Up one level
Back to Home
FM Two-Way Radios
to D-STAR Controllers
By Robert W. Meister WA1MIK
While there are many search results and blog postings concerning this topic, few details of the interfacing signals and their interconnection points are provided. This article attempts to summarize the available information.
D-STAR is apparently an acronym for Digital Smart Technologies for Amateur Radio and is presently trademarked by Icom, Inc.
D-STAR was developed by the Japan Amateur Radio League (JARL) and released in 2001. D-STAR is not owned by ICOM, but they were a major contributor to its development along with the Japanese government and other amateur equipment manufacturers. This is why JARL released D-STAR as an open source digital format. D-STAR is not a proprietary format and anyone can build and market D-STAR gear. ICOM adapted the mode. Yaesu has their "System Fusion" products.
Note that DMR signals are different than D-STAR signals. While the interfacing information presented herein applies mainly to D-STAR, it may or may not work with DMR products and things like Digital Voice Modem devices. It is my understanding that DVM uses a higher data rate and many of these older radios just can't support that without digging deeper into the circuitry.
FM vs PM:
Two-way FM radios normally pre-emphasize the transmitted audio and de-emphasize the received audio for several reasons: because the first FM transmitter was actually PM, to keep FM compatible with PM, and to improve the signal-to-noise ratio. The frequency of the carrier can be modulated or shifted either by varying the frequency (true-FM) or by varying the phase (PM) of the signal.
A Frequency-Modulated (FM) transmitter will have pre-emphasis circuitry ahead of the modulator; a Phase-Modulated (PM) transmitter won't, because the modulator circuit inherently pre-emphasizes the signal and this only has to be done once. The early FM transmitters actually were phase-modulated; this process increases the deviation of the audio frequencies above about 400 Hz by a factor of two each time the frequency is doubled. This was an unexpected result (benefit) of the phase modulator circuit but it required de-emphasis at the receiver to counteract the effect. To keep FM compatible with the existing receivers, pre-emphasis is required in true-FM transmitters.
Modulation in an FM transmitter is accomplished by varying the actual carrier frequency, either at the crystal or channel element, or at the VCO (Voltage-Controlled Oscillator). Modulation in a PM transmitter is accomplished by varying the instantaneous phase of the carrier signal after the oscillator has generated it. If you apply DC to an FM transmitter's modulator, the signal will move to, and stay at, a specific frequency, which can be seen on a frequency counter. If you apply DC to a PM transmitter's modulator, the signal will instantaneously change its phase but that's all that happens. Other than a glitch at the receiver, that's all you get. PM needs active (moving) modulation to continuously vary the phase.
The receiver will see a carrier frequency shift proportional to the amplitude of the transmitted audio signal regardless of whether frequency or phase modulation is used. The rate at which this frequency shift occurs depends on the frequency of the transmitted audio signal. Pre-emphasis increases the amplitude of higher audio frequencies. De-emphasis decreases the amplitude of higher audio frequencies. Usually this starts happening around 400 Hz and follows a 6dB per octave gain or loss. Every time the audio frequency doubles, the amplitude doubles, which causes a doubling of the amount of frequency or phase shift of the transmitted signal.
D-STAR doesn't need or want its signals to be pre-emphasized or de-emphasized, so you'll have much better luck with a true FM transmitter. This doesn't mean that a PM transmitter won't work; you'll just have to deal with the differences in the modulation circuitry. This is also why the receive data has to be flat and not de-emphasized.
Other Things to Consider:
D-STAR data must leave the transmitter with the same polarity as it enters the receiver. Some receivers or transmitters will invert the data, i.e. an increase in frequency may result in a lower output voltage, or a higher input voltage may reduce the frequency. You will need to take this into consideration to insure that the input/output polarity is maintained.
Some synthesized radio transmitters produce a higher frequency with a higher modulating voltage; others go the opposite way. Crystal-controlled transmitters can go either way. Receivers tend to vary more because you can have single or double conversion, high or low injection, and IC detectors or true diode discriminators, and these can vary from one band to another. If you have the ability to generate accurate signals, you can measure the detected output as the received frequency is varied up and down an equal amount. You'll soon discover the received data polarity.
Some D-STAR repeater controllers may have a jumper to invert the data in the event a repeater's receiver or transmitter inverts the data. This may be needed in some situations.
