Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: Prev 1 ... 6 7 8 9 10 [11]   Go Down
  Print  
Author Topic: Icoms new IC-7610 Revealed  (Read 74102 times)
K6BRN
Member

Posts: 492




Ignore
« Reply #150 on: November 19, 2017, 11:19:38 PM »

Ed (VE3WGO):

So you agree with me on everything I've said, including over the top price, way late release due to issues found late in the design/debug phase, and need for
Quote
"primitive copper/iron" Preselectors
.  (they ARE primitive - which is why major commercial DSP systems avoid them for many reasons, including cost, complexity, reliability and mass)

And... who said I thought preselector problems were the reason for the IC-7610 release delay?  You did.

But they could be - I really don't know, do you?  Love to hear your take on the matter.

Brian - K6BRN
Logged
VE3WGO
Member

Posts: 167




Ignore
« Reply #151 on: November 20, 2017, 06:53:16 AM »

Hi Brian, yes I agree.  But when I first read your post I connected your comment about the preselector to the reason for release delays and cost increase.  Now I re-read it and I can see my mistake.  Sorry about that.

In the HF range, there are not many choices for making high quality resonant circuits for preselectors, and copper wire wound on iron or ferrite toroid cores seems to be one of the most compact, stable, and low-loss solutions for those frequencies.  PIN diodes could be used to switch the filters, but they can generate IMD in a signal path so relays often get used.  It seems to be so "1980s", but the scheme works well, is reliable, compact, and cost-effective.

But preselectors are helpful, even (or especially with) a wideband direct digital sampling system such as the IC-7300 and 7610 use.  Since the A/D converter is sampling the entire spectrum from "DC" to 60 MHz, it's got to have enough dynamic range to accommodate all the signals in that range...  commercial, ham, government, military, broadcasting, and whatever else and all signal intensities. If it can't handle those then it has to rely on some form of digital AGC when it gets hit with anything in that range, and that usually cuts off the low end of the digital range (the "least significant bit(s)") which reduces the dynamic range. So the preselector improves the immunity to overload from some of those high power signals in that frequency range, and helps to reduce spurious responses and some intermod effects. It even helps improve dynamic range within the ham bands if the preselector is selective enough.  A good comparison is to look at some SDRs that use preselectors to those that don't (I think comparing the Flex 6300 to 6500 comparison shows it)... there seems to be perhaps 10-20 dB or so improvement for those that do use preselectors in some form, even for example cheap dongles compared to some next level SDR receivers that have filter banks.

The fact that SDRs haven't taken over the ham radio market yet tells me that it's still very hard to do, getting hardware for clean A/D converter based receivers and D/A converter based transmitters to work well takes a lot of work. The software development is a major investment, and probably expensive enough to be beyond what some of the current crop of radio manufacturers are willing to do themselves, so I expect some of the current brands to either merge or disappear in the next few years.  Either that, or they'll just sell hardware with some basic user interface and control firmware for survival, and rely on open source SDR software to get the job done.

Engineers and developers might be good at designing hardware and software but they don't seem to be very good at predicting when they will be finished.  (for example, at Icom and Flex)

73, Ed VE3WGO

Logged
K6BRN
Member

Posts: 492




Ignore
« Reply #152 on: November 20, 2017, 08:15:33 PM »

Hi Ed:

No problem.

I've been designing, producing and delivering DSP comms systems for many years (DSP is not new in the commercial world) and have some understanding of the dynamic range issues.

In a practical DSP receiver, the front end of the ADC (the sample and hold) is always preceded by (at least) a fixed band-pass filter to limit noise and interference bandwidth, block DC and prevent aliasing.  It’s never just “DC to Daylight”.  In a very good HF transceiver design, I would have expected a fixed band-pass filter per band rather than band-pass filters AND a variable pre-selector.  I would have used an ADC with enough bits and high enough sampling rate to deliver the needed ENOB to meet dynamic range requirements.  In front of that, I'd put the usual pre-amp and attenuator to extend the range in steps.  Not too many parts.  Dither can also be added to help extend dynamic range, but is more of a last-ditch effort to squeeze the most out of the ADC and carries penalties.  In back of the ADC I'd expect a high end Xilinx FPGA (or dedicated ASIC in commercial systems) that could handle down conversion, decimation, filtering and demodulation at the wider bit widths and sample rate of the ADC – x2 in the IC-7610.

