IMPORTANT: This article and page are under construction. Work on this page is being live posted to provide as real-time assistance as possible as this article evolves. This notice will be removed when the article is deemed to be complete.
Content contained on this page is meant to provide a top-view of station configuration, general guidance that cannot be found within the digital modem software operating documentation, and to provide off-site resources to help you get started in HF digital operations.
The following article intends to provide information on where to download free software to operate on various digital HF modes, and to provide some basic station configuration information.
This article covers multiple software applications, including:
fldigi, which supports a large number of digital modes, including the ever popular, but under-performing PSK31 mode. Also supported are CW, Constestia, DominoEX, FSQ, Heil, IFKP, MFSK, MT63, Olivia, PSK, QPSK, 8PSK, PSKR, RTTY, THOR and Throb. These modes are well suited to having a conversation. Many of these modes do not employ Forward Error Correction (FEC), and not all modes work well into the noise floor. The MT63, Olivia and the PSKR modes do employ FEC, and are more reliable than the non-FEC modes, making them particularly attractive for EMCOMM applications or operation in challenging propagation conditions.
The fldigi application is part of a broader set of applications that are known as the Narrow Band Emergency Messaging Software, or NBEMS. Collectively, the NBEMS application suite provides:
Guaranteed Delivery of Data
Guaranteed Fidelity of Data
Forms Support (such as the ICS-213)
wsjt-x, which supports the JT65, JT9 and FT8 weak signal modes, and is particularly useful for poor propagation conditions or QRP operation. However, these modes are not well suited to having a conversation, and are optimized for making a confirmable contact that can then be used for obtaining an award, such as WAS, DXCC, etc. These modes employ Forward Error Correction (FEC) and work well deep below the noise floor.
Instructions provided by radio manufacturers appear to be structured toward avoiding service calls associated with running high duty-cycle transmissions associated with digital modes, and not providing proper emissions that result not only in avoiding damage to the transmitter, but that avoid distortion and harmonics that would degrade results when operating on digital modes. These principles may not be in alignment with the manufacturer's instructions, but will provide proper emissions, avoid transmitter damage, and avoid audio harmonics and the interference that audio harmonics cause to other stations operating in the small segments of band space that are used for HF digital communications.
Before carrying out the following steps, make sure that the digital modem software (i.e. WSJT-X, JT65-HF, FLDIGI, etc.) is not keying your transmitter.
Set the transmitter POWER control on the radio to maximum.
Set the transmit audio level for digital transmission to MINIMUM. If you’re using a SignaLink USB, this is done be setting the TX knob full anti-clock-wise. If you are using a USB Audio interface that is built-in to the radio, you may have to access a menu item to reduce the transmit audio level to minimum.
Set the meter on the radio to monitor power output.
Using the TUNE button (in either WSJT-X or FLDIGI), initiate transmission.
While transmitting, slowly increase the transmit output level control (using the same control as in step 3 above) until the meter indicates 50% of the rated power output of the radio. If the radio is a 100-watt radio, the level should be set to achieve 50-watts maximum.
Terminate the transmission with the TUNE button.
This procedure achieves the following goals:
Avoid any Automatic Level Control (ALC) activity as this will modulate the transmitter and cause a high Inter-Modulation Distortion (which will reduce the ability of receiving stations to copy your signal, and especially so when your signal is week to them).
Avoid introduction of audio harmonics in the transmission (usually associated with over-driving). This will result in your signal appearing in multiple places on the receive station waterfall (possibly up to appearing everywhere, making the band unusable to others).
Avoid power levels that exceed 50% of the rated power of the radio as digital modes, depending on the mode, are anywhere from 80% to 100% modulation duty cycle. No radio is designed for such high duty cycles at full-power. The 50% level results in avoiding ALC activity, low IMD, and doesn't harm your transmitter.
The instructions provided by the manufacturer often achieve an acceptable power level, which is apparently their only goal to avoid service calls for damaged transmitters, but can still result in emissions that interfere with others or result in such poor communications that you experience difficulty establishing contact with another station. For this reason, it is recommended that you ignore the manufacturer's instructions and employ the above procedure.
Another common issue, which occurs on both Windows and Mac OS X, is that when a sound card device (i.e. SignaLink USB, or a radio with a built-in USB audio interface) is attached, these operating systems make an assumption that you want to use that device for all audio related tasks, including Voice Over Internet Protocol (VOIP, including Skype) and system alert sounds (i.e. email notifications, calendar notifications, Skype instant message notifications, etc.). This can result in your radio making transmissions that are not under your control, and the transmission of music, and you may not even realize that this occurs.
