Adam, VA7OJ/AB4OJ's IC-7700 User Review

Quick early impressions (June 2008):

Although under great time pressure due to urgent family obligations, I was able to connect the IC-7700 to my Cushcraft R8 vertical, power it up and spend a half-hour or so trying it out. The IC-7700 has the refined overall finish and feel of a first-class test instrument from HP or R&S. The receiver is extremely quiet; with no antenna connected, the internal noise was barely audible and the scope grass was just above the baseline.

I was able to hear a weak 18 MHz SSB station with an excellent SNR, even though the signal was not moving the S-meter. I tried out the roofing-filter selections, NB, NR, auto/manual notch, preamps and Digi-Sel and found them very effective at a quick trial.

The spectrum scope is the best I have ever seen on a transceiver. I tuned in WWV on 15 MHz with the scope at ±2.5 kHz span and slow sweep. I was able to confirm Matt KK5DR's minimum 100 Hz RBW measurement; the two sidebands of the 1 kHz modulating signal stood out as sharply as on a good spectrum analyser. The grass level was right at the baseline. 

Initial observations (September 2008):

On Monday Sept. 1, I installed the IC-7700 in my station, wired it up to the Quadra, set up the ALC and was then QRV! My 7700 was shipped at firmware rev. level 1.03.

I enjoyed a nice round-table QSO with some stations in California, Colorado and Oregon on 40m. Was on 40m SSB for a good couple of hours.The 7700, with its powerful DSP NR and NB, is very comfortable pulling weak, noisy signals out of the mess. It does this job better than the Pro3. Not only does the 7700 sound excellent on transmit and receive; its overall performance and operating "feel" are superb. It makes short work of the band noise on 40m. This shows that the NR and DSP NB in the 7700 are much superior to the NR and analogue NB of the Pro3.

By means of comparative 7700/Pro3 listening tests on 40m at night, I have found that the DSP NB and NR in the 7700 do a noticeably better job of quieting down the electrical power-line noise. In one instance, I was able to eliminate intractable electrical line noise on 40m by increasing NB Width to 85, and setting the front-panel NB control at 3 o'clock. This NB setting reduced the idle-channel S-meter reading from S4 to S1 (Preamp off). With NR at 60%, the noise level at the speaker fell sharply.

Another NB test: On the IC-7700 with NR at 1 o'clock, NB at 3 o'clock, NB Depth and Width at 8 and 80 respectively and Preamp 2 on, power-line noise at approx. S9  was completely suppressed. Activating NB dropped the S-meter reading from S9  + 5 dB to S6 (the band noise level). I could just hear a weak ZL at 7144 kHz. With NB off, the power-line noise swamped him completely. By comparison, the Pro3's analogue NB suppressed the noise completely with Preamp off (but I could not hear the ZL). With Preamp 1 or 2 on, the NB no longer took out the noise. In fact, the noise was unaffected on the Pro3 with Preamp 2 on and NB at 75% (and of course it overwhelmed the ZL.)

The NB impulse response test below tends to confirm the IC-7700's excellent NB performance in on-air operation.

NB Demo (July 2009): These sound clips were captured on 20m SSB, in the presence of power-line noise at S9 + 10 dB:

I find that I can pull weak SSB signals out of the noise more easily on the 7700 than on the Pro3. The 7700's much more powerful DSP also yields greatly superior IF filter performance - steeper skirts and better shape factors. The 7700's Manual Notch, with 3 width settings (WIDE-MID-NAR), can be used to "sharpen up" the flank of the passband. This can  enhance a weak signal in some situations.

On Sunday morning I was listening to some stations on 20m SSB. The Manual Notch (WIDE) completely suppressed the cable-TV spur near the Icom Net frequency (14316 kHz); the superiority to the Pro3 Manual Notch was quite evident. The band noise was much lower - on the spectrum scope and in the speaker. With Preamp off, the scope "grass" level was at baseline with 5 kHz span selected, and S1 - S2 signals "leapt out" and were easy to copy. Again, this is a big improvement over the Pro3.

