[Elecraft] FW: Sherwood Engineering

Wed Feb 25 17:01:32 EST 2015

Ed, 
I haven't checked but a few HF receivers. However, the ones I did showed noise figures of anywhere from 15 to 23 db, which is adequate for most HF work until you get up to the higher end. Pre-amps were invented for that. 

Before 2 db or better noise figure MOSFETS or eqiv. there was the parametric amplifier for those who wanted the best of noise figures for EME or other really low signal levels. I'm really glad technology makes life easier by replacing those evil little monsters. I, for one, never want to use a paramp again for the rest of my life. 

In my early years, there was CCIR-322 that defined the expected value of atmospheric noise by location, etc. I remember that 80 meters had an expected value of about 80 db Fa, or equivalent NF. and that tapered off until you hit galactic noise around 15-18 MHZ. All of which is/was below the average HF receiver noise figure. Where life gets interesting is that atmospheric noise in mostly lightning propagated from the tropics. Why is this important? because even on 10 meters the noise goes up when the band is open. This argument's premiss is that most receivers are over designed if sensitivity is the only parameter used to defined performance. 

If you find yourself in a position where your receiver noise figure at the moment is inadequate, having a pre-amp may help. Why the hedge? well, let's consider the last weekend's DX contest. 10 was crazy open with many very strong signals sitting side by side with weaker to much weaker signals. Had I turned on my pre-amp, and had I not owned a KX3, the result could have been that the strong signals might have hidden the weaker ones due to an apparent increase in the receiver noise floor caused by the mixing of strong signals with other noise. My receiver would have appeared to be less sensitive than it would have been with the pre-amp turned off. This was also the position that was written up by Rhode and others who stated that we were at a point where signal handling was as important or more important than just how sensitive the receiver was. And, ultimate filtering and shape factors have a place in this as it helps keep strong signals away from weak ones. However, there are other attributes of receiver design that help define just how good a radio will be. Unfortunately, some of them interact, and not always the best way. 

Bob Sherwood measures and publishes data that can help a perspective buyer make some guess what receiver would work best in a particular shack. The rankings are his opinion. Admittedly, buying a radio in his top rankings shouldn't disappoint, but all of the data should be looked at. And price is certainly a parameter that needs to be considered along with Bob's data. 

I still recommend that those not familiar with why receivers work and how they function in a real environment search out some of the work printed in the 60's and 70's on receivers design. They contain minimal math and are a good discussion of the real receiver issues. That should put Bob's work in perspective and help make sense of the data. And, IMHO, using a K3 or KX3 is the way to go. And for me, my requirement set drove me to the KX3, and I haven't been disappointed. 

73, 
Barry 
K3NDM 
----- Original Message -----

From: "Edward R Cole" <kl7uw at acsalaska.net> 
To: "elecraft" <Elecraft at mailman.qth.net> 
Sent: Wednesday, February 25, 2015 9:27:32 AM 
Subject: Re: [Elecraft] FW: Sherwood Engineering 

Barry, 

Yes, remembering the pentode front end of my AN-ARC1 (modified for 2m 
AM) had a horrible Noise Figure (which impacted receive 
sensitivity). One could probably have a better NF with today's 
double balance mixers (which are commonly now the front end of HF receivers). 

This is done for best Blocking Dynamic performance. HF bands may 
often have -115 dBm noise levels (or higher) so sensitivity is not 
the main thing driving the design. Lower sideband noise of the new 
synthesizer will directly improve operating in a crowded band. 

For what I mainly do in ham radio, low noise figure is mandatory and 
sensitivity levels like -170 to -180 dBm are strived for. I am 
talking about eme (moonbounce). As frequency gets higher, sky noise 
drops rapidly so that receiver internally generated noise becomes the 
limiting factor. But low phase noise from the LO is also important. 

It is still somewhat controversial whether that is only in presence 
of strong adjacent frequency signals or even important for other 
reasons. I will update my K3 because its possible, so if that has 
potential of improving my weak-signal ability, my station will 
benefit. Synthesized PLL LO's are beginning to replace xtal LO's in 
microwave equipment and particular attention is given to low phase noise. 

I see the new synth as only good stuff, and spending a couple hundred 
bucks to keep my 5-year old radio up to current technology is pretty 
cheap, vs having to buy a brand new radio to get it. 

73, Ed - KL7UW 

-------------------------------------------- 
From: k3ndm at comcast.net 
To: "Burdick, Wayne" <n6kr at elecraft.com> 
Cc: elecraft <Elecraft at mailman.qth.net> 
Subject: Re: [Elecraft] FW: Sherwood Engineering Tests 
Message-ID: 
<1691035197.17114594.1424829382558.JavaMail.zimbra at comcast.net> 
Content-Type: text/plain; charset=utf-8 

I'm old enough to remember that the most important characteristic of 
a receiver was sensitivity. Nothing else seemed to matter. Some 
receivers of the time had 2 RF amplifiers to make sure that they won 
the sensitivity battle. And, what would happened when a strong 
signal, not necessarily S9, would appear, bad things happened to your 
radio. At this point I won't define the date. 

This lunacy was being looked at by a number engineers, to include Dr. 
Ulrich Rhode, W2 something. I forget his call. He said in a series of 
papers in professional journals and Ham Radio Magazine that 
sensitivity was not the most important parameter at the time. It 
would turn out to be LO noise sidebands and dynamic range. That still 
holds today, and now, IMHO, ultimate rejection should be added. All 
of this is shown in Sherwood's data. What this all means is that all 
of the receiver parameters must be looked at to decide what makes a 
great radio. I suggest a review of some of Rhode's papers. His 
writings make extremely interesting reading as it addresses this 
discussion directly. 

73, 
Barry 
K3NDM 

73, Ed - KL7UW 
http://www.kl7uw.com 
"Kits made by KL7UW" 
Dubus Mag business: 
dubususa at gmail.com 

______________________________________________________________ 
Elecraft mailing list 
Home: http://mailman.qth.net/mailman/listinfo/elecraft 
Help: http://mailman.qth.net/mmfaq.htm 
Post: mailto:Elecraft at mailman.qth.net 

This list hosted by: http://www.qsl.net 
Please help support this email list: http://www.qsl.net/donate.html 
Message delivered to k3ndm at comcast.net 