Examples of uncontrolled transmission could include an email notification sound, or a Skype text contact became available sound, or the audio stream associated with playing a video, being transmitted by your station. And since you are probably not listening to your transmitted signal, you probably will not be aware that this is occurring.
These emissions can result in an actionable violation of FCC regulations.
IMPORTANT: To avoid these emissions, you need to set the audio default device back to the device that was used prior to attaching the USB audio sound card device that is used for radio communications (i.e. a SignaLink USB, the radio itself, etc.). This must be done for the Playback / Output device (associated with the transmitter) and the Record / Input device (associated with the receiver). On Windows or Mac OS X, you should use the Sound control panel and set the previous device as the default device. On Windows, take the extra step of setting the previous device as the Default Communications device (this applies to Voice Over Internet Protocol applications, not radio communications). If you use Skype, you should take that added step of setting Skype, from within the Skype application itself, to use audio devices that are not connected to the radio.
Some radios leave the microphone active when operating digital modes. This can result in unintended modulation, and especially so on long transmissions (which are inherent to modes like JT65-HF). It is not unusual to hear arguments with spouses, or business calls, or phone calls with explicit language, when operating on digital modes, all because the operator was not aware that the microphone was hot and did not take steps to avoid unintended modulation.
If your microphone is hot during digital transmission, take steps to avoid unintended modulation. This could include unplugging the microphone during digital transmission, or adding a muting switch to the microphone and then properly managing that switch.
You can test to determine if your microphone is hot by keying the radio for a normal digital transmission, and then monitor the power output while tapping on the microphone. If your radio has a transmit monitor mode, you can also use that feature. If you observe modulation associated with the tapping, as evident by movement of the power output meter or in the transmit monitor audio, your microphone is hot and you need to take steps to suppress its modulation during digital transmission.
NOTE: Most digital communications are carried out in the CW section of the band segment, where voice modulation is not allowed. The only exception on the HF-bands is the 60m band, where voice, digital and CW modes are all authorized. Voice transmissions in the CW section of a band may be actionable by the FCC.
On 60m, Amateur Radio enjoys a secondary allocation, with the primary allocation being for Federal use under NTIA rules. Because of the FCC Amateur Radio allocation being secondary, the NTIA rules are inherited by the FCC regulation for Amateur Radio use, and both mandate that digital transmissions on 60m be centered on the center of the channel. Unlike other bands, the operator does not have discretion as to the insertion point on the waterfall display. Because of this, 60m has a reduced capability with regard to digital traffic capabilities (i.e. traffic is serialized on a single insertion point and there can not be simultaneous digital transmissions within the channel bandwidth that use an insertion point that is not centered). Digital operations on 60m that are not centered on the channel center may be actionable by the FCC. Be aware that this is the reason why all allocation diagrams depicting 60m show both the dial frequency and the center frequency for the channel.
There are many operators in the United States of America that either do not understand this regulatory restriction, or ignore this regulatory restriction. Please do not conduct operations on 60m that may jeopardize this vital secondary allocation and maintain your digital operating practices within the guidance found in the FCC regulations.
NOTE: Because of the secondary allocation on the 60m band, it is good practice to monitor the audio on the 60m channels while operating digitally. This is so that any announcement made by a Federal station, instructing that the frequency be vacated by Amateur Radio operators, can be heard by stations operating on this band. Turning the speaker volume down guarantees that compliance with such a directive would not occur. Non-compliance could be actionable by the FCC, and may jeopardize the current allocation or possible further allocations in this band.
Some digital modes are easily identifiable on the waterfall display, while other modes are both difficult to identify and difficult to tune.
Some modes that are easily identifiable, and easily tuned, include: All of the PSK modes, including PSK31, PSK63, PSK125
Some modes that are difficult to identify and are not easily tuned include: MFSK, Contestia, Throb and DominoEX.
When new to digital operating, all of these modes might appear to be difficult to identify and tune, but experience is a great tool, and some modes will become extremely easy to identify and tune after some experience is gained. Conversely, some modes, regardless of experience level, will always be hard to identify or tune. As a primer, please visit the FLDIGI Users Manual and click on the Modems item in the table of contents to see a fairly comprehensive listing of digital modes (only Pactor, WinLink/WinMor, JT65-HF, and the modes supported by the WSJT-X software, including JT-9 and FT-8, are missing from this list). Then clicking on any specific modem will provide a description, and more importantly, a graphic that shows what that mode looks like on the waterfall display.
