> -----Original Message-----
> From: Jon Ogden
> Here is where I am now. I have successfully eliminated the interference
> on
> all amateur bands from my lighting system when it is uninstalled.
>
> The current unit I have been messing with gets mounted above a fan hood.
> The hood is basically a big stainless steel box (simplest way to describe
> it). The lighting circuitry is mounted above it. The sensing line is
> then
> attached to the metal hood. This way whenever the hood is touched, the
> lighting system is activated. With the sensing line attached to the hood,
> I
> get my 20m interference problem again. The hood is acting like a big
> antenna. The hood is not grounded but isolated from DC ground since
> grounding the hood when then ground the sense line and the switching would
> not work.
>
> I've tried adding a .01 uF cap (and other values too) from the sensing
> lead
> to ground. So far no matter what value I have tried, adding the cap ends
> up
> loading the oscillator in the dimmer unit to the point that it is no
> longer
> sensitive to touch. I've also tried winding the sensing line around a
> torroid. If I get too many wraps, I again destroy the sensitivity of the
> sensing line and it no longer does it's job. And anyhow, it still doesn't
> stop my RF problem. I've tried resistors in series with the sensing line
> to
> the hood. Above a few 100 Ohms and I lose my sensitivity just like with
> the
> torroid.
>
Have you tried a "micro" Henry choke on the sense line? Something
from 1-10 micro Henries is what is recommended. I don't think the resistor
is really needed since this is a very low current application (high current
would be bad for the human contact!). This is a very small inductance
(don't use the Radio Shack 100milliH choke!) and should not be much
reactance to the free running oscillator's sense line.
Assuming the oscillator is around 170 Khz like most touch lamps (see
if you can find out), that would be only 11 ohms inductive reactance (for 10
microH) while providing 880 ohms at 14 Mhz; a factor of 80. Don't use 10
milliH since the inductive reactance to the sensing circuit at 170Khz would
be almost 11Kohm and probably would shut it down. If 10 microH is too much
try 1 microH. That'd be only 1.1 ohms at 170Khz (assuming that is the
oscillator's freq!?). It may take experimenting with several values to get
it close. Also mount the inductor as close as possible to the circuit.
> So far, I have found no good way to RF bypass the sensing line. What I
> want
> to do is basically shunt all HF energy to ground on that line, but I can't
> load down the line otherwise the sensing circuitry won't help.
>
Perhaps you could use a capacitor to pass the 14Mhz on the hood to
ground while, at the right value could look like a higher reactance to the
sensing line's capacitance. Cap of 0.01 microF would give 1.1 ohms at 14
Mhz & 93 ohms at 170Khz. So connect the cap to the hood somewhere or at a
couple of places, then gorund the other end of the cap(s).
The problem is that the sensing line is operating on a
"Change-of-Capacitance" to change the operating state. The 14Mhz bypass to
ground might desense the "sensing line" enough to kill it. I think that's
what you've found already. So we're back to getting the RF choke right.
This all depends alot on the free running oscillator's operating frequency.
Is it spec'ed anywhere?
73,
de ed -K0iL
--
FAQ on WWW: http://www.contesting.com/FAQ/rfi
Submissions: rfi@contesting.com
Administrative requests: rfi-REQUEST@contesting.com
Questions: owner-rfi@contesting.com
|