[Elecraft] K3 high SWR at end of transmission

[Joe Subich, W4TV]

On 2026-03-25 6:09 PM, w2yf via Elecraft wrote:

> I'm not expecting firmware re-issued for the entire K-line... but if
> there is a solution that the Hamlib guys can program around that
> works it would be nice.

The hamlib guys should poll again if they get a high SWR reading -
probably require three bad polls in a 5 second period before they
shut things down.

73,

    ... Joe, W4TV

On 2026-03-25 6:09 PM, w2yf via Elecraft wrote:
> Jack,
> 
> The KX3 also exhibits this behavior. This is during FT8 transmissions using
> Hamlib that these values are given to WSJT-X, which promptly kills the
> transmission, if the stop transmit when SWR exceeds 2.5 box is checked...
> fortunately the transmission is ending anyway, but you then have to go back
> to the system and reenable transmit before continuing on. This is release
> 4.7.1 due out soon...(currently a release candidate version in the daily
> updates)
> 
> It is likely the KX2 would also do this (and K3S) although I haven't had a
> chance to test.
> 
> I'm not expecting firmware re-issued for the entire K-line... but if there
> is a solution that the Hamlib guys can program around that works it would
> be nice.
> 
> K4's do not do this by the way, but they also do not use the antiquated SW;
> command.
> 
> Tom
> 
> 
> On Wed, Mar 25, 2026 at 5:23 PM Jack Brindle <jackbrindle at me.com> wrote:
> 
>> I agree with others that because of the way Forward and Reflected power
>> are measured by the microcontroller it is susceptible to end of
>> transmission glitches, where there is a large drop-off between when the
>> forward voltage is converted by the ADC and then the reflected voltage.
>> That is probably about 40 to 50 microseconds apart, which is significant
>> when power is dropping. I don’t have access to the firmware sources, so I
>> can’t tell exactly how the SWR command is handled, nor any side-effects, so
>> all we can do is guess. I learned long ago that it is best to not
>> second-guess Wayne.
>>
>> I would suggest that you re-issue the SW; command when you encounter this
>> situation, to verify that the issue persists. I doubt if you have to do the
>> store than once, but it is possible. If the further readings show no
>> problem, then you can safely ignore it. SWR reporting is kind of a
>> hit-or-miss thing. In fast-changing modes such as CW or SSB the forward
>> power changes rapidly, and the difference between measuring points can show
>> up as an error. Of course when transmission stops this is almost certain to
>> happen, depending on exactly when the measurement cycle coincides with
>> power drop-off. I believe this is what you are seeing. A new SWR request
>> should show it. Wayne may not be correlating SWR read with PTT
>> de-assertion, so he may not be catching the situation when it occurs. This
>> is quite understandable.
>>
>> At this point in the life cycle of the K3, there won’t be any updates, so
>> it is best that you double-check responses when you see anomalies such as
>> this.
>>
>> 73,
>> Jack, W6FB
>>
>>
>>
>>> On Mar 25, 2026, at 2:38 PM, w2yf via Elecraft <elecraft at mailman.qth.net>
>> wrote:
>>>
>>> Jack,
>>>
>>> I am using all the  latest firmware versions for all subsystems of the
>> K3.
>>> As I mentioned you only see this if you are using SW;.
>>>
>>> I've never seen the bar graph indicate any change in SWR which is why I
>>> think it is a bug. If it was a true measurement I would think the
>> firmware
>>> would jump the bar graph or throw a "Hi SWR" on the display.
>>>
>>> 73s,
>>> Tom, w2yf