To help address both the identification of a mode and the tuned insertion point for communications in the selected mode, a Reed Solomon IDentifier was developed. This is a special digital modulation sequence, outside of the selected digital mode, that indicates both the mode to be used and the tuning insertion point. On face value, this sounds like a great tool, and it can be, but it can also render operations to be impossible if not used properly.
Before explaining the proper use of RSID, lets discuss how RSID is integrated into the digital modem software, using FLDIGI as an example. FLDIG includes the following features in support of RSID:
A separate checkbox control on the main window, RxID, to enable the reception of an RSID.
A separate checkbox control on the main window, TxID, to enable the transmission of an RSID.
Per modem (think digital mode) ability to enable transmission of an RSID.
Per modem (think digital mode) ability to enable receive an RSID.
When RxID (i.e. the reception of RSID) is enabled, any of the following:
Display a dialog window box, requiring dismissal by the operator, which depicts the digital mode being announced by the received RSID, the insertion point tuning, and asking the operator if they want to switch to the announced mode and frequency.
Fully automatic switching to the announced mode and insertion point tuning.
Less obvious is that in some modes, like PSK31, the RSID bandwidth exceeds the bandwidth of the digital mode, and can result in interference to an adjacent (on the waterfall) conversation. Further, RSID takes time to transmit, and reduces the amount of traffic that can be handled in a given period of time, and this should be considered during emergency communications.
Let's now imagine a typical digital mode experience on the 20-meter band, using an operating frequency of 14.070 MHz upper-side-band. There may be more than 20 PSK31 conversations going on at the same time within a 3-kHz receiver band pass, with each conversation having its own asynchronous starting and ending point. Now imagine that all of the PSK31 stations had their TxID (i.e. RSID transmission) enabled. As each of these conversations experienced a new transmission from a participating station, an RSID would be transmitted. For each of these transmissions, all receiving stations would either have to dismiss a dialog window, or retune to their original digital mode and operating frequency after receiving an RSID. If their were 20 conversations, with each conversation turning over twice per minute, this would result in 40 occurrences per minute where a receive station would either need to dismiss a dialog window or reselect digital mode and insertion point. Clearly, this is a very bad use case for RSID.
Let's now imagine a observing a signal on the waterfall that is not easily identifiable. Lacking use of an RSID, the operator may have to manually select various modes and adjust tuning until the mode is properly decoded, and this operation may exceed the time that the transmitting station is active, resulting in a lost opportunity for a contact. This is clearly a case where RSID can be beneficial.
A best use case for RSID would be to have the transmit operator apply the following guidance:
Enable TxID (i.e. the transmission of an RSID) only to those modes that are not easily identified visually on the waterfall display, and/or are not easily tuned. This would exclude using TxID on PSK modes, and probably exclude Olivia and PSKR modes too. Conversely, this would include Contestia, MFSK, THOR, Throb and DominoEX.
Enable TxID (i.e. the transmission of an RSID) only when calling CQ. Disable the TxID on each exchange occurring during a QSO, re-enabling TxID at the termination of the QSO. If in a lengthy conversation, it may be advantageous to leave the TxID on at all times to allow breaking stations to sync up and join in.
This guidance will avoid causing chaos with the use of the RSID feature while still providing for efficient and timely operations on the less popular, and less identifiable digital modes.
A recent article that appeared in QST tended to suggest full time use of RSID on all modes. When the reality of what the results of that operating practice would be were pointed out to the author, the author agreed that full time use of RSID was probably a bad idea, but no correction ever appeared in QST. It wasn't long before digital operators experienced the negative impact of this advice. Immediately following the publication of the QST article, a good number of PSK31 operators were observed using RSID on transmit, and the results were not good. Interference to adjacent conversations occurred, and all stations had to turn off their receipt of RSID if they wanted to avoid constant maintenance tasks associated with the receipt of an RSID, which occurred at the start of each transmission from each of the stations that had enabled RSID. There are still many PSK operators that are implementing this poor operating practice.
The moral of this story is that if you receive guidance, including the guidance in this article, you should apply your own critical thinking, including projecting to what the unintended consequences of such guidance might be, and determine whether the guidance is proper or well thought out. In doing so, please consider not just the benefit to your own operations, but the impact on the operations of others.
