|
This article has been published in DUBUS 2/2005 and it is protected by copyright. Any reproduction, publishing in the Internet or commercial use only with the written permission from the publisher Verlag Joachim Kraft DUBUS Web page This article has also been published in CQ VHF Fall 2005. Real life dynamic range
|
|
|
AbstractFollowing the recent general advances in receiver design, the receiver part of a typical amateur transceiver now has quite good ability to handle strong unwanted signals - but only if those signals are free from unwanted spurious sidebands (notably keyclicks, splatter and other transients). In contrast, the transmitters have been almost completely neglected. This article gives measured data for several different transceivers from different manufacturers, and it shows that the transmitters are becoming the most important source of inter-station interference. A major contribution to unwanted sidebands comes from ill-designed ALC circuits. The article also discusses what we can do to avoid generating interference to each other, by controlling the output power by other means than the internal ALC.IntroductionInter-station interference can occur when a receiver is trying to listen on a clear frequency, but there is a very strong transmitter using a frequency close by. All transmitters have unwanted sideband emissions (keyclicks, splatter and other transients - the type depends on the transmission mode). If the suppression of these sidebands is worse than the dynamic range of the receiver, then the transmitter will be mostly responsible for the interference experienced by the receiver. Modern amateur receivers have quite good dynamic range, in the order of 90 to 100 dB with the usual definitions (500 Hz bandwidth, and at frequency separations beyond a few kHz). To avoid causing interference to such receivers, the unwanted sidebands from our transmitters must be suppressed to better than -120 dBc/Hz, on the frequency to which the receiver is attempting to listen.Previous articles have dealt with unwanted sidebands due to keyclicks on CW, and splatter on SSB, and have shown that major improvements are needed [1, 2, 3]. The measurements uncovered a significant source of unwanted sidebands that are added by ill-designed ALC circuits. This article explores further, focusing mainly on the effects of poor ALC on the bandwidths of SSB and CW transmissions, and gives actual measured data for several different transceivers from different manufacturers. Speech processing and ALCAll voice modulation methods have an amplitude limit, a level that must not be exceeded. This is valid for FM and SSB as well as for AM. It is of course possible to set the microphone gain low enough to make the largest amplitude peaks that occur when speaking into the microphone so low that they will never exceed the limit. Doing so will provide the best sound quality, but only when the RF signal is strong. The average amplitude from the microphone would be far below the limit nearly all the time, and the transmission channel would be poorly used.As amateurs we are usually not very interested in a high fidelity in the reproduction of our voice in the loudspeaker of our QSO partner. What we want is the best possible intelligibility, at low RF signal levels and also in the presence of interference. To this end we use speech processors, one way or another. All amateurs transmitters contain speech processors of some kind, even though the user may not be aware of it because the circuitry may have a different label. Speech processing usually means amplitude clipping. This is a simple way of making sure that the amplitude does not go above some specified limit. However, clipping will change the frequency content of the signal, and therefore a clipper needs to be followed by a filter to remove signal components outside the intended bandwidth. This is all very well known and written about in the literature. Unfortunately the manufacturers of amateur transceivers do not follow this simple rule about filtering. When they use ALC to control the maximum amplitude of the RF envelope waveform, they change the shape of the waveform but there is no filter that removes the signal components that are generated out-of-channel. Modern ALC systems have high gain and large bandwidths, and they flatten the envelope waveform abruptly which leads to wideband interference. This means that even the emissions from transceivers used at reduced power are far from acceptable. For example, on 144 MHz the FT-817 even has problems to comply with the FCC emission rules paragraph 97.307(e), which generously allow spurious emissions with a mean power up to -43 dB with respect to the mean power of the fundamental emission at a power level of 0.5 W. The problem is common, and by no means limited to FT-817 or to Yaesu as a manufacturer. Also the problem is by no means new, but it has worsened over the years. Fortunately it is often possible to use a properly designed speech processor, limiting the amplitude of the RF drive signal so that the ALC is not