A Paradigm Shift in Amateur Radio Equipment

A Paradigm Shift in Amateur Radio Equipment

Otto J. Arnoscht, N4UZZ

Direct digital sampling in software defined radios marks a major step forward in receiver characteristics, like IF DSP two decades ago. We are witnessing the progression of standards in amateur radio. Flex Radio began this move with the development of the Flex 6300, priced aggressively to compete with the Elekraft K3 and Yaesu FT-5000, while offering superior receiver performance.

Now Icom has released the IC-7300, a self-contained direct digital sampling software defined radio with a true live band scope and receiver characteristics that are close to the best in the field at an entry level price of $ 1,049.99 with free shipping(**). No more IF, no more six-pole or eight-pole filters! No more roofing filters! All filtering is now accomplished by software algorithms operating on the digitized RF stream with virtual brickwall characteristics. The results are simply astounding. The price point is amazing. It offers a real live scope with waterfall, and a receiver that rejects signals that are 2 kc away at 94 dB in a package that requires no computer to run.

The following is a listing of receiver data(*), and it is quite revealing.

The Reciprocal Mixing Dynamic Range (RMDR) is currently the best metric for HF radio performance – the higher the number, the better a receiver could still distinguish a small signal while a very strong signal is present very close in frequency. Numbers are based on 2 KHz spacing.

  1. Flexradio FLEX-6700 116dB
  2. Flexradio FLEX-6300 110db
  3. Apache ANAN-100D – 105dB
  4. Elekraft K3S 106
  5. Elecraft K3 104
  6. Elekraft KX3 104, 96
  7. Icom 7850 ( $ 13,000) 101
  8. Flex 5000 – 96
  9. Ten-Tec Orion II – 95
  10. Icom IC-7300 94
  11. Ten-Tec Orion – 93
  12. Kenwood TS-590SG – 92
  13. Ten-Tec Argonaut VI – 92
  14. Yaesu FTdx-5000 91
  15. Ten-Tec Eagle – 90
  16. Flex 3000 – 90
  17. Kenwood TS-590S – 88
  18. Kenwood TS-990S – 87
  19. Drake R-4C/CF600/6 – 84
  20. Icom IC-7100 – 84
  21. Yaesu FTdx-3000 – 82
  22. Icom 703 – 81
  23. Icom 7800 – 80
  24. Ten-Tec Omni VII – 80
  25. Elecraft K2 – 80
  26. Icom 7700 – 78
  27. Icom 7600 – 78
  28. Icom 7410 – 78
  29. Icom 720 – 78
  30. Icom 9100 77
  31. Yaesu FT 450D 76
  32. Icom IC-746PRO 75
  33. Icom 756 Pro III – 75
  34. Yaesu FT-991 – 75
  35. Collins 75-S3B 74
  36. Icom 706 Mk II G – 74
  37. Kenwood TS-480HX – 72
  38. Yaesu FT-991 – 72
  39. Yaesu FT-1000 MP MK Field 71
  40. Icom 756 Pro – 71
  41. Yaesu FTdx 1200 – 70
  42. Signal CX-11A 68
  43. Icom 7200 67
  44. Icom 7000 – 63
  45. Yaesu FT-2000 – 63
  46. Yaesu FT-101E – 59
  47. Kenwood TS-2000 58
  48. Drake R-4C Stock – 58

As of April 2016 2 kHz RMDR Values for current entry-level transceivers below $ 2,000.00:

1. Icom IC-7300 94

2. Kenwood TS-590SG 92

3. Ten-Tec Eagle 90

4. Icom IC-7100 84

5. Icom IC-7410 78

6. Yaesu FT-450D 76

7. Yaesu FT-991 72

8. Kenwood TS-480HX 72

9. Yaesu FTdx-1200 70

10. Icom IC-7200 67

For those of us who were amateurs in the 1970s, we were used to radios like the Drake R-4C stock or the Yaesu FT-101 with RMDR specs in the fifties and sixties. Gradually, receivers moved into the 70s on RMDR, especially with the recent addition of IF DSP. Until a few years ago, QST lab reported only on 20 KHz spacing RMDR values, which now seems quaint and tells you how far we have advanced in receiver technology. With DSP taking over in the last ten years, RMDR specs improved dramatically. Many newer rigs moved up in the RMDR specs to 80 or higher.

In the 1970s some rigs were legendary receivers. One of them was the Signal CX-11A. But it turns out that the RMDR spec is not all that great at only 68db rejection. Now, that was superior to many other radios of the time, but just does not look so impressive today. The Collins 75-S3B was an incredibly expensive radio admired for its “sharp selectivity” yet today, with an RMDR of only 74 db, it ranks right down there with the Icom 706 Mark II G, a radio that is not known for highly selective receive filters.

There are newer radios on this list that are a big disappointment. The FT-991 is a relatively poor performer at only 75 db rejection, and so is the Kenwood TS-480HX with only 72 db of rejection. Other newer entry-level rigs include the Yaesu FT-450D that comes in at 76 (and beats the Icom 756 Pro III) and the Icom 7200 with a poor showing at 67 that still beats out the old Icom IC-7000 that turns in a rather surprisingly low score of 63 db rejection.

Soon no one will be looking at radios that do not have direct digital sampling and real-time band scopes and with RMDR values (or IMD DR values) in the nineties or higher. This standard will start enforcing itself on the bands because radios with high RMDR values cannot hear you even when they are right next to you, but they “walk all over you” and interfere with your reception if you operate a conventional receiver.

