Everything posted by Mod_Tom
-
Hamfest season planning: how do you work the flea market, the forums and the club tables?
On the Elmer question, clubs are probably the easiest place to start. Many clubs have a table at hamfests, and most are glad to talk about what they offer: monthly meetings, build nights, license study groups, or help setting up a first station. Ask directly if they have someone who can help with your specific interest, like HF antennas, a first digital rig, or getting on a local repeater. Specific questions get better answers than "can you help me with ham radio?" It also helps to show up at a regular club meeting, not just the big event. Field Day is another good way to meet people, since you get to see an actual station set up and operated, and you'll usually get a turn on the mic or the key. Check the club's schedule for their Field Day plans, since details vary from club to club.
-
One-station logging workflow: N1MM+ for contests, WSJT-X into Log4OM, and keeping ADIF, Cabrillo, LoTW and QRZ in sync
On CAT, the simplest arrangement is to let only one program talk to the radio and have the rest read frequency and mode from it. N1MM+ can do that for digital work through its own interface to WSJT-X, and Log4OM can read from a CAT-sharing layer. A splitter is fine if your radio and driver behave, but two programs polling the same radio at different rates can produce odd frequency jumps. If you see the VFO bounce back to the previous band, a polling conflict is the first thing to rule out. Don't run the WSJT-X log and the N1MM+ log as two masters. WSJT-X writes its own ADIF file, which is a convenient hand-off, but it is not the same as the master log. A tool like JTAlert or Log4OM's WSJT-X integration can forward each QSO as it's logged, which avoids a manual import after the session. Watch the pop-up confirmation. If you accept a QSO in two places, the dupes you create can be hard to find later. For band maps, one useful habit is to let the cluster feed run in only one program at a time. Otherwise you get the same spot twice, and in a contest the spot you work in the other program doesn't mark the station as worked in the first. Check that spots are marked as dupes from the log you're actually using. For uploads, I'd keep LoTW as a manual or scheduled batch from the master log, since a bad station location is better caught before signing. QRZ and Club Log can be more automatic, but send a few test QSOs first and look at the results on the website. Having a deliberate pause between logging and upload also gives you time to fix a busted callsign before it spreads to three services.
-
Shack controllers that glitch or reset when you transmit: ESP32, Pico and CAT builds
On the CAT side, I would treat the interface as its own small module, not wire the rig straight to the microcontroller. For CI-V, the shared bus usually needs a transistor or buffer arrangement that works with 3.3 V logic. For RS-232-level rigs, use a MAX3232 type converter powered from 3.3 V. For TTL-level rigs, check the voltage in the manual and use a level shifter if it is 5 V. Wiring a 5 V signal straight into a Pico or ESP32 pin is a common way to kill a board slowly or all at once. For PTT, an optocoupler or small transistor driver keeps the rig's keying circuit separate from the microcontroller. If you drive a relay, put a flyback diode across the coil. For Direwolf on a Pi, the same applies: a GPIO PTT line should go through a transistor or opto stage, and a USB sound card with CM108 GPIO PTT is often cleaner than driving the radio from the Pi directly. Set TX audio level carefully, because overdriving the mic or data input gives a wide, distorted APRS signal that decodes worse. For Si5351 projects, remember the outputs are square waves, so harmonics need a low-pass filter before an antenna or amplifier, and the frequency needs calibration against the crystal. The breakout board's I2C is 3.3 V, so keep the lines short and the pull-ups correct. If a display on the same I2C bus corrupts when you transmit, shortening that bus, or dropping to 100 kHz, is a good first test before assuming the chips are faulty.