The best way to make sure your D-STAR repeater is working is to try it with a pair of radios, either mobile, portable, or a mix. Make sure those two radios can communicate with each other first, then run each one through the repeater and make sure they still work both ways. If not, try inverting either the received or transmitted data but not both.
Constructing a D-STAR repeater with a couple of two-way radios and a simple controller will certainly handle the digital data, but as with any other amateur transmitter, you still have to identify the station once every 10 minutes. You can attach one of several inexpensive CW ID units to the COR, audio input, and PTT lines to accomplish this. We leave that as an exercise for the student.
D-STAR data is frequency-shift keyed with a logic 1 (one) causing a 1.2 kHz shift above the carrier frequency and a logic 0 (zero) causing a 1.2 kHz shift below the carrier frequency. The D-STAR data must be configured this way so the receiving station radio knows when there is a 1 or a 0 in the data. An oscilloscope display on a service monitor will show +/- 1.2 kHz peak deviation but an analog meter may indicate as much as +/- 1.5 kHz deviation due to the nature of the D-STAR data and the use of bandwidth-restricting filters in the metering circuit.
Time For Some Math:
So why are these deviation numbers different? It's because a square wave can be represented as a series of sine waves at odd harmonics of the desired frequency. A 1 kHz sine wave consists of just the first harmonic. If it has an amplitude of 1Vrms, the waveform peaks are actually at a level of +/- 1.414V. The oscilloscope will show these peak levels. With a square wave at the same peak-to-peak amplitude, the RMS voltage would also be 1.414Vrms, but the meter will include all energy present at all odd harmonics. The level of any harmonic can be calculated, but when you square them, add them all up, then take the square root (this is the definition of RMS), you will get the peak value of 1.414V. The amplitude of each odd harmonic (n=1,3,5,...) is:
4 sin (f * n)
-- * -----------
The first harmonic component (1000 Hz) is 4/pi times the original 1Vrms, or 1.273V. The third harmonic component (3000 Hz) is 4/pi times 1/3 times the original 1Vrms, or 0.424V. This keeps going, where each odd harmonic component is lower and lower. On a voice communications radio, the audio bandwidth is usually 300-3000 Hz, so most of these odd harmonics get filtered out and all you see is the 1st harmonic, which is 1.273 times the equivalent sine wave amplitude. Our 1.2 kHz deviated peak-to-peak D-STAR signal produces about 1.5 kHz deviation as a square wave on the analog deviation meter, which most likely also has audio bandwidth filtering.
D-STAR Controller Signals:
The radio will need to provide a flat (not de-emphasized) un-muted receive audio signal at a constant level, i.e. unaltered by the front panel volume control. This can usually be found at the output of the FM detector or discriminator circuit. This point may have some DC voltage present so you'll likely have to isolate it using a 10uF 16V electrolytic capacitor, positive end pointing towards the radio, if your D-STAR repeater controller doesn't already have one. Wideband (5 kHz) or narrowband (2.5 kHz) radios can be used depending on how congested the band is in your area.
You'll also need a spot where you can inject a digital transmit data signal, bypassing all audio processing. Most radios have a microphone (analog) input but you can't feed a digital signal into it because the audio will be pre-emphasized, clipped or limited, and filtered, and the D-STAR signal will be destroyed with all of that. You need to find a spot where the processed audio feeds the modulator circuit, possibly at the deviation pot if the radio has one, or at a spot after the deviation level is controlled. The modulator could be some varactor diodes or even a channel element. If you're using a mobile radio, this might be accessible where the audio board connects to the RF board. If there's DC present on this point, you should isolate it using a 10uF 16V electrolytic capacitor, positive end pointing towards the radio, if your D-STAR repeater controller doesn't already have one. Set the transmitter deviation to +/- 1.2 kHz maximum using an oscilloscope-type display. As the D-STAR data signal is digital in nature, you will get the most accurate deviation measurement by viewing an oscilloscope-type display; an analog meter or digital readout may produce a different deviation value, as it can't follow the square waves in the digital signal. If you have a way to measure it, you might try to get the carrier shift equal to half the bit rate. For D-STAR, this is half of 4800 bits per second or 2400 Hz shift. This is +/- 1.2 kHz from the center frequency.