When switched in, Preamps 1 and 2 raise the audible noise level in the receiver (and the scope "grass" level) significantly less than in the Pro3. It is evident that in the 7700, the preamps have been designed for highly linear large-signal handling, consistent with a low noise figure.

I have not yet encountered a situation in which the Digi-Sel preselector had a noticeable effect, but noted that the front-panel Digi-Sel control detunes the preselector sufficiently to cause observable signal loss at the ends of its range. I have selected the 6 kHz roofing filter as the default for SSB.

The PA standing current on my IC-7700 is 2.8 - 3A on start-up, rising slightly as the radio warms up. I am not concerned, but this does seem a bit on the high side as compared to my Quadra amplifier which uses 8 MRF150's with Vdd = 48V. Standing Id on the Quadra is approx. 2A. This suggests that the operating point of the IC-7700 PA is a little closer to Class A, for lower IMD.

Note on audio transducers:

I use my Heil GoldLine GM-5 and Icom SP-20 with the IC-7700. Transmit audio settings are as given in thetable, with 6 dB compression on voice peaks. The settings used with the Pro3 also work fine with the 7700. The mic is plugged directly into the front-panel MIC socket; to my way of thinking, the IC-7700's transmit audio management features (which are all in the digital domain) obviate the need for external analogue audio "boxes" with their potential for added noise/distortion and unpleasant EMC surprises.

Physical considerations:

Here is a view of my new shack setup.I left the handles on my 7700, and there they will stay. They enable me to manoeuvre the radio more easily on the table. The main tuning knob with its separate rubber grip is a bit of a pain; a knob with the ring already in place, and a hole in the grip allowing access to the set-screw would have made more sense.   

I disliked the plastic front-foot extenders. They seem cheap and nasty, and pop off when the radio is moved as they tend to stick to the table surface. They can also mar the table or scratch the lower case cover of the radio. Elevating the front of the radio by fitting the extenders places the rear edge of the case perilously close to the table surface; this risks scratching the case. At the very least, the extenders should have been shorter, and secured with screws.After a day or so I removed them, and my 7700 is now sitting horizontally on the table

Notes on Quadra integration, ALC and the RX OUT/IN sockets:

The interface with the Quadra amplifier is seamless, and works 100%.The ALC setting on the Quadra for 1 kW output is 5 bars - the same as for the Pro3. At this setting, the IC-7700 delivers 65W drive to the Quadra. The DRIVE control is set for a 50% reading on the radio's ALC scale.

As is the case for all other Icom HF transceivers, the IC-7700's rear-panel ALC jack is compatible with a negative-going ALC voltage in the range 0 to -4V. The input resistance at the ALC jack is 18 kΩ. Positive voltage should never be applied to the ALC jack.

The IC-7700 has rear-panel RX OUT and RX IN BNC sockets. These allow the insertion of an external preselector, multicoupler or other accessory into the receiver's input signal path. As these sockets can be disabled via menu, no U-link need be connected between them when they are not in use.

Display quality:

The IC-7700's TFT display is simply breathtaking. The screen  image is expansively large, bright and pin-sharp, the contrast and colour saturation are excellent and the colour adjustment menus offer many possibilities. The standard meter display emulates a moving-coil meter to perfection, right down to the ballistics and pointer shadow! I shall comment more on the display as the review develops. The overall feel of the radio calls to mind a first-class test instrument.

Note on spectrum scope:

Unlike the Pro3 scope which has a fixed 1 kHz resolution bandwidth (RBW), the IC-7700 scope offers variable RBW in the range 100 Hz to 3 kHz. The optimum RBW value is automatically selected for each span and sweep speed setting. In CENTER mode, at ± 2.5 kHz span and SLOW sweep speed, 100 Hz RBW is selected. This allows a clear display of  IMD products and AM sidebands, for example. At the narrower RBW settings, the "grass" level in the absence of signals is at or near baseline. Weak-signal spikes are thus clearly visible.

The IC-7700 spectrum scope also provides a very useful spectral-content display of the transmitted signal at the ± 2.5 kHz span and SLOW sweep settings. One can use this display as an aid when setting up the transmit audio. (Scope during TX ON in Scope Set menu.)