-
Planning a first POTA rove: UTC rollover, two-fers, 10 QSOs per park, and logging without headaches
A few operating tips for getting to 10 contacts efficiently. Pick a band with a decent chance of working your target area for the time of day. 40 m and 20 m are the usual workhorses for HF, with 17 m or 15 m possible when conditions allow. Call CQ POTA clearly, give your park reference at least once per call or two, and put up a spot as soon as you're on the air. Hunters often watch the spots page, so a good spot with correct frequency and mode makes a real difference. Don't forget that duplicates on the same band and mode generally don't count toward your 10, so if the pile thins out, changing band or mode can help you finish. Park-to-park contacts are good too because you're logging a contact that's valuable to both ends. For antennas, a simple resonant wire or vertical works fine. The main points are that it's safe, doesn't obstruct anyone, and is allowed by the park. A longer coax run can lose a surprising amount of QRP signal, so keep the feedline short if you can. As for awards, there are activator and hunter award tiers on the POTA site, and I'd check the current list there instead of relying on memory, since the criteria and endorsements can change.
-
Getting the most out of 5 watts: rigs, antennas and power for portable QRP
For kits, it depends on what you want out of it. The Mountain Topper and TR-35 style radios suit people who want a usable CW rig with minimal bulk, and building one teaches a lot about how a small superhet or direct-conversion design is laid out. Pixie-type builds are a different goal. They are great for learning about mixers, keying and audio stages, but expect wide receive bandwidth and drift, so do not judge CW QRP by that experience. If you are new to CW, I would suggest practicing copy before buying anything. Five watts and a narrow filter will get you contacts, but you still need to hear weak stations through QSB and QRM. Contests with a QRP category, including the QRP-ARCI ones, are a gentle way to build speed because the exchanges are short and predictable. For SSB, plan on a compression-friendly microphone setting and a clear band opening, because there is little headroom to spare.
-
Which spotting and lookup tools do you actually keep open while on the air? (DX cluster, RBN, PSK Reporter, POTA/SOTA and more)
For the field and POTA/SOTA, the biggest practical point is that spotting and hunting are two different jobs. As an activator you want a quick way to self-spot with the correct reference, frequency and mode, and then confirm your spot shows up. As a hunter you want to filter by band and mode and keep the reference and callsign in front of you for logging. Typos in the reference or the frequency cause most of the confusion, so double-check before you post. For CW activations, automated RBN spots can make a signal visible to hunters even before anyone posts manually, but it's still worth checking that the spot is accurate and that you're on the frequency listed. On SOTA, summit references follow an association/region-number format, so copy them exactly rather than retyping them from memory. RepeaterBook is a good starting point for repeater offset, tone and location, but it's user and coordinator maintained, so listings can lag behind reality. If a repeater matters for a trip, check the listing's status and, if you can, the club or coordinator's page. HamStudy and HamCalc are best treated as study and calculation aids: HamStudy for repeated practice of the question pools, and calculators for double-checking numbers like wire lengths and impedance matching, not replacing an understanding of the formulas. And on phone apps: anything that works offline is worth more than anything that doesn't. In a park with no signal, a cached repeater list, a stored band plan and a local logging app beat a slick web tool that won't load.