You'll need a way to make the radio transmit. A PTT input is usually present at the microphone connector. Most radios will transmit if this line is grounded.
A COR/COS signal is apparently NOT needed or required for most D-STAR controllers. They monitor the receive signal continuously and decode the digital data when present. Some even have their own squelch system. The radio usually doesn't even need a squelch circuit, however if you'll be listening to the signals on a local loudspeaker, you may want one. However, you could simply take the receiver discriminator audio and feed it through a level control directly into the transmitter data input, then use the COR/COS signal to key the transmitter, much the same an ordinary analog/voice repeater does.
Make sure you program the radio for carrier squelch (no PL/DPL etc). Similarly you want no signaling (PL/DPL) present on the transmitted signal. D-STAR does not use any sub-audible tone or digital access system such as PL or DPL or their more generic CTCSS or DCS systems.
If you're interfacing a base station, make sure you set it up to be a full-duplex station with the receiver and transmitter completely independent and active all the time. You want to treat it as two separate radios.
The signals detailed above are used by the D-STAR Repeater Controller (DRC) sold by Advanced Repeater Systems. This controller is designed to recognize D-STAR data, condition and delay the data by 250 milliseconds (to give all radios time to stabilize and lock onto the signal), and provide an active-low PTT signal when the controller senses D-STAR data. It does NOT provide any sort of station identification, linking, or call sign verification. Other D-STAR controllers may require additional signals and give you many more features, such as network connectivity and cross-band operation.
The radios mentioned below are all synthesized true-FM unless otherwise noted. The information came from articles on the web and personal examination of the various service manuals.
GE MASTR Exec II: See this article on Repeater-Builder. Even though this is a phase-modulated crystal-controlled station, it still works quite well with D-STAR.
Motorola MaxTrac, Radius, or MaraTrac with 5-pin accessory jack:
Depending on the logic board, you may have to hunt for signals equivalent to those found on the 16-pin logic board. Here are some places to start looking.
PTT: logic board front panel connector J8 pin 11.
TX Audio Input: U651A (MIC preamp) through 4.7uF capacitor and 56k resistor in series.
Ground: chassis or 5-pin connector pin 1 (closest to side of radio).
RX Audio Output: U551A pin 1 (Detected Audio Output).
Motorola MaxTrac, Radius, GM300, etc with 16-pin accessory jack:
PTT: pin 3.
TX Audio Input: pin 5.
Ground: pin 7.
RX Audio Output: pin 11. Set JU551 to position "A" for "Flat/Unmuted Audio" (closest to side of radio).
Motorola MSF5000 to external D-STAR controller:
PTT: SSCB Front-panel XMIT switch or J802 (MRTI) pin 5.
TX Audio Input: J800 pin 34 (TX Data; set appropriate jumper JU4 or JU15).
Ground: J800 pins 6, 7, 8, 9, 10.
RX Audio Output: J800 pin 39 (Quad1 Audio Output).
Motorola MSR2000: See this article on Repeater-Builder. This is a true-FM crystal-controlled station.
All connections are via the 96-pin System Connector on the backplane.
PTT: pin C10.
TX Audio Input: pin A17 (Auxiliary Transmit Audio). You'll probably have to activate this input through RSS.
Ground: pin C19 etc.
RX Audio Output: pin C17 (Discriminator Audio Output).
PTT: TX radio command board: junction of R741 and R742.
TX Audio Input: TX command board: unsolder the negative end of C607 and feed audio into the lifted end.
RX Audio Output: RX radio Command board: U551A pin 4 (flat audio).
All connections are through the radio's 6-pin mini-DIN packet (data) connector.
PTT: pin 3.
TX Audio (Data) Input: pin 1.
Ground: pin 2.
RX Audio (Data) Output: pin 4 or pin 5.
If anyone has any additional interfacing information, please pass it on to the author for inclusion in this article.
The author can be contacted at: his-callsign [ at ] comcast [ dot ] net.
Back to the top of the page
Up one level
Back to Home
This page originally posted on Wednesday 21-Jan-2015.
This web page, this web site, the information presented in and on its pages and in these modifications and conversions is © Copyrighted 1995 and (date of last update) by Kevin Custer W3KKC and multiple originating authors. All Rights Reserved, including that of paper and web publication elsewhere.