TWO TONE TESTING
In these amazing days where one can go out and buy a state of the art transceiver for just a little less than the cost of a good second hand car, much is written about such features as price and the number of bells and whistles.
Unfortunately, almost nothing is written about transceiver testing, which is the only way for the amateur to gain a "feel" for the really important front panel features such as audio AGC, audio compression, RF power output, and metering BEFORE a signal is put to air. For beginners this is particularly important, because a huge gap exists between the knowledge required to pass the amateur "ticket", and practical transceiver operation. If you are actually constructing equipment then this knowledge is mandatory.
As the major mode of operation on HF is SSB, two tone testing needs to be clearly understood. With a two tone tester, a dummy load and an oscilloscope, a transmitter can be really put through its paces. Two tone testing consists of applying two clean non-harmonically related sine waves of approximately equal amplitude to the audio input of an SSB transmitter. The sine waves are typically around 600 - 700Hz and 1800 - 2000Hz i.e. about 300Hz from either end of the audio pass band. The result, in a properly adjusted transmitter, is an RF output that varies from zero to maximum at a rate determined by the difference in frequency between the two audio inputs. Consequently, overdrive (which causes splatter), non-linearity, instability and a host of other problems are easily visible on the CRO. This thorough testing of the transmitting system becomes even more important when the transmitter is followed by a linear amplifier- if only to avoid rock throwing, obscene language and other similar behavior by understandably irate neighbours.
Other test modes can be used for SSB but they are not very realistic. For example, a single audio tone can be applied to the microphone input of an SSB transceiver. The oscilloscope display that results is a carrier envelope which does not vary in amplitude. Overdrive and other problems are not easily detected unless a spectrum analyser is available (and only a very fortunate few have one of these). Furthermore, the average RMS output power is equal to the peak envelope power, which is definitely not the case for speech, where the relationship between these two is probably around a 1:5 or 1:6 ratio. At full rated output, power supplies and heatsinks in the typical transceiver generally cannot cope as they are not designed for this type of operation, and transceiver damage is a distinct possibility.
In two tone testing, the average power output is one half of the peak envelope power, and this represents the most severe conditions ever likely to be encountered in amateur operation, which is the continuous sending of a stream of morse code dots (50% duty cycle). And this is a pretty savage test. For phone operation, heat liberated from power supplies and RF power amplifiers is very much less than that caused by two tone testing. Further, the PEP measured during typical speech will usually be significantly larger than that measured under two tone testing conditions (probably 5 - 15% larger) simply because the power supply rails rise due to the light demands and poor supply regulation.
So, in summary, intelligent two tone testing of a transmitting system and application of its results will give the amateur all he needs to operate a transceiver correctly, and put a clean signal on air to the benefit of the entire amateur fraternity.
Single tone testing is also provided by the generator described later in this article, and should generally be reserved for AM, double sideband, and FM. The dummy load and oscilloscope can be used for the first two modulation modes with the same excellent results as for SSB. FM however, requires different test gear to analyse the transmitter output. None of these three modes will be discussed in this article.
HOW IT WORKS
Operation of the generator is quite simple. The two oscillators are based around a low pass phase shift network which provides a 180 deg. phase shift at the frequency of oscillation. The circuit is unique, and is the reverse (dual) of the high pass network normally used in phase shift oscillators. It has the major advantage that the gain around the oscillator loop does not have to be critically adjusted to obtain a sine wave. In fact, the whole circuit operation simply relies on the amplifier having a very high gain to produce a square wave at its output, which is then low pass filtered to generate a sine wave at the amplifier input. Both oscillators are run for two tone operation, while one or the other is stopped for single tone work.
The sine voltages existing at pins 6 and 9 of the op amp are current summed at pin 2, which is a virtual earth due to the negative feedback from pin 1 of the output amplifier.
In order to get a good sine wave, the positive and negative half cycles of the square waves existing at either pin 7 or 9 must be closely symmetrical (better than 2%), and to cause this to happen a slightly larger negative supply is generated by the power supply splitter circuit section. This compensates for the fact that the op amp output swings closer to the positive rail than to the negative rail, and places the op amp inputs exactly at the centre of the output swing.
Note that the current swing into pin 2 can be varied by the 100K pot. This is an important feature sometimes omitted from two tone testers, and allows the amplitude of one audio sine wave to be adjusted relative to the other. This feature compensates for the fact that the transmitter audio response is usually not flat, and allows perfect zero crossings to be obtained in the RF output (see CRO photos).
The last stage is simply an inverting amplifier providing a gain which can be varied over a 50:1 range. The output impedance of this stage is approximately 600 ohms allowing transmitter low impedance microphone inputs to be driven. If the transmitter microphone input is high impedance (47K) then the generator output voltage will be approximately twice that shown on the circuit. Two diodes are included to protect the op amp output, so that it cannot be dragged either above the positive or below the negative rails by an external voltage. Finally high and low voltage outputs are provided.
CONSTRUCTION
Construction is so simple almost no description is required. Make the PCB (if you feel so inclined) and follow the component overlay using the components specified. Make sure all wiring is kept well away from the pin 7/8 end of the IC so that no glitches appear in the output sine waves due to stray capacity coupling of the fast square wave transitions. USE A METAL BOX to avoid RF feedback during testing, and coaxial cable between the generator output and transmitter microphone input for the same reason. The battery is retained by a U loop of heavy copper wire which passes through the two holes provided in the PCB (do not solder these). The copper wire is then soldered to the piece of tin plate shown on the component overlay on top of the battery, through which the battery hold down screw passes. Before switching on, check battery polarity and the orientation of electros and the IC. There isn’t much to go wrong…………..
VK5JST
PARTS LIST
Capacitors
2@ 10uF 16V aluminium electrolytics
2@ 0.1uF 100V MKT caps.
6@ 0.047uF 100V greencaps
Resistors
All resistors are 0.25Watt 5%
2@ 560R
1@ 3K3
2@ 3K9
1@ 5K6
2@ 10K
2@ 12K
1@ 22K
1@ 33K
1@ 39K
1@ 180K
1@ 220K
Potentiometers
1@ 100K linear pot
1@ 500K linear pot
Semiconductors
2@ 1N4148
1@ TL074 or TL084
Miscellaneous
1@ box- DSE Catalogue No. H2305 or equivalent
2@ potentiometer knobs
1@ 9V battery clip
2@ BNC panel mount connectors
1@ SPDT switch (on-none-on) DSE. P7654 or Jaycar ST0335
1@ SP3T switch (on-centre off-on) Jaycar ST0336
1@ 14 pin DIL IC socket
2@ 19 mm long standoff pillars tapped 3 mm both ends
miscellaneous screws, hookup wire, cable clips, scrap tinplate