-
Part 97 compliance Q&A: ID timing, third-party traffic, power limits, RF exposure, remote control, and renewal
I'll take 1 through 3. Station ID (97.119). You must identify with your FCC-issued call sign at the end of each communication and at least every 10 minutes during a communication. The 10-minute clock is a maximum, not a target. The ID has to use an emission that is authorized for the segment you're operating in, such as phone in English or CW. Tactical call signs and net names like "Net Control" or "Shelter 3" are fine for operational clarity, but they don't satisfy the ID requirement. The FCC call sign still has to go out on schedule. Phonetics are encouraged for clarity but not required by the rule. Repeaters are handled a bit differently. The repeater licensee is responsible for the repeater's own identification, which is usually done automatically by the controller. Users still identify their own stations according to the same timing rules. Third-party traffic (97.115). A person who isn't the control operator can participate in communications, but the control operator must be present at the control point and must be able to monitor and stop the communication. Messages still have to comply with everything else in Part 97, which includes the business-communication restrictions. Third-party participation isn't a loophole for anything otherwise prohibited. International traffic is where people get tripped up. Third-party traffic with a foreign station is generally allowed only when the U.S. has a third-party arrangement with that country, or when the third party is someone eligible to be a control operator themselves. Don't assume it carries over from one country to another. Under 97.117, international communications are also limited to messages of a technical nature or remarks of a personal character. Check the current list of arrangements through the FCC or a national organization before offering to relay anything abroad. Power (97.313). The ceiling for most bands is 1500 W PEP at the transmitter output, but the rule also says you must use the minimum power necessary to carry out the desired communication. There are important exceptions. On 30 meters the limit is 200 W PEP. On 60 meters the limit is expressed as effective radiated power relative to a half-wave dipole, which is a different measurement than output power. Technician-class operators have a lower limit on their HF privileges on 10 meters. Some other bands carry geographic or segment-specific restrictions, so read the section for the band you actually use instead of memorizing one number. Prohibited transmissions (97.113). The main categories are communications in which you or your employer have a pecuniary interest, music (with narrow exceptions), messages encoded to obscure their meaning, obscene or indecent language, and broadcasting. There are specific exceptions. A licensee can, for example, occasionally announce the sale of personal equipment on a non-recurring basis, and there is a limited allowance for teaching. These carve-outs are narrow, so read the exact wording of the exception before assuming a particular activity is covered. Also note that false or deceptive signals and willful or malicious interference are prohibited under separate provisions.
-
Planning your first SOTA activation: summit refs, points, VHF vs HF, and what to log
For a first activation, a 2 m FM handheld and a decent antenna is the lowest-effort route if your region has enough VHF activity. It can be as simple as a handheld, a spare battery, and a roll-up J-pole or a small Yagi on a pole. FM is forgiving, and nearby chasers can work you on simplex. It is also easy to get four contacts that way if people know you are coming, so use SOTAwatch alerts. The limit is geometry: a summit with a clear view helps a lot, but VHF still depends on who lives within line-of-sight and whether they monitor the frequency. HF widens the audience, but the kit gets more complex. A common portable setup is a QRP transceiver, a battery, an end-fed half-wave or a linked dipole, and a lightweight pole. Pick bands in advance, such as 40 m and 20 m, and carry a printed note of the antenna lengths or link positions. CW is efficient for low power, but SSB works if you are comfortable with it and the band is open. Test the whole setup in a park first so a loose connector or wrong link position doesn't cost you the summit. Weight discipline matters. Most failures on summits are not exotic, just a dead battery, a broken coax connector, or a pen that does not work in the cold.
-
Making Field Day and SET Count: Building Disaster Comms Drills That Produce Useful After-Action Reports
Good outline. I would add one thing to the after-action section: record timestamps during the exercise, not afterward. Memory smooths over delays. If the net log shows 14 minutes between a message being received at the shelter and it being delivered to the EOC, the report has something concrete to fix. Without times, it turns into "traffic handling was slow," which nobody can act on. On injects, a simple one that exposes a lot is the "wrong format" message. Hand an operator a long, rambling request and see whether they can condense it to a clean message without changing its meaning. It tests judgment as well as typing. Another is a deliberate station dropout in the middle of a net, to see whether the group notices and what happens to the messages that station was holding.
-
Keep, sell, or restore: where do you draw the line between a useful rig and a shelf queen?
My rule of thumb is to separate the "operating" radios from the "appreciating" radios and be upfront about which is which. It stops you from justifying a pricey restoration on something that was never going to be your daily driver. On the vintage safety point, I'd add that recapping and reforming advice gets repeated a lot online and doesn't always get questioned. Slowly bringing up old gear on a variac is a common practice, but the real fix for aged electrolytics is often just replacing them, because a capacitor that has dried out or drifted in value won't recover from gentle voltage. Inspect first, look for bulging or leaking parts and cracked wiring insulation, and don't just power it up and hope. And with anything tube-based, treat the power supply as live until you have proven otherwise with a meter. As for what makes me pass on a classified: vague photos, no mention of whether it has been on the air recently, and "untested" with no explanation. A seller who shows the back panel and the inside and says what has been replaced usually gets the first message from me.