activated. Many transmitters already contain a separate speech processor, so in IC7800 and FT1000D, for example, it is possible to set the controls so that the peak envelope power reaches the desired level without activating the ALC at all. The desired level then is only a few tenths of a dB below the level that would have been set by the ALC - but the transmission is a lot cleaner. Other transceivers like the Orion cause a lot of needless interference that could be easily removed by a software update - the computer inside has full control of everything, but the transmitter RF gain does not have a front panel-control. Another possibility is to control the transmitter RF gain by feeding a variable negative voltage into the transceiver's 'External ALC' input, which overrides the internal ALC. Often it is easy to make a modification to use the FM speech processor in SSB mode [2] was published over 20 years ago. Besides generating out-of-channel interference, the ALC systems of modern transceivers make them generate occasional pulses of very high power - far above the maximum rated output of the transceiver. These pulses may cause damage to equipment, may cause protection circuits to operate, or at the very least may cause key-clicks and splatter. The problem is not limited to SSB mode: a wideband pulse is generated in any mode, every time the ALC has to turn the gain down quickly - and that happens each time the power has been low for a time that is longer than the ALC decay time. For some transmitters it happens after each word space in CW, for others even after every space in a string of dots at high speed. There is a particular danger for transverters and other equipment that can only accept a low RF power input: if the power output of the transceiver is reduced by simply turning down the RF PWR control, the transceiver's ALC may still allow pulses of very high RF power. These may either destroy semiconductor devices or cause a gradual worsening in performance. The ALC systemThe ALC system is a simple servo system, and the theory should be well understood by every electronics engineer. The reason for the poor performance is that product testing in amateur journals does not look for the out-of-channel emissions of modulated transmitters, except for the two-tone test which is a static test as far as ALC is concerned. The maximum power level repeats at typically 1 ms intervals in a two-tone test, and therefore even a very short ALC decay time is long enough to keep the ALC voltage nearly constant.The principle of an ALC system is illustrated in Figure 1. The output signal is coupled through a directional coupler to the RF input at the left side of Figure 1. The diode D2 rectifies the RF signal and charges C3 to the peak value of the RF voltage in the negative direction (minus the diode forward voltage). The low-pass link R7-C2 removes remaining RF components and feeds the negative peak RF voltage to the positive input of the op-amp. If the RF power is well below the power limit, which is set by the potentiometer, the op-amp will saturate in the positive direction and the gain control output will have the voltage defined by the voltage divider R2 and R3. |
|
|
Figure 1 Simplified ALC schematic. As the RF power increases, the rectified voltage will go further in the negative direction until the output voltage of the op-amp goes below the gain control voltage by the forward voltage drop of D1. Then D1 will start to conduct and reduce the RF gain so the RF power will no longer increase, but will stay at the level set by the potentiometer. The transmitter gain can be reduced rapidly (fast attack) but with a large value for R6, it will increase slowly (slow decay). It is pretty obvious that an ALC system like this will behave nicely if the response time for gain reduction is much faster than the envelope rise time of the RF signal sent into the transmitter. A less obvious second condition for good behaviour is that the phase shift in the servo loop formed by the RF circuits and the ALC system must remain below 90 degrees at all frequencies where the loop gain is above 1. With a slowly rising signal level, the output power will be proportional to the input power (the power that comes out from the bandwidth defining filter) up to the limit level, where there is a knee after which the input power may be increased but the output power increases very little. Thus the transfer curve is essentially two straight lines (Figure 2) and for practical purposes the upper line can be taken as horizontal. For slowly varying signals, the ALC behaves exactly like an amplitude limiter in the way it affects the envelope shape, and for such signals it therefore generates intermodulation exactly like an amplitude clipper does for signals that change amplitude slowly enough. |
Figure 2 ALC is a form of amplitude limiting |
|
If the attack time constant is made very short in relation
to the rise time of the SSB or CW envelope waveform,
there will be a transient at the knee point where the ALC
system starts to reduce the gain.
