Technical Topics, January 1996
Last Updated on October 12, 2005 by Ralph Brandi
Receiver Test Results From the SWL Winterfest 1995
WINTERFEST 1995 was the scene of an interesting test. Attendees were invited to bring along their radios for a check of sensitivity. Ben Hester brought his Hewlett-Packard programmable signal generator. One of these babies cost more than a new Cadillac when originally sold by HP. Ben keeps a few of these around the house in case a hurricane hits North Carolina. Their weight will help keep the house from blowing away. Chuck Rippel contributed a true RMS meter to accurately measure noise reference levels. Ben and Chuck along with Tony Germanotta and your author spent the better part of a day checking 19 different receivers. The results were presented to the assembled throng the following day.
These tests are significant because, unlike the reviews in other SW publications, these are not new radios just taken out of the box. These are radios that have been knocked around by SWL’s in the real world. They have all survived the hard knocks of life. Some radios, like the Collins R-388 your author drug to Kulpsville, a venerable Zenith Transoceanic purchased by Chuck Rippel from Rosie Blair at the WINTERFEST, and a National HRO-500 owned by Bob Bukovsky were older than the average attendee.
For those of you who want to cut to the chase, all of the radios tested, except one, showed sensitivity better than 3 microvolts for a 10dB signal plus noise to noise ratio. In all cases the receivers were operated in the AM mode and if selectable, the selectivity was adjusted to be close to 4 KHz. That is a reasonable compromise between optimizing sensitivity and AM audio fidelity and is the bandwidth that I like to use for DXing.
The radio sensitivities are grouped into four categories and are shown in Table 1 on the next page:
- Category 1, .1 to .9 microvolts
- Category 2,.9 to 1.9 microvolts
- Category 3, 1.9 to 2.9 microvolts
- Category 4, more than 2.9 microvolts
| .1 to .9 | .9 to 1.9 | 1.9 to 2.9 | > 2.9 |
|---|---|---|---|
| Sony 2010 | Sony 2010 | Sony 2010 | Transoceanic |
| FRG-7000 | Sony 2010 | ||
| Drake R8 | Drake R7 | ||
| Europa 225 | Drake SPR-4 | ||
| Sangean 803A | Panas. RFB-45 | ||
| Drake R8 | Sony 7600G | ||
| Lowe HF-150 | Collins R-388 | ||
| AR-3000 | Grundig 700 | ||
| HRO-500 |
Table 1: How the radios tested at Kulpsville measured up.
A word of caution for new readers. Sensitivity is only one parameter contributing to the overall utility of a short-wave receiver. Don’t place too much emphasis on these results. All of the radios in the two most sensitive categories are easily sensitive enough for DXing on the SW HF frequencies. That is because the atmospheric and cosmic noise coming out of your antenna will be many times greater than the noise generated in any of these receivers.
My Collins R-388, shown in the second sensitivity category, nailed Sierra Leone on 3316 KHz during the November DXpedition to Cape Hatteras, NC. That was the first reported reception of this station anywhere in the world in a long time. This reception proves that success in SW DXing depends on luck, location, antenna, persistence, search strategy, radio selectivity, and radio sensitivity in that order.
If your radio is sensitive enough for antenna noise to be easily heard above the hiss of your radio, additional sensitivity will not improve your radio’s performance.
In fact, additional sensitivity is usually obtained at a sacrifice in dynamic range. Intermodulation, crossmodulation, distortion, and loss of effective sensitivity are the usual result when radios with limited dynamic range are connected to large, outside antennas. Note that the portable radios like the Sangean 803A, and Sony 2010 showed surprisingly good sensitivity. Remember that portables must be sensitive enough to work with a two foot whip. These and similar portable radios are the ones that are most likely to overload on an outside antenna.
We also tried to see what S9 on the meter really means in terms of signal strength. Some radios permit the RF amplifier to be disabled to prevent overload. When possible we measured both settings. As you might suspect, the readings were all over the map. Most generous, with preamp ON, was a Drake R8 belonging to Hans Johnson which registered S9 with only 14 microvolts of input signal. This R8 compared closely to a Drake R7 with the preamps off – 50 and 54 microvolts respectively. The most pessimistic reading was on Chuck Rippel’s Lowe Europa 225 which had to be pushed by 81 microvolts to register S9 with the preamp OFF. This is the same Europa that nosed out the R8 and one of the Sony 2010’s to be judged the most sensitive receiver we tested.
The moral of this story is that you can’t judge the sensitivity of a receiver by the S-meter reading. There is a temptation to believe that the receiver that has the higher S-meter reading on a given signal is the more sensitive. Don’t believe it. The only way to be sure is to measure the sensitivity with the proper equipment.
For those of you who would like to try this at home, here is how the tests were done.
All tests were run on 7100 KHz. We chose this frequency because we had a few receivers which had limited frequency coverage and because this frequency is a good compromise between the tropical bands and the higher SW frequencies. The signal generator was AM modulated at 50% with a 400Hz tone. This signal was injected into the receiver antenna input via a 50 ohm termination.
We then connected a true RMS meter which accurately measures the noise output power at the speaker of each receiver tested. If the receiver had a convenient way to disable the AGC, it was shut off to prevent gain changes from distorting the results. On radios which did not permit AGC to be disabled we ran the test with AGC on. At threshold input levels the AGC is not active according to theory. Observation of the output meter showed that there was little error resulting from this compromise. Receiver selectivity was adjusted to near 4KHz if possible.
We then adjusted the volume control for a reference receiver output noise power level with no signal applied to the 50 ohm terminating interface. The AM-modulated signal generator was then connected to the interface. The generator level was adjusted up until the output meter indicated a level 10dB above the reference. This level represents the level required for the standard 10dB signal plus noise to noise ratio. The output microvolts were read directly from the generator meter.
I would like to thank all who brought their radios including: Bob Bukovsky; Chuck Rippel; Don Cashin; Donna Ring; Hans Johnson; Janice Laws; Joe Hanlon; John Schmid; Ken Ness; Mark Meece; Michael Eilers; Ralph Brandi; Rosie Blair; Tim Maxwell; Tom Bohlander; Tom Swisher; and Tony Germanotta.
Now that Winter is upon us the 1996 SWL WINTERFEST can’t be far away. This year we plan to do something completely different (nudge-nudge, wink-wink, say no more). Chuck Rippel and I will present hints on how to get the most out of your receiver in a program entitled “The Care And Feeding Of Vacuum Tube Radios”. We hope to show why all those knobs on the old vacuum tube radios afford a degree of flexibility which makes them formidable DX inhaling machines even after 40 years of abuse. We will also provide hints to help your old vacuum tube radio to continue running in top condition.
Until next month, stay tuned.