-
Confirming contacts in 2024-style workflow: LoTW first, then bureau, direct, and Club Log OQRS for the rest?
One thing I'd add to the LoTW point: make the log clean before you upload, not after. If the time, band or mode are slightly off, the other station's upload won't match and it looks like they never uploaded. Check that your TQSL station location has the right DXCC entity, grid, and CQ/ITU zones too, especially if you operate portable or from more than one location. Each station location needs its own setup and should be signed with the right certificate. For DXCC, remember that LoTW is only part of the picture. Some people still keep paper cards for older contacts, and those have to go through the ARRL's card-checking process. For anything you plan to submit, make sure the card actually shows what's needed: callsign, date, band or mode, and a legible entity. A very decorative card that hides these details is a poor confirmation.
-
Choosing a handheld: FT-65R vs. Baofeng vs. D-STAR/DMR HTs, plus batteries, antennas, CHIRP and emergency setup
For a first radio I'd keep the analog dual-bander. The FT-65R is a reasonable choice because the menu is simple and it does the basics well, so the learning curve is about operating, not about the radio. Digital makes more sense once you know which modes your local area actually uses. A D-STAR HT is much less useful if your nearest reflectors and repeaters aren't active, and the same goes for DMR if no one around you uses your talkgroups. On CHIRP, the one habit worth building is to export your channel list as a CSV as well as saving the radio image. A CSV is easy to share with a club or an emergency group, and you can sort it by name, band, or use. Name your memories so they make sense at a glance, for example repeater call or location in the first few characters, because the display only shows so much. I'd also test the antenna and battery before an emergency, not during one. Run the radio from a full charge on a typical amount of listening and a few transmits, and see how long it really lasts.
-
From-scratch wire antenna build: what actually matters in wire, insulators, chokes, ladder line and modeling?
On the feedline side, think about what the choke has to accomplish. It should put enough common-mode impedance on the outside of the coax to keep the shield from acting as part of the antenna. A rough target is several hundred ohms at minimum and ideally in the kilohm range on the bands you use. A stack of suitable ferrite cores (the mix matters, and 31 and 43 are the usual HF choices) or a coiled-coax choke can do it, but a choke that is excellent on 40 m can be weak on 10 m. Check the choke impedance against the band instead of assuming it works everywhere. Winding turns of the coax itself through cores works, and so does coiling the cable to form an inductor with some self-resonance. Neither is universally better, because the answer depends on the core, the cable and the bands. For ladder line, the main rules are to keep it clear of metal and gutters, avoid running it parallel to conductors for long distances, and keep the twist or bend radius sensible. Common-mode current on a ladder line system usually shows up as RF in the shack or a tuner setting that changes when you touch things. A 4:1 current balun at the tuner end is an option, but a balanced tuner avoids the balun entirely, and a well-matched balun in the middle of a mismatched system can run hot. I would not reach for a 4:1 and call it done without checking what impedance it actually sees. For measuring, an antenna analyzer or a NanoVNA at the feedpoint tells you more than a SWR meter at the radio. Calibrate at the connector you are using (open, short, load) and keep power low. Feedline loss lowers the SWR you read at the radio, so a good number there does not prove the antenna is good. Minimum SWR is also not exactly the same as resonance. Resonance is where the reactance crosses zero, and the two frequencies can be a little apart, especially on wide-bandwidth designs.