Such a transient is a splatter pulse or a keying click,
and its magnitude depends on how much the gain has to be reduced.
By having a very long release time constant one can ensure that
the amount by which the gain needs to be lowered next time
is very small.
This way there will only be one interference pulse at the
onset of each transmission.
|
|
|
Figure 3 First key-down transition of a TS-2000 on 144 MHz.
A rare example of a transceiver in which the ALC does not
cause spectral broadening.
If the attack time is made a little longer, for example by increasing R7, there will be an overshoot at the onset each time the ALC gets active. If gain levels are set so that the ALC only ever needs to reduce the transmitter gain by a small amount, one can select a rather long attack time which will generate a nicely rounded overshoot that does not increase the bandwidth. Such an overshoot is completely harmless if the transmitter is connected to an antenna, but if it is used with a power amplifier it could drive the power amplifier into saturation, with a very strong interference pulse as a consequence. It could also be harmful to the amplifier, or activate protection circuit which takes the amplifier off-line. Figure 3 shows the first key-down transition of a TS-2000 on 144 MHz at full power (100 W). This is one (rare) example of a correctly working ALC circuit. If the loop gain is too high, the amplifiers within the servo loop saturate and the servo system becomes non-linear. Then transients of large bandwidth may be emitted, and also the loop will over-react, bringing the gain down too much. The slow ALC release time constant (R6*C2 in Figure 1) will then slowly allow the power to reach the desired level again. Figure 4 shows the first key-down transition on a TenTec Orion which is an example of this phenomenon. The designer clearly intended to give a nice "raised-cosine" or S-shape to the rising edge, to minimise key-clicks - but the top corner is severely distorted by the ALC transient, and the resulting spectrum in Figure 5 is far from what one would have hoped. |
|
|
Figure 4 First key-down transition of a
TenTec Orion on 14 MHz.
Note the ALC transient at the top corner.
Unfortunately most amateur transceivers are no better designed. The gain in the servo loop is often very high and the phase shift large, with stability problems as a consequence. It is a general rule in the design of servo loops that one should not have more than one RC link to set the gain vs frequency function for the loop. If R7*C2 and R1*C1 are made with similar time constants, the servo loop is close to an oscillator, and more delay through more RC links makes it even worse. It is a good idea to make all RC links except one with time constants that are at least 100 times shorter than the single large one that sets the loop gain roll-off with frequency. Manufacturers of amateur transceivers seem to be unaware of this well-known rule about having a single "dominant pole". For example, Figure #%6 shows the waveform of the first key-down transition of a FT-817 on 14 MHz. The ALC is self-oscillating at a frequency of about 35 kHz, which produces the wideband transients measured in Figure 7. In better transceivers, the front-panel 'RF Power' control is simply a manual gain control which makes a fixed change in the RF drive level. But many transceivers use the ALC system to implement the gain reduction dynamically - which means that the gain reduction that the ALC has to provide is larger at low power levels. This in turn means that the gain of the ALC servo loop becomes higher at reduced power, so that oscillations may occur. Figures 6 and 7 show typical test results for the FT-817. The IC-706MKIIG behaves in a similar way, although oscillation is lower in frequency (about 5 kHz), amplitude and duration compared to the FT-817. The FT-817 is not even stable in the steady state when the key held down, as can be seen in both Figure 6 and Figure 7. |
|
|
Figure 5 First key-down transition of a TenTec Orion on 14 MHz.
This is the peak-hold spectrum in 2.4 kHz bandwidth from the same Orion producing
the waveform of Figure 4. Because of the ALC transient,
the keying clicks are only about 20 dB below those of a transmitter with
totally unfiltered cathode keying!
.
|
|
|
Figure 6 First key-down transition of an FT-817 14 MHz.
