ICOM IC-7610 review

Review by Steve Ireland VK6VZ/G3ZZD

• E-mail: [e-mail]

he new Icom IC-7610
must be one of the most
anticipated new radios
in the history of amateur

T

K2 set the scene for an overhaul of the
superheterodyne architecture and the
release of its bigger brother K3 in 2008 ‒
which went on to dominate HF DXing and
contesting for the next decade ‒ the rst
true ‘Software Dened Radio with knobs’
the Icom IC-7300 (reviewed in August 2016
PW) has raised the curtain for the IC-7610
to challenge the K3 for its crown as arguably
the most popular radio for demanding HF
operation.

In terms of how successful the IC-7300
has been since its launch in March 2016,
the word is that currently about 20,000
have been sold worldwide. In contrast, I
understand Elecraft has sold about half this
number of K3 and K3S – still a huge amount
for an amateur radio transceiver ‒ since
sales started [1].

Time will tell how this scenario plays
out but as an HF DXer and contester, let
me say from the outset that I’m very happy
with my IC-7610 and condent that digital
architecture transceivers such as those
from Icom, Apache Labs and FlexRadio
Systems are bringing to a close the days of
the superheterodyne transceiver, even those
with a digital signal processing back-end.

It’s Not Two IC-7300s

Let’s start by getting the biggest elephant
in the room out of the way by saying that
the popular idea of the IC-7610 with its dual
receivers being like two IC-7300s in one box
is not only misleading but plain wrong.

This comparison is often used to lead into
to say that at currently around £3500 in the
UK, the IC-7610 is overpriced in comparison
to the IC-7300, selling at around £1200.

To start deconstructing ‘the elephant’, the
front-end ltering on each IC-7610 receiver
is superior to the IC-7300, with better
bandpass ltering and the one-pole Digi-Sel
preselector module, which greatly improves
the out-of-band rejection of big signals that
can cause annoying intermodulation. If you
are like me and have a powerful medium
wave broadcast station relatively nearby, this
is a really important feature.

When it comes to working DX, particular
the major ‘once every 20 years’ DXpeditions
such as the recent, sadly-aborted 3Y0Z
operation, having two equally good receivers
is priceless, because these expeditions
will generally use up to a 5 to 10kHz
‘split’ between their transmit and receive

radio. Just as the Elecraft

The Icom IC-7610
HF/6m Software
Dened Transceiver

Steve Ireland VK6VZ/G3ZZD was one of the first to get his hands
on the eagerly awaited Icom IC-7610. He reports his findings and
explains why SDRs make good DX/contest-grade HF transceivers.

frequencies. In the IC-7300, with its single
receiver and dual VFOs, you can toggle
back and forward between the receive and
transmit frequencies but this is nothing like
being able to listen to both frequencies
simultaneously, by having one receiver in
each ear of your headphones, as you can on
the IC-7610.

Dual receivers enable you to know
exactly what is happening in the pile-up and
on the transmit frequency.

The next point to make is that instead
of getting the 14-bit version of the popular
LTC2208 analogue-to-digital converter (ADC)
that lies at the heart of the IC-7300, it looks
though we get the 16-bit version [2] in each
of the IC-7610 receivers - more of this later.
16-bit ADCs are used in the vast majority of
high-end computer-assisted digital sampling
SDRs, such as Apache Labs ANAN series
[3] and Flex-Radio’s 6000 series [4] and the
LTC2208 was originally used in the ground­breaking HPSDR [5].

For those like myself who are keen CW
operators or who dislike clicking transmit­receive (TR) relays such as used in the
IC-7300, the IC-7610 has solid-state, totally
quiet, TR switching.

On the IC-7610, the CW keying waveform

generation and shaping is carried out in the
radio’s RF Field Programmable Gate Array
(FPGA) to minimise any latency (the delay
between Morse characters being formed by
a key/keyer and then actually transmitted).