Note on PSK31 operation:

I heard a strong PSK31 signal on 7071 kHz and tuned it in easily in a few seconds with the IC-7700's vector diagram and AFC. It was a local amateur of my acquaintance, so I plugged a USB keyboard and was instantly QRV. We chatted back and forth for about a half-hour at 15W RF output.

PSK31 is simplicity itself with the IC-7700's PSK facility. The vector diagram, FFT scope and waterfall display to the right of the encode/decode text field greatly facilitate correct tuning. I plan to work some weak PSK31 stations, and will report the results here.

Note on RTTY operation:

Although I am a complete newcomer to amateur HF RTTY operation, I had no trouble tuning in and decoding RTTY signals during a recent contest. The familiar tuning-bar display, together with the FFT scope and waterfall display to the right of the encode/decode text field, make correct tuning very easy. The squelch can be set to mute the audio in the absence of a received signal; this is especially useful when using the Twin Peak Filter (TPF), which raises the idle-channel audio level.

As for PSK, I plan to work some weak RTTY stations, and will report back here.

A few receiver measurements (September 2008)

As the ARRL, RSGB and others have already subjected the Pro3 to exhaustive test suites, I felt that any such effort on my part would be redundant (apart from the fact that my second signal generator is much too "dirty" for meaningful RF 2-tone measurements!)

To satisfy my own curiosity, though, I ran a few quick receiver and spectrum scope tests along the same lines as in myPro3 User Review, as follows:

  1. MDS

  2. Reciprocal Mixing Noise

  3. Two-Tone IMD3 (IFSS, Interference-Free Signal Strength)

  4. DSP IF Filter Shape Factors

  5. SHARP vs. SOFT SSB Filter Roll-off

  6. NR Noise Reduction (measured as SINAD)

  7. Manual Notch Stopband Attenuation

  8. Manual Notch Bandwidth

  9. AGC Impulse Response (pulsed RF and pulse trains)

  10. NB Impulse Response (pulse trains)

  11. Spectrum Scope Sensitivity and Amplitude Accuracy

  12. Spectrum Scope Resolution Bandwidth

  13. S-Meter Tracking and AGC Threshold

  14. DIGI-SEL Insertion Gain, Bandwidth & Passband Spectrograms

  15. REF OUT Phase Noise

  16. Calibration Marker Leakage

  17. Transverter Output Level

The following test setup was used in tests 1, 2,4 - 6 and 11 - 14:

HP8640B signal generator, calibrated against Millivac MV723B RF power meter with 50Ω 0 dB probe, then connected to ANT4 (configured as RX).Sinadder 3 connected to [SPKR] jack*. NB off, NR off, ATT = 0 dB, AGC MID, Twin PBT cleared. Offset tone pitch = 1 kHz (SSB), 600 Hz (CW). SSB and CW "Sharp" filter shapes selected. 6 kHz roofing filter selected. (Note: BPF Indicator is on for CW filters.)

* Sinadder 3 set to "SINAD" for Test 5, and to "AC Volts" for all other tests.

For Test 3, the following test setup was used: Marconi 2018A and 2019 signal generators, each driving a Mini-Circuits ZHL-32A amplifier. Each amplifier output was connected via a 15 MHz LPF and a 20 dB pad to one input of a Mini-Circuits ZSC-2-1B combiner. The combiner input was routed via a 0-110 dB step attenuator to the IC-7700 ANT1 input.

1. MDS (Minimum Discernible Signal). In this test, MDS is defined as the RF input power which yields 3 dB unweighted (S+N)/N at the output. The results obtained are very close to those in Ref.2. Table 1a shows the effect of the 6 and 3 kHz roofing filters on MDS; these filters have higher insertion loss than the 15 kHz filter.