Note the self-oscillation in the ALC circuit.
As mentioned above, the use of ALC for regulating the output power may have other side effects. Table 1 shows measured power levels from a IC-706MKIIG at various power settings. The pulse emitted when the PTT button is released in SSB mode may be fatal for a solid state power amplifier [4]. |
|
Figure 7 Spectrum corresponding to Figure 6.
This is typical of FT-817 when run at reduced power on 14 MHz.
The upper curve is peak-hold in a bandwidth of 2.4 kHz and
the lower curve is the average power spectrum at narrow bandwidth.
Setting CW Carrier SSB PTT-off
peak continuous peak peak
(W) (W) (W) (W)
-----------------------------------------------------
H 75 49 108 120
5 55 33.8 78 116
2 16.4 9.5 32.4 97
1 10.0 5.0 11.3 79
L 5.2 2.5 9.0 46
Table 1.
Power output transient levels at the antenna connector
for an IC-706MKIIG.
Effect on inter-station interferenceThere are two possible causes of inter-station interference on the air - receiver overload, or transients from transmitters - and it is sometimes difficult to tell which is responsible. Receiver overload has been extensively reviewed, but transmitted transients have not.State of the art in amateur transmitters is illustrated in Table 2 which shows the results of many measurements of peak hold spectra in a bandwidth of 2.4 kHz. The first entry of Table 2 shows the dynamic range of a typical receiver, an IC-706MKIIG on 144 MHz. If any of the transmitter performance figures in the rest of the table is smaller than the receiver dynamic range figure at the head of each column, it means that the transmitter would be the dominant cause of inter-station interference, rather than receiver overload. Any transmitter performance figures that are equal to or better than this criterion are shown in bold... and you can see there are very few of them! Note the dramatic improvement in the IC-718 and IC-7800 that was achieved by disabling the ALC as discussed above. |
Splatter level below PEP at
Model Power Ser.no. Band 5kHz 10kHz 15kHz 20kHz 30kHz 40kHz 50kHz
or ALC (MHz) (dB) (dB) (dB) (dB) (dB) (dB) (dB)
-----------------------------------------------------------------------------
Typical RX (IC-706MKIIG) 144 56 61 66 70 74 77 79
-----------------------------------------------------------------------------
DX70TH T005735 14 15 15 32 32 51 55 68
DX77 T002056 14 31 50 51 51 51 59 68
FT-1000D 3G3300126 14 39 59 66 66 75 77 79
FT-1000MP MkV 4D570081 14 33 35 35 46 46 59 66
FT-736R 9E260294 144 31 50 55 59 67 74 81
FT-817 0.5W 1E270433 14 15 15 15 15 15 29 29
FT-817 0.5W 1D240059 14 13 13 13 13 13 27 27
FT-817 0.5W 1E270433 144 20 20 20 20 31 31 40
FT-817 0.5W 1D240059 144 15 15 15 15 32 32 32
FT-817 5W 1E270433 14 13 13 13 13 13 26 26
FT-817 5W 1E270433 144 40 49 49 49 61 61 75
FT-817 5W 1D240059 144 36 44 44 44 56 56 67
FT-847 2W 81100231 144 19 19 19 38 38 41 41
FT-847 low 81100231 14 18 18 18 18 35 40 40
FT-847 81100231 14 27 27 27 34 42 50 54
FT-857 3J130041 144 30 50 54 60 69 75 79
FT-857D 4D200054 144 33 52 60 66 71 79 84
FT-897 14 34 52 68 74 80 80 82
IC-706MKIIG 06230 144 28 48 58 62 75 81 84
IC-718 alc ON 03011151 14 41 52 56 58 61 62 63
IC-718 alc OFF 03011151 14 49 58 69 75 84 84 85
IC-765 40W 02576 14 36 52 59 64 64 64 64
IC-765 02576 14 34 37 37 37 37 37 37
IC-7800 alc ON, 0301012 14 38 46 54 61 71 81 88
IC-7800 alc OFF 0301012 14 43 68 88 88 89 89 89
IC-910H 01533 144 32 53 64 68 69 69 69
Orion 03C10433 14 37 41 46 47 53 58 64
TR-9130 3040284 144 33 42 50 56 67 75 82
TS-2000 30400028 144 44 53 63 76 85 88 89
TS-2000 50600050 144 32 48 55 61 71 86 89
TS-2000 25W 50600050 144 35 48 64 79 84 86 89
TS-2000 30400028 14 31 45 57 66 74 75 75
TS-2000 50600050 14 32 53 59 68 77 78 78
TS-2000 25W 50600050 14 32 49 61 69 79 79 79
TS-50 41000988 14 50 67 77 82 85 85 85
TS-711E 8070268 144 17 27 32 42 53 58 67
|
|
Table 2. Peak-hold spectra of some amateur transceivers in
SSB mode. With the exception of the cases shown in bold,
the transmitted signal quality is likely to be the dominant
cause of inter-station interference.
