Re: [Elecraft_K3] Compare the K3 with the IC-7300

Rob Sherwood

Hi Dick,


With all the whirlwind testing of the IC-7851 and IC-7300 for my Visalia and Contest University presentations, I haven’t gotten a chance to test the transmit IMD of the 7300.  Adam Farson has the 7300 transmit data, and it looks pretty much like a typical 13.8-volt solid-state PA.  If one wants to make a generalization, the 50-volt PAs are about 6 dB better than the 13.8-volt PAs.  What is much better?  Yaesu in class A, but their ALC completely ruins the class A clean signal.  So if you want a clean class A with Yaesu, you have to run zero ALC and very little or no processing.  When I talked to Yaesu about this problem, their response was, “our very fast ALC decay make the signal louder”.   Processing should be done before ALC, at least in my opinion.


What else can really make an SSB signal clean?  Pre-distortion, only available as far as I know in the Apache ANAN series, called PureSignal.  Also the rather obscure ADAT ADT-200A has pre-distortion, but no market share.  I tried to run the ADT-200A in the January CQWW 160 meter contest, but I couldn’t solve several problems, so after a few contacts I gave up.  Of course pre-distortion wasn’t an issue on CW.


I have several radios to put on my list, which I will get to after a trip to California and Visalia DX Convention. 

IC-7851, IC-9100, IC-7300 and maybe the ADAT, but it has some really odd dynamic-range results. 


Will hams pay for a cleaner signal ?  I asked Jack Burchfield of Ten-Tec that question a few years ago, and he though not.  Personally I want a clean SSB signal, and a clean CW signal without key clicks and excessive CW bandwidth.  Many radios have a CW rise time that is far too fast.  I could adjust the rise time of my TS-990S to 1 ms.  That is nuts, and the Icom can adjust to 2 ms, but the way they measure rise time is odd.  When I measured their default 4 ms, I got a real number of 2 ms, and 8 ms came out a real 4 ms. 


Bob Allison has a new sidebar in the May QST product review, and he hits on the problem that a lot of OEMs are allowing the operator to pick inappropriately fast rise times. His explanation of new ARRL testing methods is a must read.


What puzzles me is why some rigs that will never be run mobile are running on 13.8 volts with today’s 50-volt FET power amp transistors available.


Rob, NC0B   


From: Elecraft_K3@... [mailto:Elecraft_K3@...]
Sent: Monday, April 11, 2016 11:38 AM
To: Elecraft_K3@...
Subject: Re: [Elecraft_K3] Compare the K3 with the IC-7300



I've been thinking about the signals of the two pictures Rob sent me, and I have this comment;

I started in the hobby before we had solid state rigs.  With me it is only a hobby, not an obsession. I have forgotten much,
but who can forget how we had  to manually resonate the final tank circuit each time we changed frequency.  Tubes took a beating.
I couldn't wait to get a radio with solid state finals. In solid state HF rigs, I've had Heathkit, Kenwood, and TenTec, but no other brands.

I wanted a nice radio for what may be my last sun-cycle, and I chose a K3.  I was active on CW at the time.  I'm not the nectar type,
but during the past five years, I have read gripes about things on a K3 that were so much better than what I had experienced on other radios. 

But, I am disappointed to see the picture of the K3 transmit signal.  I've been thinking about what it would take to clean up the signal in a K4
as it has been suggested.  Then, the question arises; "Will the average Joe Amateur even care?"

I think we will see a stampede to buy the IC-7300 with it's toys, at it's price, instead of those models at, or near, the top of Rob's list.
(Though at this point, I do not know where the 7300 will be found on the list, or what the transmit signal will look like.) 

Dick, n0ce

On 4/10/2016 10:30 PM, Rob Sherwood. rob@... [Elecraft_K3] wrote:


Hi Dick,


Odd-order transmit distortion products are within a few 10s of kHz of your transmit frequency.

We are not talking about harmonics.

I have attached two small 100K pictures, if they make it through the reflector.

The span of the spectrum analyzer is 40 kHz, + 20 kHz and – 20 kHz of the transmit frequency.

If they don’t come through, e-mail me direct, rob@....

One is of a Collins 32S-3 and the other a K3.

This is what Knut is talking about.

The comparison is typical of a solid-state PA vs. a pair of 6146A tubes with negative feedback.

Most any solid-state rig today would look about the same.






Check out the automatic photo album with 2 photo(s) from this topic.




If this email is spam, report it to

Join to automatically receive all group messages.