Table 1: MDS (Min. Discernible Signal)
MDS at 14.1 MHz (dBm)
Preamp SSB 2.4 kHz CW 500 Hz BPF
Off -120 -126
1 -132 -138.5
2 -136.5 -142

Table 1a: Roofing Filter & MDS

15 -3 -3
6 0 0
3 0 0

2. Reciprocal Mixing Noise. Reciprocal mixing occurs in a superheterodyne receiver when the noise sidebands of the local oscillator (LO) mix with strong signals close in frequency to the wanted signal, producing unwanted noise products at the intermediate frequency and degrading the receiver sensitivity. Reciprocal mixing noise is a measure of LO spectral purity.

In this test, a strong "undesired" signal is injected into the receiver's RF input at a fixed offset from the operating frequency. The RF input power is increased until the receiver noise floor increases by 3 dB. The reciprocal mixing noise parameter, expressed as a figure of merit, is the difference between this RF input power and measured MDS (factoring in ΔMDS, per Table 1a.) The test is run with the 2.4 kHz SSB filter selected, and preamp off. The higher the value, the better.

Table 2: Reciprocal Mixing Noise
  Recip. Mixing (dB)
Offset kHz 3.6 MHz LSB 14.1 MHz USB
Roof Filter 15 6 3 15 6 3
2 74.5 73 75 78 75 78
3 79 78 80 83 81 82
5 84 85 87 85 86 89
10 87 96.5 96 94 96 96

3. Two-Tone MD3 (IFSS, Interference-Free Signal Strength). tested in CW mode (500 Hz), ATT = 0 dB. Test frequencies: f1 = 14100 kHz, f2 = 14102 kHz. IMD3 products: 14098/14104 kHz. IMD3 product level was measured as absolute power in a 500 Hz detection bandwidth at various test-signal power levels, with 12 dB ATT selected. The ITU-R P-372.1 band noise levels for typical urban and rural environments are shown as datum lines.


Fig.1: IC-7700 2-tone IMD3 (IFSS) vs. test signal level.
Fig.1: IC-7700 2-tone IMD3 (IFSS) vs. test signal level.

Note on 2-tone IMD3 test: This is a new data presentation format in which the amplitude relationship of the actual IMD3 products to typical band-noise levels is shown, rather than the more traditional DR3 (3rd-order IMD dynamic range) or SFDR (spurious-free dynamic range). The reason for this is that in a direct-sampling SDR receiver employing an ADC, SFDR referred to input power rises with increasing input level, reaching a well-defined peak (“sweet spot”) and then falling off. In a conventional receiver, SFDR falls with increasing input power.

If the IMD3 products fall below the band-noise level at the operating site, they will generally not interfere with desired signals.

The SFDR behaviour of an ADC invalidates the traditional DR3 test for a direct-sampling SDR receiver. Our goal here is to find an approach to SFDR testing which holds equally for SDR and legacy receiver architecture. See Reference 5

This test has been run on the IC-7700 with 12 dB RF attenuation inserted. The reason for this is that it allows direct comparison of IFSS for a legacy receiver and a direct-sampling SDR  by bringing the conventional receiver's noise figure into line with that of the SDR.

4. DSP-IF filter shape factors. A quick check of -6/-60 dB shape factors was performed for the 2.4 kHz SSB, 500 Hz CW and 250 Hz CW  IF filter settings. (Note: BPF Indicator was on for CW filters.) Shape factors for all 3 filter settings met or exceeded Icom's advertised specifications. The measured -6/-60 dB shape factor for the 2.4 kHz SSB filter was 1.24; the spec is 1.5.

Table 3: Shape Factors
Filter Sharp Soft
2.4 kHz SSB 1.2 1.25
500 Hz CW 1.3 1.4
250 Hz CW 1.4 2.5

Tip: To achieve optimum S/N ratio, the selected IF-filter bandwidth should not exceed the occupied bandwidth of the received signal.

5. SHARP vs. SOFT SSB filter roll-off.  Test conditions: 10.000 MHz LSB, 2.4 kHz filter selected, RF input power -96 dBm (below AGC threshold), AGC off, preamp off, ATT = 0, NR off, NB off, Twin PBT neutral. Table 3 shows the roll-off for the [SHARP] and [SOFT] filter settings.

