[Elecraft] KX3 SWR meter accuracy on 50 MHz?

[Wes]

There are so many variables here that it kind of defies analysis.  First of all, 
the KX3 uses a BNC connector. Almost always, as you have done, connecting to a 
good load, that will handle some power, probably requires adapting to type N.  A 
reasonable load might be something like a NARDA 366NM which has a VSWR spec of 
1.2:1.  I personally use a series of Narda 766 attenuators which are 1.1:1.  
Adding a typical BNC to type N adapter, Pasternack PE9002, which is rated at 
1.3:1.  (A lot of manufacturers like Amphenol don't even rate for VSWR)

Converting these two values to reflection coefficient, adding them (worst case) 
and converting back to VSWR gives 1.57:1. So the load could be the 1.5:1 that is 
measured.  But this is only a part of the issue.  The KX3 VSWR bridge has 
unknown directivity and sensitivity since it uses diodes at very low power.  
Furthermore, it's not driven by a good 50 ohm source, the amplifier output is 
unknown and there is a BPF in the circuit.  At the output side there are at 
minimum some relays before getting to the output spigot.

In summary, I wouldn't worry about it. If you really want to know what the load 
Z (antenna) is, measure it with an appropriate instrument calibrated with known 
standards.

Wes  N7WS


First, On 9/17/2020 7:36 PM, Drew Arnett wrote:
> Tech support said they saw 1.5 to 1 on 50 MHz with their dummy load.
> HF looked great, of course.  They didn't mention if they also used a
> microwave load.  :-)
>
> Best regards,
>
> Drew
> n7da
>
> On Sun, Sep 13, 2020 at 2:20 AM Drew Arnett<arnett.drew at gmail.com>  wrote:
>> I did before shipping in for repair with a directional coupler.  Looked great.
>>
>> On Sat, Sep 12, 2020 at 7:13 PM Nr4c<nr4c at widomaker.com>  wrote:
>>> Well, it depends on the “Dummy” load.  Have you swept the load with a good spectrum analyzer or VNA?  It may not be 50 ohms at 50 MHz
>>>
>>>
>>>
>>> Sent from my iPhone
>>> ...nr4c. bill
>>>
>>>
>>>> On Sep 12, 2020, at 11:51 AM, Drew Arnett<arnett.drew at gmail.com>  wrote:
>>>>
>>>> After getting my KX3 back from repair (before my next planned contest
>>>> operation, thanks!), I ran some measurements to both see how great the
>>>> repairs were and to get a new baseline for future measurements.  TX
>>>> efficiency problem was gone.
>>>>
>>>> However, I'm left to wonder about the accuracy of the internal SWR
>>>> meter for 50 MHz.  Using a microwave dummy load of adequate power
>>>> rating and just a coaxial adapter, I measured the following:
>>>>
>>>> 14 MHz @ 1 W setting
>>>> VSWR 1.0 to 1 (per KX3)
>>>>
>>>> 28 MHz @ 1 W setting
>>>> VSWR 1.0 to 1
>>>>
>>>> 50 MHz @ 1 W setting
>>>> VSWR 1.5 to 1
>>>>
>>>> 50 MHz @ 4 W setting
>>>> VSWR 1.6 to 1
>>>>
>>>> Not horrible.  Not great.
>>>>
>>>> It's fine if it's not as accurate on 50 MHz.  I just want to know if
>>>> this is normal.  Has anyone else looked at that and can share what
>>>> they see?
>>>>
>>>> Thanks and best regards,
>>>>
>>>> Drew
>>>> n7da
>>>> ______________________________________________________________