Equipment Reviews, January 2000
RF Systems P-3 Preselector
I jumped at the chance to review this unit when Fred Osterman of Universal Radio gave me the opportunity, having been alerted by a mention on Radio Nederland’s website that the P-3 was on the way. Why all the excitement about another black box? Because the one fault of modern receivers is their lack of front-end selectivity.
Back in the old days before digital frequency synthesizers, most quality radios had a tunable radio frequency amplifier as the first stage after the antenna input. This amplifier was tuned either in conjunction with the main tuning or, in some radios, the stage was tuned with a separate “preselector” control. The main reason that this stage was required was to reduce the problem of “images”-signals that were twice the intermediate frequency away from the desired signal. With the advent of high (>30 MHz) first IFs, image rejection could be accomplished by a simple low pass filter at the front end of the receiver. Along with the switch to digital frequency synthesis and equipment minaturization, mechanical parts such as variable capacitors, wafer switches and slug-tuned inductors became expensive and difficult to obtain. As a result, the tuned front-ends of yore were replaced with fixed bandpass filters covering an octave or more, or no filtering at all.
The ultimate job of a radio receiver is to permit “single-signal” (did National Radio and James Millen have a trademark on this term?) reception of the desired station. This requires that the receiver reject interfering stations and noise, as well as providing adequate sensitivity to allow reception of weak signals. Most hobbyists think of selectivity as that provided by filters in the intermediate frequency stages of the radio. These filters reject adjacent channel interference. There is also a need to reject signals which are far removed from the desired frequency, such as the local broadcaster down the street or WEWN and other powerhouse SW broadcasters. The problems that such strong signals can cause when they enter the circuitry of receivers with poor front-end selectivity are spurious signals (signals appearing on frequencies they shouldn’t), cross-modulation (an undesired signal appears to be mixed with the desired one), intermodulation (two or more signals mix to produce spurious signals), blocking (overall reduction of receiver sensitivity) and a general increase in the background noise level. A properly designed preselector can eliminate these problems.
What’s an antenna tuner?
Many people confuse antenna tuners (also known as transmatches or matchboxes) with preselectors, although the two have completely different functions. An antenna tuner is designed to transform or match the characteristic impedance of the antenna and feedline to the impedance of the receiver’s antenna input. An antenna tuner serves as a transformer to correct for impedance mismatch and allows for maximum signal transfer from the antenna to the receiver. Depending on the circuit design, an antenna tuner may function as a low-pass filter, but lacks the narrow selectivity afforded by a preselector. A preselector serves as a bandpass filter (a “window” which passes only a portion of the radio spectrum) for the radio’s input. A good preselector can compensate for the deficiencies of the wideband front ends of modern receivers (or older receivers with no front end filtering). The marks of a good preselector are a narrow bandwidth (i.e. a narrower window on the frequency spectrum), smooth tuning, minimal insertion loss and no added noise or distortion. High quality components are required for optimum preselector performance. A measure of the quality of tuned circuits is the “Q” factor-essentially the ratio of bandwidth to signal loss-the higher the number, the better.
There is a role for both antenna tuners and preselectors. Since most passive antennas present widely varying impedance over the radio spectrum, an antenna tuner can compensate for this and allow for maximum signal transfer to the receiver. This aspect of signal transfer is most important for transmitters, however. Filters, including preselectors, are designed to have a specific input and output impedance. So, when a preselector is used in a receive-only situation, there is at least a theoretical advantage to using an antenna tuner between the antenna feedline and the preselector to insure optimum preselector filtering. Most active antennas are designed to provide a constant 50 ohm output, so tuners are not necessary with them. Preselectors can be of benefit with active antennas since the antenna’s wideband output can overload receivers, especially if their output gain is set too high.
The RF Systems Approach
To achieve the narrowest bandwidth practically and economically possible, RF Systems used high quality inductors using ferrite core material developed in a research program sponsored by the Dutch government. Using these cores for the coils of the P-3 has resulted in a preselector with very low loss (averaging 1.5 dB in the HF range, 4 dB in the MW band) and no spurious outputs, since there is no amplifier required to overcome insertion loss and the cores are highly resistant to saturation with strong signals.
RF Systems was also meticulous in other design aspects of the P-3. A high quality air spaced variable capacitor is used and is tuned with a 1:3 reduction drive. Tuning is smooth and precise, which is not only nice, but also necessary due to the narrow bandwidth of the preselector. The coils are enclosed in shielded modules which are subsequently filled with inert foam to provide mechanical and thermal stability. The top three ranges (4 to 32 mHz) have an additional high pass filter placed prior to the preselector circuit for additional attenuation of undesired signals. Protection against voltage spikes for both the P-3 and the connected receiver are provided by a static leak path and a gas discharge surge arrestor.