Table 3: SSB Filter Roll-off
Offset Hz SHARP dBr SOFT dBr
250 -4 -10
300 0 -5
400 0 -5
500 0 -4
750 0 -1.5
1000 0 0
2000 -1.5 -2
2500 -2.5 -3

6. NR noise reduction, measured as SINAD. This test was intended to measure noise reduction on SSB signals close to the noise level.

Initial test conditions: Sinadder configured for SINAD measurement. 10.000 MHz LSB, 2.4 kHz filter selected,  AGC MID, preamp off, ATT = 0 dB, NR off, NB off, Twin PBT neutral.

The receive frequency was offset +1 kHz to produce the test tone, and RF input power was adjusted for a 6 dB SINAD reading (-127 dBm). NR was turned on, and SINAD read at 30% and 50% (max.) NR settings. The results are given in Table 4.

Table 4: NR vs. SINAD
NR setting % SINAD dB
0 6
30 10
55 (max.) 16

This shows an S/N improvement of 10 dB with NR at maximum for an SSB signal roughly 6 dB above noise level. This is an approximate measurement, as the amount of noise reduction is dependent on the original signal-to-noise ratio.

7. Manual Notch Stopband Attenuation. Test signal: -72 dBm (≈ S9) at 10.100 MHz CW. NR off, NB off, 500 Hz CW filter selected, Preamp 2 on. Manual Notch set to MN/NAR. Measured MDS: -142 dBm.

The Manual Notch nulled out the signal completely. Thus, Manual Notch stopband attenuation is at least 70 dB.

8. Manual Notch Bandwidth. Dave, KI4KQ, has been kind enough to contribute the results of his Manual Notch bandwidth test.

9. AGC Impulse Response. The purpose of this test is to determine the IC-7700's AGC response in the presence of fast-rising impulsive RF events. Two types of event are applied to the receiver input;  RF bursts with a fast-rising wavefront, and pulse trains with short rise times. (Icom's DSP-derived AGC is discussed here.)

IC-7700 configuration: 14.100 MHz USB for Test 8a (2.000 MHz for Test 8b), 2.4 kHz SSB filter (Sharp),  NR off, NB off, AGC VR on, AGC control at minimum (or AGC FAST, with decay time set at 0.1S.)

9a. Test with RF bursts. An HP 8011A pulse generator applies a pulse train to the AM input of the HP 8640B, with the AM selector set to [PULSE]. The test is performed with Preamp off at two steady-state RF power levels: -20 dBm (S9 + 50 dB) and -10 dBm (S9 + 60 dB) at 14.100 MHz. The pulse generator is adjusted to generate RF bursts of 1.2 µS duration. Burst rise time (to -3 dBr) is 200 nS. Pulse period is 200 mS. (The RF burst train is displayed on a Tektronix 455 oscilloscope.)

The AGC recovers completely during the preset decay time; there is no evidence of clamping. This is evident from a sound file captured during the test. (The 1 kHz tone is due to signal leakage through the HP 8640B's pulse gate.) View a screenshot of the test.

9b. Test with pulse trains. Here, the HP 8011A is coupled to ANT4 via the pick-off port of a line sampler. The sampler's main port is terminated in 50Ω. The IC-7700 is tuned to 2 MHz, as the RF spectral distribution of the test pulse train has a strong peak in that band. AGC decay time is set to minimum (0.1S) as before, but Preamp 2 is on.

The pulse rise time (to 70% of peak amplitude) is 10 nS. Three pulse durations are used: 50, 70 and 125 nS. In all cases, pulse period is 600 mS. Pulse amplitude is 16Vpk (e.m.f.)

Table 5: Pulse Train Test Summary
Pulse Duration AGC Response S-Meter Peak
50 nS barely detectable no indication
70 nS recovers in < 0.1S S3 - S4
125 nS recovers in < 0.1S S7

As in Test 8a,  the AGC recovers completely during the preset decay time; there is no evidence of clamping. This is evident from these sound files. View a screenshot of the test.

