ESD protection for microcontroller pin that is to be touched

[Nico Coesel]

I have read it so many times that my eyes starting hurting ;-).




I know, it is a capacitive system. However, what I wanted to make after
giving up on Qtouch, is a sensor that /is/ to be touched.

Why would you do that? Capacitive sensing isn't that hard to get right
with the chips available. There must be something else you are doing
wrong. Maybe switch to a different brand. From what I've heard the
capacitive touch chips from Cypress are the best. This chip was highly
recommended:
http://www.cypress.com/?id=3650

I never got round to testing it myself because the client choose to
use a touchscreen + TFT.

 

[Tim Williams]

The BAV199 is very low leakage and "budget-priced".

Indeed, though the voltage drop is kind of high. A subsequent series
resistor will easily cushion the leftovers against the uC's input diodes.

Tim

 

[Joerg]

Indeed, though the voltage drop is kind of high. A subsequent series
resistor will easily cushion the leftovers against the uC's input diodes.

That's the trick, the "subsequent resistor". Then it won't matter
whether the diode drops 1.5V or 2V.

 

[Robert Macy]

With a full plane you can live with a few sprinkled 0.1uF plus on 10uF
or so. Because the plane itself is a small and very good distributed
capacitor.

I know I'm 'preaching to the choir' but the plane is a capacitor at
low enough frequency. Higher frequencies it's actually more like a
transmission line whose characteristic impedance starts out as a
function of plane separation [that's why thinner is better, because
this Zo is THE limiting factor] It's like transmission lines in 3D, in
parallel, so difficult to envision. At high frequency there's not much
area involved so the capacitance is low, then at lower frequency as
the wave spreads, there is more area and there is more capaictance.
Think like a transmission line whose Zo drops as the wave spreads,
kind of like Zo(f) When f is high, Zo goes to some value, then when f
is low, the Zo is dropping like a rock 10 ohm, 1 ohm, 0.1 ohm etc
depending on size of the board. The trick with caps is to place them
judiciously so that the board planes 'look' like they continue out to
infinity.

From reading your posts, I thinkyou know this stuff already. Has
anybody calculated the Zo(f) for PCB planes?

If you can get a copy of the AppNote from Ansoft describing using
their HFSS to reduce number of bypass caps it's worth reading. Going
from something like 60 caps down to 12 and placing them properly [may
have misremembered] substantially IMPROVED the performance of the PCB
by reducing noise on the planes.

 

[Rene]

Aha, thanks!

With a full plane you can live with a few sprinkled 0.1uF plus on 10uF
or so. Because the plane itself is a small and very good distributed
capacitor.




That was in the last century

Rene, plus one very important thing in addition: If you get a barrage of
hits and your system is extremely low in power consumption you might
want to have a limiter that bleeds excess charge. This is because
regulators can typically only source current but most (except for some
sync bucks) cannot sink. A TLV431 is nice for this, just set it a few
hundred mV above your nominal rail voltage. High enough that tolerances
won't make it come on but low enough so it does come on before chips go
kaputt.

Thanks to you as well. I have a TL431, I think I will put that into the
device.

Things are starting to sound rather difficult, e.g. in Roberts
analysis... I suddenly remember why I gave up electronic engineering a
long time ago and switched to technical software engineering ;-), I do
not like all these physics, why can't components not just behave as
ideal components? Again ;-) (but I *am* glad that I usually do not have
with these matters - and all the more gratefull for the assistance of
the members of this group).

Thanks again to all of you!

Yours sincerely,
Rene