Re: KAT3 (not KAT3A) issue - older K3 take note

jeff stai wk6i

Glad I said something! G3PQA noted the explanation in another email here:

But in a nutshell from WE6R:

"Overview; There is an 8.215 trap on the RF board, the trap on the KAT/KANT3 boards was an additional trap that was later found to be way overkill. 

"On the KAT3 ATU, it is on the antenna side of the tuner, so it is subject to high SWR and can heat and burn.

"On the KANT3 you cannot cannot run with high SWR or the radio will back off, so not a problem on these boards.

"Early ATUs had a small coil L10 and single pink/brown capacitor C10. The cap can fail internally invisibly or catastrophically! The coil can get burnt."

And yes you should redo the TX calibration after making this mod. :) 73 jeff wk6i

On Tue, Jun 15, 2021 at 5:49 AM Geert Jan de Groot <pe1hzg@...> wrote:
[re-re-send; groups,io doesn't accept posting, but there is no way to upload pictures either - not cool!]

Hi Jeff,

On 15/06/2021 03:56, wrote:
TL;DR: Got a KAT3 in your K3? Next time you have your K3 open, pull out
KAT3 and preemptively remove L10 and C10, and replace with a jumper. You
don't need them and they are probably messing you up at times.

Upon reading your email, I made a mental note to do this next time the radio is on the desk, but then google found twitter messages from F8CRH and on the bench my K3 went.

When I removed my KAT3 I found traces of impending doom and pyrotechnics and made a few pictures. IMHO, if you have a KAT3 (not KAT3A), you want to open your radio and make the modification NOW.

Please find attached (I tried uploading this to, but either don't have privilege or it's my ten left thumbs) pictures of what I found of my still-working KAT3: the board, details of L10 singed pretty badly, and L10/C10 after I removed them.
(I also left them on but my ISP will probably break that in a few months, so that URL is not google-safe)

In my case, the previous owner of my K3 was an avid 6m EME-er. On the other hand, my recent experiments on 6m were not so hot, indeed with a 6m dipole that I could not make to resonate - perhaps I should try again.

I still wonder why these parts were on their way to destruction.
96nH // 3900pF makes 8.215 MHz, these were a trap for the IF of the K3 - why would they fail while working on these higher bands?

I note that the KANT3 (the PCB you have if you don't have an antenna tuner), the same parts are on the board, and they are missing on KANT3A.
And if you bought a MFJ950 K3 antenna tuner, you will find the same parts (L1 and C9, where C9 even has a lower working voltage of 500V instead of the 1600V of the Elecraft part). As I don't understand the failure scenario, I don't know if they same problems apply in these boards, but YMMV.

Come to think of it, you probably want to re-calibrate TX power after making this mod...


Geert Jan PE1HZG

Jeff Stai ~ WK6I ~ wk6i.jeff@...
RTTY op at W7RN
Twisted Oak Winery ~

Join to automatically receive all group messages.