10. NB Impulse Response. As the IC-7700's noise blanker is a DSP process "upstream" of the AGC derivation point, the NB should be very effective in suppressing impulsive RF events before they can stimulate the AGC. To verify this, the NB is turned on during Test 8A. The NB parameters (Threshold, Depth and Width) are adjusted for best suppression of the test pulses. (Ref. 3, p. 5-16). Table 6 gives the results. A sound file is available for the 125 nS case, the pulse is barely audible. The S-meter deflection is also completely suppressed in both cases, showing that the impulsive events never reach the AGC derivation point.

Table 6: NB Impulse Response Test Summary
Pulse Duration NB Settings Result
  Depth Width Threshold Blanked?
70 nS 6 75 60% Totally
125 nS 8 100 80%

The NB also suppresses isolated single events. This can be verified by selecting PULSE PERIOD/EXT and pressing SINGLE PULSE on the HP 8011A.

11. Spectrum Scope Sensitivity (minimum visible spike). In this test, the RF input signal level is adjusted to produce a spike which is just visible above the scope "grass" level. SPAN is set to ± 2.5 kHz, with SLOW sweep in CENT mode. As the scope is independent of the main receiver, the IF filter setting is irrelevant. ATT = 0 dB, Scope ATT = 0 dB. Table 7 gives the results:

Table 7: Spectrum Scope Sensitivity
Minimum Visible Spike for SPAN = ± 2.5 kHz
Preamp Level dBm
Off -114
1 -126
2 -133

11a. Spectrum Scope Amplitude Linearity.The spectrum scope dynamic range is 80 dB (Scope ATT = 0 dB). The scope graticule has 8 vertical divisions at 10 dB/div.

In this test, SPAN is also set to ± 2.5 kHz, with SLOW sweep. The vertical amplitude is noted for each 10 dB increase above the minimum visible spike. The scope display tracks the input signal power  accurately over the entire 80 dB range.

12. Spectrum Scope Resolution Bandwidth. In a spectrum analyser, the resolution bandwidth (RBW) determines how far apart in frequency two (or more) signals must be to be resolved into separate and distinct displays on the screen.

To measure RBW on the IC-7700 spectrum scope, the Calibration Marker is activated, and a test signal is injected into the antenna input at a level sufficient to produce a spike whose vertical amplitude is equal to that of the Marker. Initially, the test signal is approx. 10 kHz above the selected Marker spike. (Example: Marker at 14100 kHz; test signal at 14110 kHz.). To ensure an accurate amplitude display, sweep speed is set to SLOW for all SPAN settings. For each SPAN value, the test signal is moved closer to the Marker spike until two distinct spikes are just observable. Table 8 gives the measured RBW value for each span setting:

Table 8: Scope RBW
Span ± kHz RBW Hz
2.5 100
5 200
10 500
25 1k
50 1k
100 2.5k
250 3k

Note: From the foregoing, it will be seen that the IC-7700 is also useful as a bench spectrum analyser.

13. S-Meter Tracking and AGC Threshold. This is a quick check of S-meter signal level tracking. Initial IC-7700 settings: 2.4 kHz USB, Preamp off, 15 kHz roofing filter, AGC SLOW.

A 14.100 MHz test signal at MDS is applied to the antenna input. The signal power is increased, and the level corresponding to each S-meter reading is noted. (S9 readings are taken with Preamp off, Preamp 1 and Preamp 2 in turn.)

To measure AGC threshold, the HP 8640B is offset -1 kHz to produce a test tone in the speaker. The RF output is turned down to MDS.  The IC-7700 AF Gain control is adjusted for -6 dBr test tone level. The input signal power is then increased until the test tone no longer increases. The test is then repeated with AGC OFF. The actual AGC threshold (knee) is the point at which the AGC OFF test tone level first exceeds that for AGC SLOW.

The test results are given in Table 9:

Table 9: S-Meter Tracking & AGC Threshold.
S S1 S2 S3 S4 S5 S6 S7 S8 S9 S9+10 S9+20 S9+30 S9+40 S9+50 S9+60
dBm -91 -89 -87 -85 -83 -80 -78 -75 -72 -62 -52 -42 -32 -22 -11
Preamp 1: S9 = -83 dBm. Preamp 2: S9 = -90 dBm.
Measured AGC Threshold (Preamp OFF): -90 dBm.


