SK3 and touch paddles (long!)


From: Richard Hieber (
Date: Fri Jan 05 1996 - 11:05:20 EST

Hi gang,

there's been much talk recently about the CMOS Super Keyer 3 on this
list, but I simply have to join the chorus: It's great! I received it at
the 31st of December last year, built it on the January 2nd and have
been playing with it ever since ;)

The other keyer that I own is based on a 8749 chip and has a good
reputation in the local ham community. I always thought this one is
probably the best keyer on the market. But the SK3 beats it hands
(paddles) down!

Some questions

OK, enough ranting. There are a few very very minor things that are
unclear to me.

1) If I press button 5 and 6 to enter the inquiry mode, then paddle in a
   command, *sometimes* there is no response. When I repeat it, there is
   *always* a response at the second try. Do others experience the same?

2) Is there an inquiry function to find out what the current speed is?
   The R command only gives the range which can be varied by the 100
   kOhms pot.

3) Currently I am powering the keyer with a 9V battery. The
   voltage is reduced to 5.4 Volts with a resistor and a Zener diode in
   series. (This is only a temporary setup - I will probably buy a 6V
   Lithium battery for it and drop the voltage with a diode or a LED).
   The battery that I am using is probably weak and I suppose the
   voltage has gone a few times below the minimum 3.5 Volts. I noticed
   corruption of a few configuration parameters as an result. The text
   memories didn't seem to be effected.

   I am *sure* this is only because of the barbaric power feeding
   arrangement, having the SK3 in parallel with a 5.4V Zener - and I
   don't even have a meter here to measure if this is the real voltage
   the keyer is seeing. I might think the standby current running
   through the zener is too low really to keep the voltage stable,
   especially when the sidetone is on and draws additional current. Due
   to an car accident three weeks ago, I don't have any instruments at
   my study QTH right now and have to make do with what I find in the
   clubstation - that's the reason for the temporary setup.

I don't have any real contest experience so far, but this weekend there
is a contest from the German CW club AGCW-DL, and I am thinking about
participating to make use of the new toy ;) I will be using my
high-quality Schurr paddles and the QRP+.

After a sufficient period of playing with it, the SK3 will find its
future working place in a companian box for the QRP+, together with the
Schurr paddles, with a big gel cell and a QRP tuner. This was the idea
why I ordered the kit. But after having seen how small and how good it
is, I am tempted to order another kit and make a standalone keyer out of

Touch paddles

For this one I would like to have touch paddles. I plan to mount them on
the first half of the sides of a small case, with just enough space for
the SK3 circuit board, a battery, a minature speaker, the speed control
pot and the circuit for the touch paddles. The six memory buttons will
be mounted on top. The base plate will be something heavy to keep the
keyer on the table ... but I could always hold it at the back part with
with my right hand (I am keying with the left). I imagine the shape of
the keyer to be conical but not necessarily in steps like in this small
diagram (couldn't do better with ASCII):

                            VIEW FROM TOP

     Touch paddles +------------+
        DIT +-----------+ | B |||Speaker
      +--------+ SK3 circuit | A |||-+
FRONT | | T E ||| |Speed BACK
      +--------+ | T R |||-+control
        DAH +-----------+ | Y |||

Some more questions

I once built touch paddles with a few cheap CMOS chips. The working
principle was the conductivity of the skin, and that has inherit
problems. One is that to measure conductivity, you need two points of
contact. So it wasn't enough to touch the plates with thumb or finger,
in addition to that the hand had to rest on a base plate. Moreover, if
my hands were extremely dry, it worked erratically. Luckily normally the
excitement of a CW conversation keeps my palms quite moist! ;)

I have heard about other principles. How exactly do capacitive touch
paddles work? I haven't heard excited reports about it so far on the
list, only a passing mention from time to time. Do capacitive touch
paddles work *really* well for somebody? Please tell me ...

I searched the archives and found a mail from Robert Gobrick (Bob),
VO1DRB/WA6ERB, from 16 Nov 1994:

> I am still working on a project of mounting the Curtis keyer in a little
> case and mounting a pair of $4 Ramsey capacitance-touch paddles and
> circuitry in the same case - that would make it small, light and cheap.
> I have used touch paddles before (have a paddle called the Copperhead
> that was in May 1991 73 magazine and it really does work half way
> decent.

I think the Copperhead design works on the resistive principle, but how
about the Ramsey paddles? Do they work well? - I am not really looking
for a commercial kit, but rather for a good, small, reliable design that
I could reproduce myself. Low power consumption is essential.

In another archive mail from 10 Apr 1995, Joe E., N2CX,
<JEVERHART@cayman.VF.MMC.COM>, wrote:

> Briefly, the "Copperhead" Keyer and a few others like the Ramsey kit are
> not true "capacitive" paddles, they rely on skin conductivity and/or
> stray filed pickup from local electrical lines conucted throught the
> opoerator's body. They can be very very sensitive to changes in
> humidity, skin conductivity ambient electrical fields, etc. Although I
> have used one off and on (no pun intended) for the last couple of years,
> I don't recommend them.
> The design from SPRAT is probably better, but I haven't tried it. The
> best approach I've seen is a digital desing I saw when I worked for the
> Pulsar watch folks in another lifetime. It compared the delay in two
> signal paths one of which had a capacitive touch plate. When the touch
> plate was contacted by a finger, the extra capaictance added to the
> delay in that path which was sensed by digital circuitry. The result was
> a very stable repeatable touch sensor. AND it was tested for no damage
> up to 50 kV. If I get achance, I'll attempt to dig up the circuit and
> share it with the list.

Joe, I'd be very interested to learn more about it. It might be not easy
to reproduce though - I don't know. Can I expect something *simple* to
work repeatable? I'd also be interested in other principles. Someone
once mentioned an optical keyer written up in QST where you tap your
finger between a light beam. A friend of mine suggested tension strips
as something to put some research into.

Hope I didn't put somebody off with such a long mail. The other ones
that I read this afternoon were so much shorter (AutoResponse from


Richard Hieber, DL8MFQ/AA8CP

Search QRP-L Archives

[ QRP-L Archive | ]
[ 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 ]


This archive was generated by hypermail 2b29 on Fri Jun 02 2000 - 11:29:37 EDT