[Elecraft] K3 SWR Anomaly

[Mike Harris]

Looking at the KAT3 schematic indicates that in bypass all the relays 
are set such that all the series inductors are individually short 
circuited and all the capacitors are open circuit. This potentially 
introduces strays due to the short circuit inductors, relays and pcb 
tracks still in the RF path.

I believe it is mentioned on the Elecraft web site that this is the 
reason for the KAT3A having a "true bypass" or words to that effect.

It was suggested that you could tune out the strays by tuning the ATU 
into a dummy load rather than simply putting it in bypass.

I've just checked my KAT3 action by tuning each band via the Ant1 port 
into an Oak Hills Research 100W dummy load and comparing the result with 
simply bypassing it.

All the way from 160m to 10m both states "Auto" and "Bypass" read 1.0-1
On 6m both read 1.1-1.

I normally have the KAT3 in "bypass" given that much of the time my K3 
is looking at the input circuit of an Acom 1000 or the antennas are 
already an acceptable match. There were clearly some long forgotten 
tuning solutions still sitting in the atu memory which were tuned out 
whilst feeding the dummy load.

Regards,

Mike VP8NO

On 21/01/2019 12:40, Grant Youngman wrote:
> The revised KAT3A has a hard bypass with relays.  The KAT3 does a “soft” bypass by placing the tuning network in some kind of neutral position.
> 
> I don’t know what the KAT3 actually does to achieve this, but is it possible this is somehow related to the anomaly you are seeing?  With the KAT3 in  “neutral” (whatever that actually means), since the SWR bridge sits between the KPA3 output and the KAT3 input, maybe that’s something to look at — or at least get more info from Elecraft about how it works and if that might be related.
> 
> Grant NQ5T
> K3 #2091 KX3 #8342
> 
>> On Jan 21, 2019, at 8:34 AM, Ed G <ed.n3cw at gmail.com> wrote:
>>
>> Hi Barry,
>>      I think what I am seeing is related to your experience. I ran through some additional testing. I see perhaps some slight differences between ANT1 and ANT2 positions as far as SWR readings, but nothing significant, at least on 40 meters.
>>      I do now see something I’m even more puzzled over. While collecting SWR readings into a dummy load on 10 meters, I initially did see a high (2.0:1) reading on the K3 meter. I switched in the K3 tuner, and brought the SWR down to 1:1.  But when I then bypassed the tuner, the low 1:1 SWR remained.  This was true for either ANT1 or ANT2.  I really don’t understand that.
>>      So back to 40 meters, where I first noticed the high SWR anomaly. I thought I would try the tuner again, and then bypass it, to see if SWR would stay low on the K3 meter as it did on 10 meters.  It did not.  So I still have the anomaly on 40 meters, and it is the same on either ANT1 or ANT2 positions.
>>      To summarize:
>> 1.  SWR using a dummy load on 40 meters remains constant using an external LP-100A meter, but varies greatly using the K3 meter. As you tune up in frequency on the 40 meter band, the SWR on the K3 meter rises to almost 3:1, and the K3 output power drops back significantly.
>> 2. Antenna tuner settings appear to “stick” on 10 meters, even when the tuner is bypassed.
>>
>> I believe some users, assuming there are others that have K3s which would exhibit this behavior, may not have noticed this anomaly. I do believe it may be responsible for what some users have noticed when using their K3 with an amp such as the KPA1500. That is, power overshoot. I suspect the K3, at least for me on 40 meters, is folding back power in response to a false SWR reading, and this can be seen in how the KPA1500 amp correspondingly drops back in power as the user transmits.  That is perhaps why (again, for my situation) I see the power overshoot only on 40 meters when using my KPA1500.
>> --Ed—