[ICOM] Sherwood Engineering test numbers

[Tom Norris]

On Oct 15, 2006, at 8:26 PM, Jan Robbins wrote:

> Hi John, et. al.,
>
> There is NO MORE RELIABLE set of third-party measures of receiver  
> performance ANYWHERE in the world  than Rob's.  Whether you, in  
> your particular environment and set of needs, care about what he  
> says, is entirerly

[snip]

Until recently I worked for a usaf (contractor ran) pmel, and not  
only did we use many/most of the methods he uses in his testing, we  
were familiar with most of types of test instruments he uses, and we  
used -and calibrated -  such things in the course of our daily work.

I can certainly say Mr Sherwood's methods and data are quite valid  
and reliable.  I'd much rather use his data as a reference that that  
published by the ARRL lab.

That said, I don't care for the receiver performance of my 706MkIIG  
compared to my 756ProII or my R-390 or R-390A.


*John, W5TD sez -- *
"... he measures the 2khz IMD DR for the Icom 706MKIIG at 74db.
This places it right in line with the PROII and PROIII at 75db, and
better than the 756PRO (71db) ... and the Icom 746 (70db). "

The 706* series isn't meant to compete against anything else other  
than those in its class. The IC-706* suffers from design "faults"  
inherent in a transceiver packed into a small box.  These things  
aren't reflected in the numbers.  One thing that comes to mind is  
filter "blow by" that affects selectivity.  Compare practical things  
such as limited number of bandwidth choices, heat dissipation of a  
smaller package. It's physical size  doesn't allow the same circuit  
layout of, lets say a 756, and is not going to have the same  
performance overall, and so on.

Don't simply compare numbers, compare design and purpose.  While the  
data on Sherwood's page accurately show the best-case numbers for the  
receivers tested, it *is* a bench test, and real world even my lowly  
vacuum tubed R-390A's can outperform many modern receivers - even  
that in my ProII under some conditions.

One specification doesn't the whole radio make.  Back to the original  
question - yes, his testing methods are correct.

Hopefully Adam will chime in here, as he has a good deal of technical  
information on his pages on the subject I believe.

73

Tom NU4G