Here are some negative points about the IC-7300 that have been discussed on various boards and in various reviews:

  • The IC-7300 shows a full MHz on the band scope, but it does not serve that up to software in the computer. For programs like CW Skimmer, only the band width of 3.6 kHz is served up through the USB cable.
  • The IC-7300 has one band scope – no dual band scopes are available.
  • Pops and impulses that should be handled by the noise blanker will trigger the AGC. This is of concern when operating on 40m or 80m and lightning storms are common along the Gulf coast. Oddly, the Icom 746Pro and 756 Pro III do not have this problem. (Reviewers indicate that nevertheless the noise blanker handles static crashes pretty well.)
  • Some reviewers say that Icoms tend to be somewhat noisier receivers than some other brands, and the IC-7300 is not an exception.
  • Strong signals from out-of-band can cause receiver overload, forcing a reduction in RF gain. Engagement of the IP+ feature essentially automatically reduces RF gain in the face of strong out-of-band signals. This can be of concern for some European operators who are near broadcast stations and may be of concern for Field Day operations.
  • Some operators report that weak signals that are strong enough to copy and work still do not show up on the scope or waterfall.
    • Users and reviewers indicate that SDRs need to be handled differently than conventional radios, and that this applies to Flex radios as well. They say that best weak signal results can be achieved when the RF gain is turned back until background noise becomes barely audible. This “sweet spot” then provides the greatest discrimination between noise and weak signals.
    • There are band scope and waterfall settings that maximize weak signal displays
  • The IC-7300 has a reed relay T/R switch which is noisy during cw QSK. Some users have opened the radio and applied acoustical putty or coax seal putty to the relay to quiet it. Others have noted that the way the relay is mounted allows the relay click to resonate through the board on which the relay is mounted.
  • The CW relay will probably not effectively QSK past 27 wpm between dits. There is still effective QSK at 35 wpm between words.
  • The radio has several birdies below 3.6 MHz.



*The data for this listing were gathered from the Sherwood Engineering Receiver Data list, and from QST lab tests.

**As of July 2018 the IC-7300 is available for an astounding price of $ 1,049.95 after manufacturer’s rebate of $ 100.00.

Antenna Comparisons:

Antenna Comparisons:

End-fed Half-Wave versus Loop

Using FT-8 Log Data

Otto J. Arnoscht, N4UZZ

Three radio amateurs provided their FT-8 log listings for 07/29/18 005430 on 20m for comparison of antenna performance. Station A (N4UZZ) has a 350 foot rectangular loop at 27 feet. Station B (KJ6MTJ) has an end-fed half-wave antenna (EFHW from myantennas.com) mounted north-south at 6 feet above ground on a wooden fence. Station C (KS4TAC) has the same EFHW antenna running from a 6-ft high position near the house sloping upwards into a tree about 35 feet high. About 20 feet of the 130-ft wire extends over the tree and descends downwards on the other side of the treetop. The slope points towards the southwest.

Between the three stations, 27 stations were decoded. An arbitrary signal value of -30db was assigned to any stations that were not decoded. The data from the FT-8 log are provided in tabular form and graphic form on Page 2.

The average signal reports for the three stations were:

Station A: -14.59 (350 ft loop at 27 feet)

Station B: -13.63 (EFHW on 6-ft fence)

Station C: -14.30 (EFHW sloping to 35 ft)

Five stations were not decoded by Station A that were decoded by Station B. Of these five, three were also not decoded by Station C. One dx station in Portugal was decoded by Station A and not Station B or Station C. One dx station in European Russia was decoded by Station B but not by Station A or C. Stations A, B, and C failed to decode 6, 6, and 7 stations respectively that were decoded by one or both of the other stations.

There were a total of 27 received stations. In comparison to Station A with the loop, Station B with the end-fed on a fence received 13 stations stronger; 5 stations about the same; and 9 stations weaker than Station A. In comparison to Station A, Station C received 6 stations stronger; 14 stations about the same; and 7 stations weaker.

The averages reported above were influenced by the arbitrary assignment of a signal report of -30db for stations not decoded. This appeared as a reasonable value to assign since stations with signals below

-24db are unlikely to be decoded. (It should be noted that the db values given here are WSJT-X signal-to-noise ratios and are not related to antenna gain values.)

Overall, the results are that the EFHW antenna strung along a fence at 6 foot height above ground winds up with the best performance on this comparison. It beats out a 350-ft loop at 27-ft height, and beats the other EFHW that is sloping up to a tree about 35 feet up. The overall perspective is that all three antennas performed very close to each other. With FT-8, there is a bit of a random factor in which stations are decoded or not decoded at any specific time period, and in this comparison stations that failed to decode a signal were punished in their scores.

This test underscores the idea that no special antenna structure is needed to work dx and that even a wire strung along a fence can be quite competitive. This should serve as a strong encouragement to amateurs who want to explore hf but do not have the means for expensive antenna installations.

Table 1

WSJT-X signal reports for 27 stations logged

Additional columns are

subtractions of SNR values

for the purpose of comparison

between the antennas.

Below: Chart 1

Graphic presentation of table data. Each three-color cluster of columns represents one signal. The columns present the SNR of the received signals. Blue columns are N4UZZ, orange is KJ6MTJ, and gray is KS4TAC. Numbers correspond to line position in the table.