14. DIGI-SEL Insertion Gain & Bandwidth. This is a simple test to determine the gain of the DIGI-SEL preselector at its centre frequency, and its  -6 dB bandwidth. Initial IC-7700 settings: 2.4 kHz USB, Preamp off, DIGI-SEL off, DIGI-SEL control at 12 o'clock, AGC SLOW. Spectrum scope SPAN is set to ± 5 kHz, with SLOW sweep in CENT mode.

A 14.100 MHz test signal at approx. -70 dBm is applied to the antenna input and adjusted for +40 dB vertical amplitude on the spectrum scope. DIGI-SEL is then activated and the signal-generator output reduced to restore the vertical amplitude to + 40 dB.

Here, the required reduction is 2 dB, indicating 2 dB DIGI-SEL insertion gain.

Next, SPAN is increased to ± 250 kHz, with MID or FAST sweep, and the input power readjusted for +40 dB on the scope. The test signal frequency is slowly increased until the scope amplitude decreases by 6 dB*.   The test signal frequency is then reset to 14.100 MHz and slowly decreased until the scope amplitude again decreases by 6 dB. The total frequency excursion between these two points is the -6 dB bandwidth of the DIGI-SEL (500 kHz in our test.)

*This is checked by increasing the signal-generator output by 6 dB and noting that the scope again reads +40 dB.

14a. DIGI-SEL passband spectrograms. A NoiseCom NC6110 RF noise generator applies a broadband noise spectrum at a power spectral density of -91 dBm/Hz to the ANT 1 input. Initial IC-7700 settings: fo = 14.100 MHz, 2.4 kHz SSB, Preamp off, DIGI-SEL off, DIGI-SEL control at 12 o'clock, AGC SLOW. Spectrum scope SPAN is set to ±250 kHz, with FAST sweep in CENT mode. Fig.2 shows the resulting noise spectrum.

Fig.2: Noise spectrum with Digi-Sel off.
Fig.2: Noise spectrum with Digi-Sel off.

Next,  Digi-Sel is activated. Fig.3shows the resulting noise spectrum. Note the 6 dB roll-off at the -250 and +250 kHz span edges.

Fig.3: Noise spectrum with Digi-Sel on.
Fig.3: Noise spectrum with Digi-Sel on.

15. REF OUT Phase Noise (November 2009). At the request of IØWTD, I measured the phase noise of the IC-7700's 10 MHz reference oscillator output  on an HP 8563E spectrum analyser with the HP 85671A phase-noise utility. The analyser was connected to the REF I/O socket, withREF IN/OUT (ACC menu) set toOUT. Click here for the phase-noise plot. (Note that the displayed phase noise is close to the limit of measurement.)

16. Calibration Marker Leakage (September 2010). The calibration marker comb signal is applied at the inputs of the LPF's preceding the HF and 50 MHz 1st mixers.

I measured the marker signal leakage from the ANT1 antenna socket with a Millivac MV-723B broadband RF millivoltmeter terminated in 50Ω. The leakage is negligible with either preamp (or the Digi-Sel) in-line.  The meter indicated the composite signal power of all the 100 kHz harmonic spikes passing through the BPF for the selected frequency range. The test results are given in Table 10. (Note: Test frequencies were just outside the amateur bands, to prevent damage to the meter's sensor in the event of accidental keying.)

Test conditions: Calibration Marker on, Preamp off, Digi-Sel off.

Table 10: Cal. Mkr. Leakage
Freq. MHz Mkr Level dBm
3.4 -60
7.4 -56
10.4 -54
14.4 -52.5
18.4 -51
21.6 -51
25.0 -55
30.0 < -60
56.0 < -60

If you are concerned about this leakage, you can enable Preamp 1 or 2 (or Digi-Sel) whilst the marker is on.

17. Transverter Output Level (August 2014). I measured the transmit power output at the X-VERTER socket with a Millivac MV-723B broadband RF millivoltmeter terminated in 50Ω. Transverter Mode was set to ON (OTHERS menu), and the output for various RF Power control settings recorded at 28.1 MHz in RTTY mode. The DRIVE control was nominal (50%). The test results are listed in Table 11.