The P-3 has a rated frequency coverage of 30 kHz to 32 mHz. The longwave range of 30 to 450 kHz is not tunable, but is filtered by a low-pass filter which provides 40 dB of attenuation of signals above 450 kHz. The tunable ranges are: 0.45 to 1.6, 1.6 to 4.0, 4.0 to 8.0, 8.0 to 17 and 17 to 32 mHz. The average insertion loss from 1.6 to 32 MHz is 1.5 dB, with a specified maximum of 3 dB. Over the range of 0.1 to 1.6 MHz, the average loss is 4 dB, with a maximum of 7 dB. The specified average 3 dB bandwidths are 15 kHz for the range of 0.45 to 1.6 MHz, 50 kHz for 1.6 to 4.0 MHz, 200 kHz for 4.0 to 8.0 MHz, 700 kHz for 8.0 to 17.0 MHz and 3 MHz for the upper range 17 to 32 MHz. There are no values specified for -20 dB or -30 dB bandwidths, but a frequency plot included in the manual shows good selectivity down to -20 dB. The second order intercept point is specified as +80 dBm and the third order point is listed as +40dBm.
RF Systems has deliberately failed to include a preamplifier circuit in the P-3. Other preselectors add a preamp in order to overcome the insertion loss of the tuned circuitry, such a preamp can suffer the same risks of overload and spurious signal generation as the receiver itself. The RF Systems philosophy was to minimize insertion loss and avoid the need for a preamp and its potential problems.
The P-3 is housed in the same type of enclosure that seems to be standard for RF Systems products, a black enclosure with silver front and rear panels measuring 160 x 150 x 60 mm (WxDxH). SO-239 coaxial connectors are provided for the antenna and receiver connections. There is also a binding post for an earth ground connection which is required for the lightning protection to work. The front panel contains a “Pass-Selective” switch to put the P-3 in and out of the circuit, a six position range switch and the 1.25 inch diameter tuning knob. No power connection is required, as the unit is completely passive. The current price from Universal Radio is $219.95, plus shipping.
The P-3 is simply a joy to use, as the tuning is smooth and precise and the signal peak is sharp and unambiguous. Although in the majority of cases, the preselector will be adjusted to provide the maximum signal level for the desired station, there are times when better reception can be obtained by off-tuning the preselector to reduce the overall level of interference and improve the signal-to-noise ratio. The instruction manual details how to use the preselector in this fashion.
The P-3 is such a well-built and designed unit, that I wish I could report that it magically made weak signals crystal-clear-unfortunately the effects are more subtle than that. With the P-3 in the antenna line, I was able to peak the tuned signal easily and there was almost no noticeable insertion loss for shortwave signals. I could detect a slight reduction in the underlying noise with the P-3 in use and it did help make signals more readable in the lower shortwave ranges-I noticed less effect above 10 MHz. My results may be explained by the overall lower signal strengths here in the West. I tried the P-3 with an AOR 7030+ and a Lowe HF-225E, both of which are designed with front-ends that are resistant to overload. I also used the P-3 with a Kenwood R-1000, an older generation receiver which can be prone to overload, but my results were similar to those with the more modern receivers.
Effects most pronounced on MW
I did notice a definite difference on MW, however. Here using the P-3 reduced the underlying “babble” of those several stations that can be heard under the desired signal. Switching in the P-3 made them virtually inaudible and made the desired signal more intelligible. I was concerned that what I was hearing was just due to the signal attenuation from the inherent insertion loss of the preselector, but comparison of the audio with the preselector switched in versus switching in the receiver’s built-in 20 dB attenuator indicated that the effect was due to enhanced front-end selectivity, rather than signal attenuation alone. I noticed a similar, but less pronounced effect on 49 meter band signals. My last test receiver was a Radio Shack DX-160-a single conversion solid-state dinosaur from the 1970’s. Here the P-3 made a definite improvement in reducing images from the 49 meter powerhouses that tended to appear 910 kHz below the actual frequency.
More information can be obtained from Universal Radio-6830 Americana Pkwy, Reynoldsburg, OH 43068; (800) 431-3939 (orders) and +1 (614) 866-4267 (information). Universal’s catalog webpage on the P-3 can be found at: <http://www.universal-radio.com/catalog/preamps/3790.html>.
The following graph shows selectivity curves for the P-3 tuned to the mid-point of each range:
The top photo of a spectrum analyzer screen shows a sweep of the mediumwave band with the P-3 tuned to 1008 kHz. The bottom photo shows the spectrum analyzer response with the P-3 switched out. Notice that the desired signal has been reduced in amplitude by about 5 dB, but the undesired signals have been reduced by about 40 dB; also look at what happens to the low level noise in the upper trace-it disappears.