KPA100 Problem

[Jim Stoneback]

Dealing with the last unresolved problem with my K2/100:

When using the K2/KPA100 on CW, regardless of power setting (above 11 watts,
that is), the power output fluctuates above the set power level by about
10%.  With the power setting at 90 watts, the output will alternate between
about 98 and 88 watts, at about one cycle per 2 seconds.  At 30 watts, its
+/- about 4 watts, etc. Using a dummy load.  It will do this indefinitely.

All other aspects of the KPA100 operation are normal.  With PA turned OFF in
the menu, the barefoot K2 does not exhibit this power fluctuation.  Any
suggestions on what to look at?

73

Jim Stoneback
K4AXF
Drake SW8
K2.ssb.100.at sn 6194
K2.at sn 6263 (under construction)

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

[Don Wilhelm-4]

Jim,

A bit of operational theory first:
The K2 power control circuit is a feedback control loop which includes
the gain of the entire transmit chain in its control loop.  As a result,
when the KPA100 is added, the loop gain for some K2s is increased so
much that the loop oscillates.  A quick study of feedback control
systems will shed some light on the reasons - feedback control loops
should ideally be 'critically damped', but one cannot do that easily
while still allowing a change (KPA100 in/out) in the gain of the loop.  
A compromise between the high power operation and low power control
performance must be made - the loop must be overdamped for low power,
while being slightly underdamped at high power.

The "fix" (it is really a performance customization for each K2):
Resistor R98 on the bottom of the K2 RF Board can be used to control
damping of the power control loop.  Increasing the value will provide
more loop damping (and reduce the oscillations).  Too much resistance
will cause the loop to be severly overdamped and the low power K2 will
never get up to the point of full control.

Increase the value of R98 until the base K2 power output on 10 meters
drops significantly.  It is the choice of each builder just how much
power loss will be tolerable on 10 meters.  It will require at least 9
watts to drive the KPA100 to full power output on 10, so one would not
normally want to increase R98 above the point where the power output on
10 meters is much less than 10 watts, but if R98 is too low, the high
power output can oscillate forever.

A temporary variable pot can be used to determine the correct value for
your K2 - use a 2k (or larger) pot in place of R98, and after you
determine the correct point, remove it and measure its resistance - use
the next lower standard value resistor size for R98.

The proper value for your K2 is that which reduces the high power
oscillation (on 40 meters where the transmitter gain is highest) after 2
to 3 seconds while not reducing the 10 meter base K2 power output
significantly.

73,
Don W3FPR

Jim Stoneback wrote:

> Dealing with the last unresolved problem with my K2/100:
>
> When using the K2/KPA100 on CW, regardless of power setting (above 11
> watts, that is), the power output fluctuates above the set power level
> by about 10%.  With the power setting at 90 watts, the output will
> alternate between about 98 and 88 watts, at about one cycle per 2
> seconds.  At 30 watts, its +/- about 4 watts, etc. Using a dummy
> load.  It will do this indefinitely.
>
> All other aspects of the KPA100 operation are normal.  With PA turned
> OFF in the menu, the barefoot K2 does not exhibit this power
> fluctuation.  Any suggestions on what to look at?
>
>

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com