out of the club's log. As at every other contest, a contact logged on the host laptop under your own call, with Contest mode turned back on to reach Club Cabrillo, now stays in that laptop's own log instead of going into the Club Cabrillo under the club's call. The GOTA station and the other laptops send their contacts as before.

  • Field Day club log survives a power cut

    The Field Day host now writes each contact to disk before confirming it to the laptop, so a sudden power cut can't lose the newest ones.

    Full details

    A Field Day host's club log survives a sudden power cut. The host wrote each contact the laptops sent into its club journal but left it to the operating system to put on the disk, so a hard power cut could lose the newest ones (the laptops sent them again when they rejoined). The host now puts each one on the disk before it tells the laptop it has it, and the contacts of a laptop's file it imports before it shows what it added. Each laptop's own contest log, the message queue and the JS8 inbox also make sure the file's new name is on the disk after every save, so a power cut can no longer bring back the version from before it (Linux and macOS; Windows has no such step).

  • Field Day ×1 power hint corrected

    The hint now says ×1 is for more than 100 W from any power source, as ARRL's rules give, not only commercial or generator power.

    Full details

    The ×1 Field Day power hint is right. It said "Over 100W — commercial/generator power", but ARRL's rules give ×1 for more than 100 W from any power source. It now says so.

Every version