Unlike the IC-7300, the IC-7610 comes
with superb audio peak ltering, adjustable
in frequency, width and gain and available
on both receivers, which is better than any
analogue or digital audio lter that VK6VZ
has ever used. This includes the excellent,
well-regarded ones on the Elecraft K3, Ten­Tec Orion 2 and Yaesu FT-1000. The IC-7610
lter does an amazing job of cleaning up any
CW or digital signal – not just in helping to
dig weak ones out of the noise.

The high denition touch screen (7in
diagonal) is substantially larger and easier
to use than the one on the IC-7300 (4.3in)
and you have the additional ability to use an
external VGA monitor by plugging a cheap
DVD-I to HDMI socket adapter (typically
around £5) into the IC-7610, which in
turn plugs into a HDMI to VGA converter
(usually around £20) into which the screen
is plugged, giving you a massive view of the
IC-7610 spectrum scope and transceiver
functions.

While it is possible to give the IC-7300

10 Practical Wireless April 2018

an external display by plugging it into a
personal computer running the N1MM
logging software [6], as most IC-7300 users
will know, you cannot simply plug a monitor
screen into it.

Other advantages of the IC-7610 over
the IC-7300 are its larger physical size,
which results in a much more comfortable,
spacious and easy to use front panel.
It is great, for example, to have a large
Independent Receiver Tune control directly
adjacent to the main tuning knob. The rear
panel offers much better connectivity than
the IC-7300, particularly where antennas
concerned – you get two PL-259 antenna
ports, plus a BNC-format receive antenna
input and output.

While the IC-7610 is not a dual IC-7300,
the interface of the original software written
for controlling the IC-7610 is very similar
to that of the IC-7300. What this means in
practice is if you have used a IC-7300 and
customised its menu settings, in particular
the spectrum scope and receiver ltering
parameters, setting up the IC-7610 how you
want it is a breeze.

Even if you haven’t, the super videos on
setting up and using the IC-7300 available
on YouTube ‒ in particular, my favourite ones
by Steve Ellington N4LQ [7] – make life very
easy for new IC-7610 users.

Design and Construction

The Icom IC-7610 uses a direct sampling
SDR architecture, with two identical,
independent digital down conversion (DDC)
receivers and a digital up conversion (DUC)
transmitter. By independent, I mean that
the receivers can operate independently
on different frequency bands and different
modes. As standard, the receivers both
share the same main tuning knob but you
can buy a second stand-alone tuning knob
(Icom RC-28) to tune the second (designated
SUB) receiver (UK price to be announced but
probably around £250).

The tuning of the two receivers can be
tracked/coupled together, enabling you to
carry out diversity reception by, for example,
plugging separate horizontal and vertically
polarised antennas into each receiver and
then listening to one receiver in your left ear
and the other in your right ear.

A look at the block diagram and
schematics which come on the CD supplied
with the IC-7610, Fig. 2, shows that each
receiver front-end has its own ADC, ADC
driver/preamplier, stepped attenuator,
group of bandpass lters and Digi-Sel
automatically tracking preselector. If you
listen very carefully while tuning one of the

The IC-7610 in use at VK6VZ.

receivers, you can faintly hear its associated
preselector operating/tuning.

In addition, there are separate digital­to-analogue (DAC) converters, using 14-bit
ISL5961 chips, and audio chains on each
receiver, enabling binaural reception of
signals. The encoding of the analogue
transmit audio and the decoding of the
receive audio bitstream is carried out by an
Ashai-Kasei AK4621EF dual 24-bit 192kHz
stereo audio CODEC, which interfaces with
the FPGA.

As most who have read about direct
sampling receivers will know, it is the ADC
that forms the heart of any digital up/down
conversion or ‘digital sampling’ radio,
converting RF signals to digital data by
rapidly sampling them. The symmetrical
data lines coming from the two LTC2208

©

ADCs go to the main IC-7610 Altera

RF
Field Programmable Gate Array (FPGA)
which then carries out most of the RF and
digital signal processing and frequency
management in the IC-7610.