-
EME across 2 m, 70 cm, 23 cm and microwaves: what each band demands in antennas, power, polarization and scheduling
Polarization is where the bands really split. On 2 m, Faraday rotation in the ionosphere can turn the echo's polarization by a large amount, and it changes over minutes to hours. That is why you see slow, deep fades that have nothing to do with the Moon itself. Even with perfectly matched antennas, the two stations can be badly misaligned in polarization. This is sometimes called spatial offset, and it comes from the geometry between the two locations and the Moon. A cross-polarized array with switchable or phased polarization lets you hunt for the best angle, and many 2 m stations use it. On 70 cm the Faraday effect is much weaker, because it falls off roughly with the square of frequency. By 23 cm it is usually negligible, so many dish stations use fixed linear or circular feeds instead. With circular polarization, remember that reflection from the Moon reverses the sense. A station that transmits one sense receives the opposite sense, so check the convention your contacts use before you build the feed. Libration fading is a separate effect. The Moon rocks slightly, which spreads the echo in frequency and produces fast flutter. It is mild on 2 m, clearly present on 23 cm, and severe at 10 GHz. That is why higher-band Q65 submodes use wider tone spacing, and why a signal that looks steady on 2 m can look smeared on the waterfall at 10 GHz. For a first station, this argues for starting on a lower band while you learn to read what the decoder shows you.
-
Gain structure, overload and HF options on budget SDRs: RTL-SDR vs SDRplay vs Airspy vs HackRF, and what the software changes
I'd start with gain, because it is free and the test is quick. Set the software gain low, then raise it until the noise floor just lifts a few dB when the antenna is connected, compared with a terminated input. If the band looks better at lower gain, you are overloading something. A telltale sign is signals that appear in several places and move in odd ways as you change gain. Mirrors, ghost carriers and wide skirts around a strong station all suggest the front end or ADC is being driven too hard. In that case add a notch or band-pass filter ahead of the receiver, such as an FM broadcast notch, before you buy a different radio. On the AGC point, it is convenient for casual listening, but it can hide what is happening. For testing, use manual gain so you can see the effect of each change. For weak-signal work, you want the noise floor from the antenna to just exceed the receiver's own noise. More gain beyond that point doesn't improve sensitivity, it only reduces headroom. Also separate the local noise from the band noise. Move the antenna away from the computer, monitors, LED lighting and switching power supplies. If the noise changes when you unplug a charger or turn off a light, you have your answer. For HF, a common-mode choke on the feedline and a short USB cable, or a USB extension with ferrites, can make a surprising difference. A USB 3 port close to the receiver is a known source of interference in the 2.4 GHz range and sometimes lower.
-
Planning a first POTA activation: park references, the 10-QSO rule, spotting and logging checklist
One thing I'd add for logging: build the park reference into your logger template before you leave. Typing a reference from memory in the field is where typos happen, and one wrong digit puts your contacts under the wrong park. If you're roving, set up a separate log or a clear marker for each park so you don't have to untangle them later. On the UTC point, write down your local time for 0000 UTC before you start. If you're near the cutoff, it can be worth finishing the 10 contacts before the date changes rather than hoping to get them afterward. If you do cross the boundary, a QSO made after the rollover belongs to a new activation day, so be careful when you export. For hunters: if you're chasing a particular park, don't just call blindly. Listen long enough to hear whether the activator is working split, calling CQ POTA on a clear frequency or taking only a certain region or band. Then give your call once, clearly, and wait for the response.
-
Building a 100 W HF shack: how much does the radio matter versus power, tuner, filtering and bonding?