The receiver dynamic range of the IC-706MKIIG is not especially good (for example, the TM-255E is about 20 dB better in dynamic range) so it is not a very demanding standard for comparison. Even so, most transmitters in Table 2 failed to meet that standard, which shows how poor is the typical performance of today's transmitters. There is no good reason for this, because it should be much easier to make a good transmitter than to make a good receiver. The problem, of course, is that they all use the ALC to limit the RF envelope waveform. At narrow frequency separations the linearity of the final amplifier may affect the transmitter bandwidth, but above 15 kHz the interference mainly originates in the transmitter's ALC loop. To use ALC to limit the envelope waveform of a signal that has already gone through a speech processor is ridiculous, as discussed earlier, but very common in amateur equipment. ALC might perhaps have been a clever way of controlling the power level in the vacuum tube era, but this aspect of transmitter design has stood still for 30 years. With appropriate knowledge about what the service menu functions do, or with a software update, most modern rigs can probably be run without the ALC as a speech processor. The computer inside a modern rig should be able to set the gain correctly for the constant amplitude signal that comes out from the SSB filter when a speech processor is used. Speech ProcessingAlthough best readability in SSB mode is obtained without speech processing, the peak power may then go as high as 100 times the average power, and practical transmitters cannot deliver such extreme power levels. In reality there always is some engineering or legal limit on the peak power - and therefore, as pointed out in the introduction to this article, speech processing is necessary for optimum intelligibility in voice communication. (Saving energy when operating from a battery is a very special case. With cleverly managed bias currents, a battery operated SSB station would be best used without speech processing for maximum battery life, but for the rest of this article it is assumed that transmitters are limited by peak power and that the total energy consumption is of no concern.)It is well known from amateur literature that an RF clipper is much better than an audio clipper. This is not quite true, however: the RF clipper is better, but the difference is small as long as the clipping is not harder than necessary for optimum intelligibility. The drawback of audio clipping is often illustrated something like this: "Let us assume that the passband is 0.2 to 2.4 kHz, and that the signal from the microphone is 300 Hz at a given moment. An audio clipper will convert the waveform towards a square-wave that contains odd harmonics. The frequencies 900 Hz, 1.5 kHz and 2.1 kHz will fall within the audio passband and make the sound very different from the original sine-wave. An RF clipper will of course also convert the sine-wave which is present at e.g. 10.7 MHz to a square-wave but the overtones at 32.1 MHz and higher will not pass through the filters, so only the original sine-wave will remain and be transmitted, and so the output from the loudspeaker at the receive side will be exactly the original sine-wave (assuming a correct BFO setting). The only effect of RF clipping to a sine-wave is to reduce the amplitude to make it fit the power limitations of the power amplifier." That line of argument, that a sine-wave will not be distorted by an RF clipper, may sound convincing but it is not really valid. The very purpose of the speech processor is to change - to distort - the voice waveform. Indeed, if it fails to distort the waveform, the