Table 11. Transverter Output
100 -15.5
70 -18
50 -22.2
 30 -29
20 -31.5
10 -35
0 -36


Transmitter Measurements:

My friend Matt KK5DR has tested IC-7700 transmitter output and 2-tone IMD. Please refer to pages 5 and 6 of his IC-7700User Review.

Note on Firmware Ver. 1.04 (released December 11, 2008):

Idownloaded Ver. 1.04 from the Icom Japan website, and installed it without difficulty. Icom's description is rather brief: Improvements in the software performance for the DSP UNIT.

I noted a very subtle improvement in NR operation; the NR seems to track voice peaks a little faster and attenuate "highs" somewhat less than in ver. 1.03. In addition, the "noise tail" which I had previously observed in ver. 1.03 - a rise in background noise level on voice peaks, which trails back to silence between syllables - is much less noticeable.

Another NB Note (February 2009):

During a regular Sunday-morning session on  20m and 17m, I encountered bursty, rasping electrical noise at S9 + 10 dB. The noise spikes peaked at 40 - 50 dB above baseline on the Spectrum Scope. NR was at 1 o'clock, NB at 3 o'clock, NB Depth and Width at 8 and 80 respectively and Preamp 2 on.

The NB took the noise out completely, dropping the idle-channel S-meter reading from S9 + 10 to approx. S3. Here is a live demo.

Comment by John Plimmer, a South African SWL and IC-7700 owner:

As a mainly MW broadcast DX'er, I had a great problem with listening to AM stations in SSB/ECSS mode as the audio created loud transients of high notes that were painful to the ear, making listening in this mode difficult.

In the new v1.04 upgrade this appears to be rectified, so I can now listen to AM BBC World Service on SSB/ECSS during the day during non-DX time and hear it without the previous annoying distortion. The upgrade is a significant improvement to the previous problem and does seem to solve it satisfactorily. If I now walk away from the radio for a while and come back it is not even evident I have the NR on, whereas previously it was shrieking at you and you knew it was the old NR.

I never had a problem with listening to the amateur bands in SSB only, so those of you who listen only to SSB signals may not have noticed a big difference.

Note on Firmware Ver. 1.11 (released September 14, 2009):

I downloaded Ver. 1.11 from the Icom Japan website, and installed it without difficulty. Icom's description is succinct: Improvements in the TX monitor audio level.

There was a marked improvement in monitor level. The monitor volume in the Heil ProSet Plus is now quite adequate with Moni Gain at 50% and AF Gain at 30%. Previously, the monitor level was marginal with Moni Gain at 75% and AF Gain at 40%.

Note on Firmware Ver. 1.12 (released April 1, 2010*):

I downloaded Ver. 1.12 from the Icom Japan website, and installed it without difficulty. Icom's description is as follows: Improved DSP filter shape selection function to work correctly when changing the operation mode.

There was a subtle improvement in DSP IF filter performance. Filter skirts appeared a little steeper, with less ringing, and there was a greater contrast between Sharp and Soft shape-factor selections in SSB and CW modes. These are subjective impressions; I did not perform "before-and-after" measurements.

Firmware Ver. 2.00 (released March 28, 2014)

Firmware Ver. 2.10 (released May 21, 2015)  Details

* No, Geraldine, this is not an April Fools' joke!


  1. ICOM IC-7700 HF & 50MHz Transceiver Review, Peter Hart G3SJX, RSGB RadCom, June 2008
  2. ARRL IC-7700 Product Review (QST, March 2005 - members only)
  3. IC-7700 User Manual
  4. IC-7700 Service Manual
  5. Issues and Advances in HF Receiver Testing


To be continued as I learn more about the IC-7700...

Copyright © 2008-2010 A. Farson VA7OJ/AB4OJ. All rights reserved. 
(Copyright in any contributed items reposes with the respective contributors.)
Last revised: September 04, 2022

Expression Web 4