[Elecraft] Problems on 40M with 100 watt section

[Tom Hammond]

Rich:

>I just noticed that when I am on 40 meters and I am on the 100watt 
>output, I cant transmit. I think something is resetting the MCU 
>because I see "elecraft" flash. It works up to 10 watts without any 
>problems on 40, but as soon as I go above that acts up. All other 
>bands work fine on the 100 watt part.
>
>I dont see anything in troubleshooting but plan to look through the archives.

I know this is gonna sound weird, but it's happened before... and 
it's load-specific... so you won't see it on all loads...

If you have an o'scope, set the sweep to fairly slow, just fast 
enough that the trace is solid across the screen. Set it to DC and 
set the trace somewhere near the bottom of the graticule, so you have 
essentially all the vertical height the vertical deflection to 
2V/div, or whatever will give you almost full vertical deflection 
when 12VDC is connected to the probes. Then, hang it across the power 
supply line to the rig.

Try transmitting on 40M CW and see if you can see the DC voltage 
being pulled VERY LOW just at the instant you start to transmit. 
It'll be a very short downward pulse, but it may pull the voltage 
down toward 6V or less (I've seen it pulled as low as 3VDC). It will 
occur at the instant that the K2 resets, then it'll go right back up.

IF you see this action, and what you've described is 'textbook' for 
this occurring, then the KPA100 is going into parasitic oscillation 
for just an instant, pulling more current that your PS can provide 
(possibly as much as 30A or so), in response to the heavy current 
load the PS is going into foldback current limiting and dropping the 
voltage to near zero for an instant.

This recently happened with the K2/100 belonging to NN0B (a local to 
me). What we found was that the 1mH RFC which is installed back AT 
the KPA100 ANT jack had been installed too close to the SO-239 (and a 
bit 'into' the cutout for the SO-239. The solution to the problem for 
NN0B was for us to physically slide the RFC to the left, as far away 
from the cutout slot as possible. That's all we had to do.

Though this is an UNcommon problem it is not an isolated case.

If you don't have access to an o'scope, double-check that the RFC is 
physically not 'into' or over any portion of the cutout slot for the 
SO-239. If it is, move it as far away from the slot as possible.

Please let me know what you find.

73,

Tom Hammond   N0SS