But in the Icom IC-7610 we ALSO have two very (physically) complex and expensive electro-mechanical pre-selectors shoe-horned into the box.  Look at the sheer number of relays in the Icom IC-7610 pre-selector design (its MOSTLY relays). Then multiply them by two.  Then add in the relays in the antenna tuner.  These are SWITCHED pre-selectors, and not the variable capacitor/regenerative or variable iron core designs of yesteryear - which were much simpler.  Relays are neither cheap nor highly reliable.  And they are in addition to the antenna tuner relays.  LOTS to go wrong - reminds me of the old Icom IC-745 with its rotary stepper switch architecture - few are still running without a major rebuild (helped do that to one).  The wire-wound cores used in the IC-7610 do not seem to have been designed for pick-and-place automated assembly either – bet there is plenty of hand assembly in those modules.

Yes - I know Flex does something similar.  And they are known to be expensive. 

My main observation is that (today) it must be cheaper to put pre-selectors in an HF radio than to use better (and more expensive) ADCs and FPGAs – because the parts count and assembly cost would have dropped dramatically without the pre-selectors.  Plus, ADC and FPGA costs drop over time – relays and cores are a more mature technology and do not.  So Icom could have had a roadmap with automatic cost reductions built in to a larger degree.

Regarding keeping to schedule when designing a DSP system – Agree.

Best Regards,
Brian – K6BRN
Logged
N4UE
Member

Posts: 711




Ignore
« Reply #153 on: November 21, 2017, 10:02:41 AM »

Hi Brian, a slight correction to your excellent post.....
Icom's 745 does not use a stepper mechanism. I bought one brand new and still have it on one of my operating positions.
I believe you are thinking of the 701 or 720 series. I heard they were very noisy.....

regards,
ron
N4UE
Logged
K6BRN
Member

Posts: 492




Ignore
« Reply #154 on: November 21, 2017, 05:32:50 PM »

Hi Ronald (N4UE):

You're very likely correct.  A friend found the radio - I think a 720 series unit (IC-720A?) - at a garage sale by  an SK's XYL.  It did not work.  So I convinced him to bring it over so that we could open it up, fix it and get him on the road to full HF privileges.  Boy was that a Rube Goldberg device inside!  Short story - he's now an Extra and uses an Icom IC-7300.  The refurbished IC-720 (something) now has a new home.  And I have a new ham buddy.  It worked out.

Best Regards,

Brian - K6BRN
Logged
ZS5WC
Member

Posts: 635


WWW

Ignore
« Reply #155 on: November 23, 2017, 03:34:10 AM »

 :)I have the IC-720 and 720A with LPF stepper relay.. (No preselector).
Restored both and they work FB.
Also does not suffer from the IC-7300 AGC overshoot pop!..
You can see example here:
https://www.youtube.com/watch?v=mI6Q8niXKvc
Also, any HF rig on an antenna should never have an 'S' meter read Zero.
ICOM runs their rigs with 30dB ZERO "S" meter offset. "S" meter only responds at -90dBm.
Normal start point is -121dBm.
Reason why my IC-7700 was drop kicked over the fence. Hi!
73 de William
ZS4L
Logged
N8FVJ
Member

Posts: 378




Ignore
« Reply #156 on: November 29, 2017, 07:19:58 AM »

IC-7300 AGC overshoot. Another reason I sold my IC-7300. I highly doubt the new IC-7610 will have this issue.
Logged
Pages: Prev 1 ... 6 7 8 9 10 [11]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!