speech processor is not doing anything! The relevant question is whether the distortion to a voice signal that an RF clipper introduces is more favourable for intelligibility than the distortion produced by an audio clipper. The human voice is not a sine-wave. If it were, an audio AGC would be the perfect speech processing, fully equivalent to RF clipping. With short pulses sent into the microphone it does not make much difference whether clipping is made at AF or RF. Likewise, if two signals at say 800 and 900 Hz were sent into the microphone input, the third order intermodulation at 700 and 1000 Hz would be the same for RF and audio clipping. One could argue that the human voice is much more like a series of pulses than a sine-wave, and that the difference between AF and RF clipping is small. The only way to really know is to make tests with a real voice signal. I have done such tests some 30 years ago and nobody was able to say whether I was using RF or AF clipping. I did these tests at marginal signal levels only - for strong signals it is easy to hear the difference, but you have to remember that intelligibility is not the same thing as a pleasant 'hi-fi' sound. Recently I have verified these findings with computer simulations. There is a difference, but it is not large. The speech processing simulation is included in Linrad-01.25 (and later versions) as part of the setup for transmit routines. You may download it from [5] and make your own tests to find out how clipping and filtering affects intelligibility with your own voice. The reason for bringing up the merits of audio clipping is that modern transceivers actually use RF clipping without filtering after the clipper by means of a fast ALC system. It has been used for decades, in transceivers like the FT225 for example [2], but it is and it has always been a bad idea, because of all the off-channel interference generated. All rigs designed like this should be modified to make the speech clipping to occur on the correct side of the bandwidth-defining filter. For the FT225 this is particularly easy [2] but it is pretty easy in other rigs also. Basically one can reduce the gain of the amplifier immediately after the filter until the ALC becomes inactive. There will still be something ahead of the filter that limits the signal level and serves as a clipper, but it does not matter if it is an audio amplifier, the SSB generator, an RF amplifier or mixer. Anything that limits the peak power at the right side of the filter - before the input- is fine. It is extremely easy to reduce the gain after the SSB filter. Just send some DC into the ALC input to make the ALC meter permanently show its normal peak voltage. Carrier sideband noiseThe wideband noise surrounding a strong carrier is often the limitation on VHF bands. Such sidebands are usually referred to as phase noise sidebands because it is assumed that they originate in the phase noise of a local oscillator. Transmitters are typically not even as good as the quality of the LO would allow, because inadequate noise figure of the transmit amplifiers plays an important part, and such noise modulates the amplitude as much as the phase.ù As was pointed out in [6], the first article of this series, in order to make sure that no part of the transceiver unnecessarily limits the overall performance, the two-signal receiver dynamic range (in dB for 1 Hz bandwidth) should be equal to the LO phase noise suppression (in dBc/Hz). The transmitter sideband noise should also have the same value, if it is correctly designed and does not add needless noise from poorly designed amplifiers. Remember also that each of these performance figures is valid only at a given frequency offset from the signal frequency. Table 3 shows the two-signal dynamic range and the transmitter sideband noise levels of some typical 144 MHz transceivers, at three different frequency offsets. |
