[Elecraft] P3 woes - repaired but now claims my K3s, puts out 873W!

[Bob McGraw]

My question would be "where is the sensor for the P3 located in the feed 
system?"  It should be between the KPA500 output and the KAT500 input.   
Therefore, when the KAT500 resolves a match, then the line between the 
KPA500 output and KAT500 input is nominally 50 ohms.

If the sensor is located after the KAT500, thus in the feedline to the 
antenna, the impedance and thus the voltage on the line, will impact the 
the indicated value.   Most all power reading sensors are voltage 
reading devices, referenced to 50 ohms nonreactive and should be located 
in a 50 ohm line which has a nonreactive load.  Most power indicating 
devices are calibrated for 50 ohms.   There are some exceptions.

As example;  if the impedance of the antenna is reactive ( most are ) 
and about 200 ohms Z  being fed though 50 ohm coax for the feedline and 
the power is 500 watts, the voltage is about 316 volts.    If the 
impedance were 50 ohms and the power is 500 watts the voltage would be 
about 158 volts.   Here it is easy to see, using a voltage sensing 
circuit, the power indication would be incorrect by some significant 
magnitude.

I've considered adding a power/SWR monitor to my P3 but it would be in 
the coax to the antenna.  I use the internal tuner in my K3S to resolve 
a match, thus the power measurement would be after the ATU in the radio 
and the actual load Z is really unknown.   The only way to determine if 
the power indication is correct with the P3 is to use a 50 ohm dummy 
load or place it in the RF path which is 50 ohms nonreactive.

73

Bob, K4TAX


Message: 2
Date: Fri, 7 Apr 2023 16:15:31 -0600
From: Alan Bloom <n1al at sonic.net>
To: Elecraft Reflector <elecraft at mailman.qth.net>
Subject: Re: [Elecraft] [P3] P3 woes - repaired but now claims my K3s
	puts out 873W!
Message-ID: <3fbef504-357c-f46e-748f-d11e1a822018 at sonic.net>
Content-Type: text/plain; charset=UTF-8; format=flowed

Hi Dave,

There is a "SensorCal" entry in the Transmit Monitor Menu, but I don't
think it has enough range to correct such a large error.

There is also a "MtrScale" entry to set the full-scale power on the
meter display.? Does the reading change when you change the scale?

In addition to the I/O board, the problem could also be on the Sensor
board (the small board that connects the sensor RJ45 connector to the
I/O board).? If one of the resistors on this board went open or shorted
it could cause a power error.? Is the reverse power reading way off as
well?? (i.e. does the SWR read wrong for SWR not close to 1:1?)? If so,
that could be due to a resistor problem on the Sensor board.

  > I'm wondering if the new revision I/O board is somehow incompatible
with the older TX Monitor board,

No I don't think so.? There was a minor change in the sensor design at
one point but it was just a capacitor change to improve the detected
bandwidth so the envelope would display correctly.? (The same sensors
are used for the W2 wattmeter.)

73,

Alan N1AL