[Elecraft] Flaw in the 1.8 kHz roofing filter - Final for 2009

[Geoffrey Mackenzie-Kennedy]

Dave,

Before I list a couple of home truths, may I say that I agree with your 
comment.

1)
The IMDDR3 data  from ARRL and others is a report of the receiver's IMDDR3 
performance, not that of the filter's IMDDR3 performance. The data does 
provide some indication that the IMDDR3 performance of filter "A" might be 
"worse" than that of filter "B", "C" or "D", but filter "A" itself might not 
be the cause of the problem. The filters should be tested as stand-alone 
items before putting blame on any particular filter.

The designer of a receiver should know the IIP3 of the filters to be used, 
otherwise he is flying blind.

2)
The effects of SLC (Surface Layer Contamination) on the performance of 
crystals have been known for many years, and methods used during the 
manufacture of crystals to avoid SLC have also been used for many years. SLC 
is usually caused by dirt or particles of quartz which has / have not been 
cleaned out during the manufacture of the crystal. So called "computer 
grade" crystals, which are cheap, can be expected to be contaminated, 
because most are produced for non-critical oscillator applications.

"Dirty" crystals if used in a filter will have a bad effect on the filter's 
IMD performance and its loss.

73,

Geoff
GM4ESD


David Gilbert wrote on Monday, October 19, 2009 at 8:13 PM

> Personally, I find it easier to believe that Inrad fixed the problem
> based upon Elecraft's subjective confirmation of such than I do your
> totally subjective implication that they might be collectively and
> intentionally lying to us.
>
> Dave   AB7E