Value Line '430 Cap Touch I/O - Max. Cap Drive Question

I don't think there is a limit. However, the oscillation gets slower, the larger the capacitance is.
And the capacitance change you can achieve is relatively low. So if your base capacitance is 1nF and the difference between touch and no touch is only 1pF, you'll only have 1 0.1% change in capacitance (and frequency). Which will well be in the noise range.

Best is if your capacitance change is roughly in the range of the base capacitance. If th ebase capacitance is much smaller, even changes in the air moisture may be detected as a touch, whiel if it is much bigger, the frequency change is so small, that you canntot tell it from noise or jitter.

Frequency change is relative to capacitance change, while a resistor change shifts both frequencies and therefore does only affect the speed at which you'll get a sufficient number of oscillations to work with (or for a given time interval, the granularity) but does not influence the relative frequency change caused by the touch.

Jens-Michael,

Thanks for the response - so do you think the capacitive sense capability (Pin Osc) is completely decoupled from the Pin I/O max. capacitance rating?  If so, that is good - but I thought there would be some practical limits - i.e. - if I tried to measure a 1F supercap would that work?  I suppose the RC time constant could still be held fairly easily to be sub 1-second.

Thanks,
John W. 

johnw said:

if I tried to measure a 1F supercap would that work?

The practical limit is the timer. If one oscillation takes longer than your measuring interval (which i slimited by your capability to measure the passed time), then you won't get a result.

Also, the higher the capacitance (and therefore the oscillation period), the more do things like self-discharge, input trigger jitter etc. affect the result. Also the port pin driving capability has some limits.
And then, supercaps have some battery-like charging/discharging behaviour and do not follow the normal RC formula. It takes minutes to fully charge a supercap (99.9%) while with appropriate current driving capability (which a power supply surely has) it should be done in seconds. R*C*ln(2^10), where C=1(F) and ln(2^10) is ~7, so if the supply has an impedance of 1Ohm it should be charged in <7 seconds from 0V to VCC*0.999 if it were a plain capacitor.

Jens-Michael,

Yes - agreed on the issues of the supercap - just meant to be an extreme example.

On a related question:

Does anyone know what the capacitance design value is for the new '5529 experimenters boards with the 100K resistors and capacitive pads?

Thanks!
johnw