I’ve recently discovered [8] the FPGA is
congured as a digital down converter and
delivers a digital 12kHz ‘IF’ to each of the
receiver DACs, which converts the digital
signal back to audio. All signal processing
functions are performed in the FPGA.

The use of the FPGA after an ADC in this
manner is the conventional, time-tested way
to deal with the huge amount of data that
comes out of an ADC – a crucial and difcult
task that takes a huge number of gates and
some very special programming ability. My
good friend and regular co-writer on SDR
Phil Harman VK6PH uses the analogy of
‘drinking from a re hose’ when it comes to
carrying the necessary function of an FPGA.

In the case of a 16-bit LTC2208 ADC that
samples analogue signals at the rate of 130

mega samples per second (Msps), the data
output stream would be 16 x 130Msps - over
two gigabytes per second!

To deal with this huge amount of data
and take advantage of the high dynamic
range the 16-bit LTC 2208 offers, the FPGA
rst carries out a specialised form of ltering,
known as decimation, using digital ltering
formed from gates in the array, before
carrying out other processing tasks, using
further gates.

In addition to the main FPGA that
supports the two ADCs, each IC-7610

©

receiver has a separate Lattice

common
FPGA, which carries signal processing tasks
that specically relate to the associated
receiver.

Other separate, independent functions
for each receiver include spectrum scope/
waterfall displays with 100dB dynamic
range, audio and squelch controls and
external speaker jacks.

Measurements and On-Air Performance

Just about every radio amateur who is
interested in receiver performance and the
associated measurements would be aware
of the famous receiver test data table [9]
published by Rob Sherwood NC0B, founder
of Sherwood Engineering, which is compiled
from equipment testing carried out in Rob’s
extremely well-equipped laboratory.

Rob’s table rates receivers/the receive
sections of HF transceivers produced over
the last 30 years or so according to their
ability to deal with strong signals that are
relatively closely spaced (mostly 2kHz apart).
He calls this parameter ‘Dynamic Range
Narrow Spaced’ or ‘DRNS’ and if a radio
has a DRNS of 75 to 80dB in his table, my
experience is this is good enough for general
(DX and ragchewing) operation unless you

April 2018 Practical Wireless 11

Review: Icom IC-7610

The IC-7610 rear panel has plenty of interfaces to antennas and external devices.

are unlucky enough to have a radio amateur
who lives a kilometre or two away, runs (at
least) the legal UK power limit and is active
on a daily basis when you want to go on the
air.

If this is the case, then you are likely
to need a radio with a DRNS of at least
6dB higher (81 to 86dB) to operate within
a kilohertz or two of your high-powered
neighbour.

Just before Christmas, Rob put an Icom
IC-7610 through its paces in his laboratory
[10] and was kind enough to allow me to use
the associated measurements in this review.
The sample IC-7610 performed very well,
displaying a DRNS of 98dB (with the IP+
control switched on) and 90dB (with the IP+
switched off) at 2kHz spacing.

When it comes to Rob’s very long and
thorough table, this DRNS currently puts the
IC-7610 in eleventh place - some ve places
above the IC-7300, which has a DRNS of
94dB (IP+ on) and 81dB DRNS (IP+ off).
However, to put this into context, the leading
radio in Rob’s table is a FlexRadio 6700 that
he measured at 108db DRNS (preamp on)
back in 2014. A second Flex 6700 sample
measured by him in March 2017 measured
96dB DRNS (preamp on) and 99dB DRNS
(preamp off).

As Rob said in a recent interview [11]
on the Ham Talk Live internet programme
with Neil Rapp WB9VPG, the top direct
sampling/digital down conversion receivers
and transceivers he has tested all have
similar, consistently-high DRNS gures. The
2017 Flex 6700 sample measured 99dB,
the Apache Labs ANAN 200D 99dB, the
Microtelecom Perseus receiver 99dB, the
Icom 7610 98dB (IP+ on) and the Icom
R-8600 98 dB (IP+ on).