5kHz 20kHz 100kHz
RX TX RX TX RX TX
(dBHz)(-dBc/Hz) (dBHz)(-dBc/Hz) (dBHz)(-dBc/Hz)
-----------------------------------------------------------------------------
IC706(02803) 92.6 91.0 107.7 108.3 125.8 125.4
IC706MKIIG(04668) 106.8 103.8 118.7 117.2 132.3 125.0
IC821H(01942) 97.8 95.8 113.7 113.1 129.0 127.7
IC970H(LA3FV) 102.7 100.1 123.7 121.6 140.7 132.0
FT100(9E032006) 108.6 107.6 118.9 119.0 130.6 129.4
FT817(0N110101) 103.3 101.3 118.2 117.2 132.6 130.4
FT847(LA9CM) 99.9 96.0 116.9 115.0 131.9 130.4
TM255E(51100675) 128.8 116.2 136.9 122.3 144.5 125.5
TS850S+conv(LA6MV) 112.6 113.9 125.6 129.2 138.8 133.8
|
|
Table 3. Comparison of receiver reciprocal mixing and transmitter composite noise. The data is from the Scandinavian VHF/UHF meeting in Gavelstad Norway June 2003. I think it is because this kind of performance information is not made public in transceiver testing that the manufacturers see no reason to design the transmitter carefully. I am sure it would be very easy to cure problems of this kind at the design stage and that it would not lead to higher production costs. At the 2003 Scandinavian VHF/UHF meeting I only made systematic measurements on receivers. At subsequent meetings I have mainly made measurements on transmitters, because transmitter spectral purity is becoming the limiting factor for dynamic range on VHF bands. Table 4 shows the sideband noise of various transceivers collected at several amateur meetings [7]. HF transmitters typically have lower noise at close separation than VHF transmitters. This is a natural consequence of a lower LO frequency. At large frequency separations on the other hand, VHF transmitters are typically better than HF transmitters. Maybe it is because engineers designing at VHF frequencies are more aware of amplifier noise performance, but HF engineers simply assume it will be OK. State of the art: what is good and what needs attentionIt seems to me there is a general consensus that HF receivers have adequate dynamic range [8]. HF receivers are also not often limited by the two-signal dynamic range and therefore some degradation may be harmless. On crowded HF bands the challenge is in the summed power from a large number of strong signals (including broadcast signals on 40 metres). The performance at very close frequency separation may need some attention, e.g. CW operators on the extremely crowded low-frequency bands may need a good intermodulation dynamic range at strong-signal separations as close as 0.5 kHz, and I have been told that many modern rigs fail badly at such close separations.But HF transmitters are very unsatisfactory. The bad habit of using the ALC as a wideband modulator distorts both CW and SSB waveforms, and generates emissions that often degrade the transmitter dynamic range by 40 dB and even more. This is not quite as bad as the number indicates, because the ALC generated sidebands have a high peak to average power ratio and it is possible to hear signals that are much weaker than the splatter peaks or keying clicks - unless, of course, these peaks are strong enough and long enough to capture the AGC of the receiver. On VHF it is different. Neither transmitters nor receivers have the dynamic range required for several operators in the same city area to operate simultaneously without mutual interference. Even at large frequency separations, they are far from achieving this goal. However, there is no good reason why any VHF LO synthesizers should