On the tuner and filtering questions, a tuner makes the radio see 50 ohms, but it does nothing about loss in the feedline caused by high SWR. A built-in ATU such as the ones in the G90, IC-7300 or IC-7610 handles modest mismatches. A badly mismatched antenna on coax can waste a lot of power in the line, and a tuner at the shack end will not fix that. If you run a multiband wire fed with ladder line, a tuner suited to balanced feeders is the intended solution, and that is a different design choice from a mismatched coax-fed antenna. With an amplifier, tune up at low drive with the tuner first, then bring up the amp, and make sure the tuner is rated for the power you intend to run. Band-pass filters are mainly for situations where a strong signal on another band reaches the front end, for example multi-transmitter setups, SO2R, or a close neighbour's transmitter. In-band overload (a strong station a few kHz away) is not something a BPF solves. Those cases are handled by the radio's roofing filter and AGC behaviour, and by the attenuator, preamp and bandwidth settings. If you do add BPFs, check the power rating, the insertion loss and whether they sit in the transmit path or only on receive. The superhet-vs-SDR argument tends to sound more decisive than it is on typical antennas. Receiver dynamic range matters most with strong local signals or crowded contest conditions. For a quiet-location home station with a modest antenna, the antenna and noise environment will often limit you before the radio does. A radio-by-radio ranking is probably less useful than asking what your limiting factor is: noise floor, adjacent strong signals, RFI in the shack, or TX power and antenna efficiency. On layout, keep RF feedlines apart from USB, audio and computer cables, and cross them at right angles where you must. Put ferrites on USB and CAT cables. Mind ventilation for the amplifier and supply. Leave room to reach the back of the gear, because most troubleshooting ends up back there.
-
Remote station design: VPN or port-forward, where to put the latency, and what the rules expect from you
One thing worth separating is audio latency from control latency. Many people blame the internet when the real delay is buffering in the software. Most remote audio clients let you set a jitter buffer. A small one gives snappy audio on a clean wired link but chops up on a congested one, and a large one sounds smooth but makes PTT turnaround feel slow. Try it at your worst time of day, not just when the network is quiet. For CW, a station-side keyer is the sensible answer. If you only ever do digital modes, the problem mostly disappears because the radio, decoder and clock are all together. For contesting, I'd read the rules for each contest before building anything, because some limit where the transmitter and receiver can be relative to each other, and some treat a remote station differently from a local one. It is easy to build a great remote station and then discover it is not eligible for the category you wanted. On the SCU-LAN10 and Icom RS-BA1 side, check the firmware and software versions together. These interface units and the radios' firmware have been updated over time, and mismatched versions are a classic cause of audio dropouts that look like network trouble.
-
Breaking pileups on rare DX: split, timing and what changes for Clipperton or Bouvet-type targets
For the low bands, the limiting factor on 160, 80 and sometimes 40 is usually receive, not transmit. Signals from a remote island may be weaker than the local noise, so a receive antenna (Beverage, flag, pennant, small loop or similar) can matter more than extra transmit power. Propagation timing matters too. Grayline and sunrise or sunset enhancement can open short windows, and the signal may rise and fade quickly. Be ready at your own sunrise or sunset rather than assuming the band will be open all night. Check which time windows the target would be on the air at their end as well. Spots and tools help, but they are hints, not facts. DX cluster spots can carry busted calls or wrong frequencies, so confirm by ear before calling and read the comment for split or listening details. For CW and RTTY, the Reverse Beacon Network can show whether anyone near your location is hearing the DX, which tells you when to start listening. Don't spot yourself or post speculative frequencies. Club Log is useful before and after the event. Before, it helps you see how a target ranks for your own needs and which bands and modes you still lack. During and after, many DXpeditions upload logs to Club Log, so the log search can show whether you are in the log, though updates may lag. Avoid working the same band and mode repeatedly just to be sure, because dupes cost the operator time. If the team has an Online QSL Request System (OQRS) set up, use it according to their instructions rather than emailing individually. Finally, on FT8, if the expedition uses Fox/Hound mode, follow the instructions in the software and on the team's page. As far as I know, hounds are expected to call above the Fox's own audio range, and you shouldn't call on the Fox's frequency. Read the team's operating notes, since details differ between events.
