Subject:
|
Re: multiplexed touch sensors [was: Re ideas for a RIS 2.0]
|
Newsgroups:
|
lugnet.robotics
|
Date:
|
Wed, 25 Aug 1999 18:06:19 GMT
|
Viewed:
|
837 times
|
| |
|
|
John A. Tamplin <jat@liveonthenet.com> wrote:
> > If that were true, then the following multi-touch-sensor adapters would not
> > work:
> >
> > http://www.lugnet.com/robotics/?n=1897
> >
> > http://www.plazaearth.com/usr/gasperi/3rmux.htm
>
> Put a touch sensor port into raw mode and watch the values you get as you
> gradually depress the sensor. There is not a sharp cutoff like there is
> with a typical contact switch.
>
> I tried building a touch sensor multiplexor, but I was unable to reliably
> get even 3 to work because the minimum resistance is quite large and
> because the transition from infinite to that minimum resistance is quite
> gradual. By switching to micro-contact switches, I was able to reliably
> get 4 switches to work, including operating with a wide range of battery
> freshness.
Having revisited this entire sub-thread, it seems people started rambling
on (constructively, I think) about switch times, debouncing, etc.
Getting back to multiplexed touch sensors, I bet it's possible to get the
Lego touch sensors to work fine with custom firmware and some sensible and
reasonably simple debouncing logic to throw away readings during
transitions. The caveat occurs in the hopefully special case where
multiple buttons are being pressed and released all at the same time;
readings might never settle long enough to determine the state of the
buttons.
If micro-contact switches work, then I would suppose it is because their
switch time is substantially less than your sensor sampling rate, thereby
making more complicated debouncing unnecessary. I mention this to contrast
with the Lego sensors. When you tried the Lego sensors, what was your
sample rate? Was it the ROM/firmware 3 ms period / 330 Hz rate, or was the
LegOS .1 ms period / 10 kHz max ADC rate? The LegOS rate is insanely high,
and could be a source of problems. I believe 3 ms ought to be sufficiently
slow for many cases, but of course, I have not actually tried to multiplex
sensors, so maybe 3 ms is still too fast for multiple sensors, and you need
to drop an even lower rate or use a more complicated algorithm to get
reasonable results.
-Kekoa
|
|
Message has 2 Replies:
Message is in Reply To:
 | | Re: ideas for a RIS 2.0
|
| (...) gradually depress the sensor. There is not a sharp cutoff like there is with a typical contact switch. I tried building a touch sensor multiplexor, but I was unable to reliably get even 3 to work because the minimum resistance is quite large (...) (25 years ago, 24-Aug-99, to lugnet.robotics)
|
22 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
This Message and its Replies on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