be notably worse than the best ones (for example the synthesizer used in the TM255). It is typically easy to modify the oscillator of any 144 MHz transceiver to this performance level. As an example, LA6LCA has modified the VCO in his TR9130 and the two-signal dynamic range of the modified unit is 147.8 dBHz at a frequency separation of 100 kHz and above. I think the weak-signal VHF community would benefit greatly if the manufacturers would receive this message from the market: "We will only buy transceivers that are have a transmitter and receiver dynamic range better than 140 dBHz at a separation of 100kHz as soon as at least one such unit has become available." Compared to the current state of the art, the improvement needs to be 15 to 30 dB - especially on the transmitter side - and it would make a significant difference to many of us. Finally, there is no reason why such improvements should make the rigs significantly more expensive. The parts that are presently missing are not costly hardware, but some thought and attention from the designers. |
Model Band Noise sideband in -dBc/Hz
(MHz) 5kHz 10kHz 15kHz 20kHz 50kHz
--------------------------------------------------------------------
DSW40(SM4MJR) 7 130.5 130.5 130.5 130.5 130.5
DX70TH(T005735) 14 105.1 113.8 118.0 120.5 127.7
DX77(T002056) 14 101.4 112.1 117.4 120.5 128.9
IC706MKIIG(06034) 14 110.9 118.4 - 124.3 129.3
IC706MKIIG(04668) 14 112.3 118.8 - 122.2 123.7
IC706MKIIG(06230) 144 103.6 112.0 115.9 118.1 125.6
IC718(03011151) 14 111.7 118.6 122.2 124.3 130.6
IC756PROII(01690) 14 117.4 125.4 129.8 131.7 136.3
IC765(02576) 14 121.3 126.7 128.4 129.0 130.1
IC7800(0301012) 14 120.9 131.9 136.1 137.8 142.4
IC910H(01533) 144 96.9 106.2 111.1 113.7 121.3
FT1000D(3F320079) 14 108.1 116.1 - 127.9 130.4
FT1000D(3G330126) 14 107.7 115.0 117.8 120.0 124.7
FT1000MPMV,200W(4D570081) 14 114.8 123.7 126.8 128.4 130.0
FT1000MPMV,20W,AB(4D570081) 14 112.3 114.8 115.0 115.0 115.5
FT1000MPMV,20W,A(4D570081) 14 112.1 114.2 114.4 114.2 114.2
FT726R(3I050222) 144 111.3 123.6 128.2 129.5 130.7
FT736(9E260294) 144 115.7 123.7 126.7 128.4 130.8
FT817(1E270433) 14 107.3 115.2 119.6 122.8 128.8
FT817(1D240059) 144 101.7 110.6 114.8 118.0 126.7
FT817(1E270433) 144 101.0 109.6 114.2 117.4 126.0
FT847(81100231) 14 105.6 117.2 124.9 129.3 136.4
FT847(81100231) 144 94.3 107.3 112.7 116.1 125.2
FT857(3J130041) 144 101.2 111.2 116.1 119.6 126.7
FT857D(4D200054) 144 101.2 111.4 116.5 119.8 126.9
FT897 14 109.9 120.2 125.8 128.4 127.3
K2(03903) 14 114.6 117.8 118.9 119.1 120.0
K2+conv(03903) 144 113.7 119.3 121.9 123.9 128.0
MFJ9020(SM4MJR) 14 127.3 133.1 135.3 136.6 138.7
Orion(03C10433) 14 128.2 127.1 126.2 125.2 119.8
SW30+(SM4EPR) 10 134.6 136.1 136.8 137.0 137.0
TR9130(3040284) 144 116.3 125.6 129.7 132.3 135.9
TS2000(21000340) 14 109.4 117.8 - 123.1 126.2
TS2000(30400028) 14 108.6 117.8 119.6 121.1 124.1
TS2000(50600050) 14 110.3 118.3 121.5 123.0 125.4
TS2000(30400028 144 105.3 115.3 119.8 122.6 131.0
TS2000(50600050) 144 104.7 112.9 117.2 120.6 129.7
TS450S(60700160) 14 110.6 120.0 - 125.6 128.4
TS50(41000988) 14 109.6 114.2 115.2 115.9 116.7
TS711E(8070268) 144 114.0 121.1 124.3 126.0 130.8
|
Table 4. Transmitter sideband noise levels.
References1. Leif Asbrink, SM5BSZ, Transmitter Testing, DUBUS 2/2004 p. 9-45. 2. http://www.sm5bsz.com/txmod/ft225eng.htm 3. http://www.sm5bsz.com/txmod/rt0282eng.htm 4. http://www.sm5bsz.com/dynrange/pulse.htm 5. http://www.sm5bsz.com/linuxdsp/linroot.htm 6. Leif Asbrink, SM5BSZ, Receiver Dynamic Range, DUBUS 4/2003 p. 9-39. 7. http://www.sm5bsz.com/dynrange/rig_compare.htm 8. Peter Chadwick, G3RZP, HF Receiver Dynamic Range: How Much Do We Need? QEX May/June 2002, p36-41. |