Now I have my own brutal means of on­air testing how well a receiver performs on
very strong signals, which is: “Can I operate

within 1kHz of the formidable contesting
station of my good friend Kevin Smith

VK6LW, who lives just a couple of kilometres
away, whichever way ‘Kev’ is beaming and
whatever band he is operating on?” To give

an idea of his signal strength, on the 160m
band where Kev’s signal is loudest, he is a
true S9+40dB.

To get a feel for how receiver DRNS has
improved over time, in the 1990s I used top­of-the line radios for that decade, including
the FT-1000MP (68dB DRNS) and FT-1000
(69dB DRNS), neither of which were ‘Kev­proof’.

Nowadays DRNS for new top HF radios
produced by all the manufacturers are well
over the 81 to 86dB milestone and the
problems of front-end overloading have,
as a result, generally faded into history, at
least at my QTH. My IC-7610 was tested out
during the 2017 CQ WW CW contest and the
2018 CQ 160 CW Contest and proved totally
‘Kev-proof’, as did a friend’s ANAN 200D
(99dB DRNS in Rob’s table), which was used
alongside the IC-7610.

Not only I could tell no difference in Kev­proofness between the two radios, but they
performed at least equally well in this regard
to their predecessors at VK6VZ, the Ten-Tec
Orion 2 and Elecraft K3.

However, for me, using the IC-7610 on-air
was more fun and effective than any of its
predecessors. As Rob Sherwood said in
the Ham Radio Live interview: “It is crazy to
judge a radio by one parameter” and, to me,
it’s particularly crazy to do this just on DRNS
when all new contest/DX-grade radios work
so well on strong signals.

Why SDRs Make Good Radios

One of the reasons why the Elecraft K3S,
Icom IC-7851, Yaesu FT-5000D and other
superhet-based transceivers are at the top
of Rob Sherwood’s table is because they
use crystal ltering as the rst point in their
architecture after the antenna to provide
signal-width selectivity. Crystal lters placed
in this position are known as roong lters

and protect the rest of the receiver from very
strong signals outside of the very narrow
width of the multi-pole lter.

However, what may seem to be the
strongest point of this latest variation of
the superhet architecture can also be
arguably its ‘Achilles’ heel’ when it comes
to producing very clean and clear-sounding
audio in your speaker or headphones.

All radio frequencies that radio amateurs
use are covered in noise – atmospheric,
ionospheric and man-made. When noise
spikes or pulses pass through a crystal
lter, the phase response of the lter varies,
depending of the frequencies of the noise
passing through it. However, when the same
spikes or pulses pass through an ADC, it
responds to them in a linear manner and the
phase response stays the same, irrespective
of the frequency of the noise.

As a result, noise can sound harsh and
grating to our ears from a superhet equipped
with a crystal lter(s) but the same noise will
sound mellow and easy on the ears when
heard on an SDR.

Superhets also generate additional
internal noise that gets overlaid onto signals,
exacerbating the situation. This is because
their architecture makes RF signals audible
by stepping down (or sometimes, by
temporarily stepping up) the frequency of
the desired signal. This means mixing the
signals with other signals – and every active
superhet stage of mixing adds noise onto
the signal.

The Elecraft K3 and other recent top-line
superhets cleverly minimise this problem by
having only one true intermediate frequency
(IF) stage and thus one mixer. However,
some superhet transceivers have three IF
stages and thus three mixers and sometimes
add more crystal ltering to improve the
steepness/narrowness of the receiver’s
selectivity as part of these stages, which all
adds additional phase distortion.

To get around this problem of mixers
adding noise and to minimise the varying
phase responses of crystal lters, modern
superhet transceivers reduce the number
of mixers and crystal lters and, instead,
get more selectivity and noise reduction by
turning the analogue signal into a digital one
after one IF stage.

The end result is that weak signals
on modern superhets can sound highly
processed and thus sometimes hard for us
to distinguish.

Direct sampling SDRs and their ADCs
respond to weak and strong signals in a
totally different way to analogue designs
using (non-linear) mixers and crystal lters

12 Practical Wireless April 2018