Many of the digital modes can accomodate a large number of simultaneous QSOs within the approximately 3.0 kHz receiver bandpass on the frequency of operation. It is not unusual to observe, on the waterfall display, more than 20 simultaneous QSOs occurring with PSK31 or FT8 operating modes within the receiver bandpass.
This ability to support multiple simultaneous QSOs on frequency presents some unique problems when in the receive phase of the QSO. Any of the signals that fall within the receiver bandpass may be strong enough that the receiver's Automatic Gain Control (AGC) may reduce the gain of the receiver in order to avoid receiver overload. If you are working a weak station within the receiver bandpass and the strong signal transmits, the reduction in receiver gain may attenuate the signal of the weak station to the point where copy becomes impossible.
You can use your radio filters to help mitigate the reduction of receiver gain through AGC operation. You can use any combination of the following methods:
Reduce the width of the bandwidth of your bandpass filter to a width that is no smaller than the width of the digital mode you are using, and center the filter over the signal you are working. The goal is to place the strong signals outside of the receiver bandpass so that the strong signals are attenuated, reducing their contribution to AGC operation.
Place the notch over the strong signal. This will reduce the signal strength of the strong signal, reducing its contribution to AGC operation.
Use the IF SHIFT control to move the radio IF so that the strong signal is placed outside of the receiver bandpass. This attenuates the strong signal, reducing its contribution to AGC operation.
The operation of each of these controls is clearly visible on the waterfall display. In fact, using the waterfall display may be the best method for an operator who is unfamiliar with filter operation to gain some insight into how the filters operate and their application in station operation. In general, the attenuated frequencies will appear as a darker background, and signals within the span of the attenuated frequencies will also be rendered as a dimmer image (e.g. a red signal may be drawn as yellow).
You can play with your filters and observe their impact on the waterfall display. This is advisable to gain experience and familiarization of the operation of the filters on your radio.
Although not a filter control, if you are using a rotatable directional antenna (i.e. Yagi, Quad, Beam, etc.), you may be able to rotate the antenna to place a strong signal into a null while maintaining the strength of the desired signal to minimize AGC action from a specific direction.
Lastly, your radio may have the ability to disable the AGC altogether. This should be used as a last resort, but may result in increased receive signal. Unfortunately, all received signals will become stronger and the receiver may overload.
These procedures can make the difference between success or failure in working that far away and rare DX station.
Transmitter Power Levels are usually less than 50-watts, and often in the 25-watt to 35-watt range. This low power level stems from transmitter survivability during a high duty-cycle transmission, and as a courtesy (in part, to minimize the impact of AGC operation with other stations). In fact, the digital modes are particularly well suited for QRP operation, and a great deal of success can be had from portable operation, using a sub-optimal antenna, with just one or two watts of transmit power.
Operators tend to get very upset when their AGC reduces the signal of the station they are attempting to work, to the point of communications failing, when a station transmits with high power.
That being said, one has to question whether they would rather have one station call a station 30, 40, 50 times or more, without success, and inhibiting other stations from working the target station, or would you rather hear the calling station use higher power for a short QSO and return to reduced power operation, giving other stations the opportunity to work the target station? One has to ask as to which is more disruptive to other operators? The subjective answer to these questions is left to the reader.
For instruction on how to configure FLDIG, please click HERE.
The important applications to install for full NBEMS support include fldigi, flarq, flwrap, flmsg and flamp. Although you’re probably going to use fldigi, in most cases, it is a good idea (if you want to have full EMCOM capability available) to download and install the full NBEMS suite.
More information on NBEMS can be found by clicking here.
For the links for NBEMS software below, you want to download the image that is for Mac OS X while running on an Intel compatible processor. These are indicated by seeing _i386.dmg in the file name. Windows users would use the filenames that include _setup.exe.
When configuring fldigi to communicate with your radio, you will need to have a serial connection (requires a USB to Serial converter on the Mac) and the appropriate setup files. For Windows, if the radio has a USB interface, a Virtual COM Port driver may be available and would avoid using the serial interface. An xml file will be needed to configure fldigi to interface to the radio. You can download an xml file for your radio by clicking here, and then selecting the folder for the manufacturer of your radio.
Once you have the xml file downloaded, fldigi can be easily configured to talk to the radio.
Please note that the above version of WSJT-X is a preliminary release (which adds FT8). I would expect that a final release will supersede the above link at some time in the near future, and should be available from the main WSJT-X page. The main page, and on-line documentation, can be found by clicking here.
For instruction on how to configure WSJT-X version 1.8.0, please click HERE.