[Elecraft] K2 dual keying problem

[Don Wilhelm]

David,

I would rather speculate that it is a uP timing situation coupled with
contact bounce at the key and the forward voltage drop of the diodes.  If
one of the diodes is a bit slower than the other, the voltage seen at the uP
pins may actually look like one pin has gone to a low state before the other
(that is the condition to generate the iambic alternating dots and dashes).
This problem can be caused by contact bounce at the hand key - cleaning the
hand key contacts will help, but one can never cure contact bounce at a
mechanical closure point.

The net of the story is that the uP must believe that both the dot and dash
contact closures occurred together, and the only way that can be known is
for it to sense a low voltage on both its input pins.  If the polling of
those pins happens to occur during a contact bounce period and one of the
diodes is a bit slower than the other it will appear to the microprocessor
that only one of the contacts was closed.

I am surprised that two diodes in series appears to work better - in fact it
will make the low state at the uP pins at an even higher voltage than one
diode.  I would suggest that you try using Shottky diodes (like 1N5711)
instead to lower the voltage at the uP pins when the hand key is closed.

So clean your key contacts and go back to one diode for each side and if
still a problem switch to Schottky diodes.

73,
Don W3FPR

> -----Original Message-----

>
> Mindful that I am neglecting my hand keying, my sending being spoilt by
> the ease of using an iambic paddle, I have connected a straight key
> across the paddle keying line using a pair of 1N4148 diodes as shown in
> the K2 Manual.   This works reasonably well, but occasionally when
> pressing the straight key I get a string of alternate dots and dashes.
>
> I presume that the problem is caused by the reverse resistance of the
> diodes being too low.  I tried putting two 1N4148 diodes in series in
> both lines. This is better but has not cured the problem completely. Has
> anyone else had this problem? Or does anyone have a suggestion as to how
> both a paddle and straight key may both be connected to the K2 with
> reliable results.
>
> 73 de David G4DMP
>
>