-
APRS setup sanity check: paths, Mic-E, SmartBeaconing, and getting a Direwolf iGate onto aprs.fi
My debugging order is to check each hop separately, because 'not on aprs.fi' hides several different failures. First, did the radio actually transmit? Look at the TX indicator and the radio's own beacon log. Second, does the packet look right? A second receiver or a Direwolf log shows the raw frame, including the path and symbol. Third, can an iGate hear you? Look on aprs.fi for nearby iGates and digipeaters, and remember that a handheld with a poor antenna may be heard by a fill-in digi but never reach a wide-area one. Finally, search aprs.fi by exact callsign-SSID, including the raw packet view. A packet that was heard and gated but doesn't appear on the map may have a malformed position, a bad symbol or a filtering issue rather than a radio issue. On paths, I'd keep it simple. In a dense area a single WIDE2-1 or even no digipeating is often enough, since multiple stations will hear you anyway. In a sparse area WIDE1-1,WIDE2-1 is a reasonable default, but I wouldn't go to WIDE2-2 or higher without a local reason. Each extra hop multiplies the traffic, and that adds up quickly on 144.390.
-
How do you separate real ham radio news from hype? Gear launches, band openings, and rule changes
My rule of thumb is to sort every item into one of four buckets: rumor, announcement, shipping, and verified. A rumor is anything without a manufacturer or regulator document behind it. An announcement is a company saying a product is coming, which can still slip. Shipping means people can actually buy it. Verified means someone with test gear has published results. Most hype lives in the first two buckets, and it is fine to be interested without making a purchase decision there. For rules, go to the primary document. The FCC's filing system and the actual rule text will tell you whether something is a petition, a notice of proposed rulemaking, or a final order. Those are very different things, and headlines often blur them. ARRL articles are a good readable summary, but linking back to the source docs is a good habit. For firmware, I'd read the changelog for the specific features you use. If it only touches a menu you never open, there is no rush. If it changes CAT behavior, digital mode audio levels, or the ATU, wait a bit and check owner forums for regressions, and export your settings first.
-
Planning a 24–48 hour EMCOMM field station: how are you splitting power, antenna and noise decisions?
For noise, I'd make it a test step in the setup sequence. Power up everything in the order you expect to run it, then tune across your working bands with the generator on and off and note what changes. Common culprits are switching power supplies, laptop chargers, USB hubs, LED lanterns, and the generator's own inverter. Fixes tend to be mundane. Separate the antenna from power and data cables, use clip-on ferrites or chokes on DC leads and USB cables, and put a common-mode choke on the coax. Also try moving the antenna farther from the shelter. A few extra meters of separation often does more than any ferrite. On coax, use the shortest run that gets the antenna clear of obstructions and the shelter. Loss depends on the cable type, frequency and length, so check the manufacturer's attenuation chart for your cable at 2 m and 70 cm. It is usually the VHF/UHF runs where cheap thin coax hurts most. Add weatherproof connectors, strain relief so the mast doesn't pull on the radio, and a drip loop where the coax enters the shelter. For masts, the safety checks should be unambiguous. Look up for power lines before raising anything, and keep the full mast height, plus a generous margin, clear of any line in every direction it could fall. Guy the mast where the manufacturer specifies, stake it so it can't be dislodged by a vehicle or foot traffic, and put flagging on guys. If thunderstorms are near, the plan should say to stop operating, disconnect and lower the antenna if it's safe, and move people to a proper shelter. A tent is not lightning protection. For the teardown checklist, add: power down radios before disconnecting, stow coax with connector caps on, count stakes and guy lines, photograph the site for a final check, bag trash and cut cable ties, and log battery state of charge and generator fuel so the next team knows where things stand. A final line item I like is a short written note on what failed or was missing, so the checklist improves after every deployment.
-
SDR front-end overload and dynamic range: RTL-SDR vs SDRplay vs Airspy vs HackRF, and how to tame it
One distinction that gets lost in these discussions is a receiver that is sensitive versus one that is robust. A dongle with a good tuner and a decent LNA can hear very weak signals when the band is quiet, and still fall over when a strong broadcast signal is in the same 2 MHz window. That is why a bit-count comparison often misleads. What you really want to know is the blocking and intermodulation performance, and whether the filtering sits before the first active stage. For a basic test that does not need lab gear: watch the display while changing gain in small steps. A real signal rises and falls by roughly the same amount as the gain change. Intermodulation products usually rise faster than real signals, because third-order products grow about 3 dB for every 1 dB of input increase. If a suspect signal moves a lot more than the genuine ones when you touch the gain, it is probably being created inside your receiver. Then try adding attenuation or a band-pass filter and see whether it disappears. On the upconverter point: always double-check the offset sign and value in the software. A wrong offset gives you a perfectly working receiver that shows every frequency incorrectly, and then decoders misbehave and it looks like a software bug.
-
Weak-signal digital station checklist: clock, audio levels, CAT, and what decode thresholds really tell you
Start with time, because it costs nothing to check. FT8 and FT4 decoders assume both ends transmit on the period boundary. The DT column in the decode window shows how far each received signal is from your clock. If most stations show a similar offset, the error is probably yours. If only one station is off, it is probably theirs. A PC that only syncs occasionally can drift by a second or more, and FT4's short periods make that worse. Run an NTP client set to sync frequently, or use a GPS-disciplined source if the machine will be offline in the field. Next is receive audio. WSJT-X's documentation suggests setting the input so the receive level meter reads about 30 dB. Too low and you are working with quantization noise. Too high and strong signals clip and raise the noise floor for everyone else. Use the sound card's native 48 kHz, 16-bit format, and turn off any Windows or macOS audio enhancements. Also make sure system alert sounds do not go to the same device that feeds the transmitter, or one notification will end up on the air. On transmit, set the rig to USB (or the DATA-USB mode if it has one), and turn off the speech processor, compression and any audio equalization. Then set the drive so ALC shows little or no movement. FT8 is close to a continuous carrier for most of each period, so run well under the rig's continuous rating. A clean 20 to 30 watts beats a splattering 100, and a waterfall with spurs beside your signal is the proof that it is overdriven. Keeping the audio tone in the middle of the passband, around 1500 Hz, also helps keep harmonics out of the rig's filtering, and WSJT-X's Split Operation setting (Rig or Fake It) can manage that for you. For the shared-port problem, one serial port can usually only be opened by one program. The usual fixes are rigctld from Hamlib, flrig, or OmniRig on Windows, so the rig is controlled by one process and everything else talks to that. Check that the baud rate, stop bits and handshake in the program match the rig's menu. A mismatch often shows up as intermittent frequency readout rather than a clean failure.
-
Getting from 5 to 20 wpm: what to do about plateaus, head copy and when to put the pencil down
On keyer settings, a few practical defaults to start from. Use a sidetone you find comfortable, usually in the 500 to 700 Hz range, and keep it the same as your receive pitch so you are not retraining your ear. Standard weighting is dash equal to three dits. Spacing matters more than raw speed: a sender at 15 wpm with correct gaps is much easier to copy than one at 20 wpm with mushed letter spacing. For reference, at 20 wpm one dit is about 60 ms, so tiny timing errors are very audible. For iambic A versus B, it mostly comes down to what feels natural. B adds an extra element when you release the paddles after squeezing, which some find helpful and others find annoying. Try both for a few days each rather than switching constantly. Set paddle tension light enough that you are not fighting the spring and the contact gap small but not so tight that it double-triggers. On nets and contests: low-pressure formats are a good middle step. Slow-speed nets and weekly short contest-style events like the CWops CWT sessions have a fixed exchange, so you only need to copy a name and number or callsign. That makes them a controlled way to practice real-world copy. Check the current times, speeds and rules from the organizers, since they change. If you can copy a few exchanges before a slot, do that before trying to jump in and send.