Searching \ for '[PIC] Using N.C. Vibration Switch for Low Power PI' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/power.htm?key=power
Search entire site for: 'Using N.C. Vibration Switch for Low Power PI'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] Using N.C. Vibration Switch for Low Power PI'
2007\03\13@202346 by Brooke Clarke

flavicon
face
Hi:

If I used a Normally Open vibration switch as a PIC input it could be
connected between Vcc and the input.  A high value resistor between the
input and ground would pull down the input and I think the current used
would be minimal.

But if a Normally Closed vibration switch is used is there a circuit
that would not draw current while waiting for vibration?  A coin cell
will be powering this circuit.  This type (2 ball) is much more
sensitive to vibration than the spring over rod type that's N.O. See:
http://www.pacificsites.com/~brooke/Sensors.shtml#Vibration

Have Fun,

Brooke Clarke

--
w/Java http://www.PRC68.com
w/o Java www.pacificsites.com/~brooke/PRC68COM.shtml
http://www.precisionclock.com

2007\03\13@205940 by Gerhard Fiedler

picon face
Brooke Clarke wrote:

> If I used a Normally Open vibration switch as a PIC input it could be
> connected between Vcc and the input.  A high value resistor between the
> input and ground would pull down the input and I think the current used
> would be minimal.
>
> But if a Normally Closed vibration switch is used is there a circuit
> that would not draw current while waiting for vibration?  A coin cell
> will be powering this circuit.  

I think there's no way around it: a NC switch carries current when
connected to a voltage and not active :)

But depending on your timing requirements, you may be able to power the
switch only once in a while (say every 50ms for a few tens of microseconds)
by powering it from a PIC port.

Gerhard

2007\03\13@211948 by David VanHorn

picon face
>
>
> But depending on your timing requirements, you may be able to power the
> switch only once in a while (say every 50ms for a few tens of
> microseconds)
> by powering it from a PIC port.


It's even more fun than that..
There's a thing called "wetting current" that you rarely see specified
anymore.
Below this current level, the switch isn't guaranteed to close, or to stay
closed!

Thereby hangs a tale:

Years ago, I used to do alarm systems for a living.  We had a litton
glassbreak system that was pretty good, but it would false alarm for no
apparent reason.  We had probably 50 systems installed, and they were all
trouble.

The sensor was a gold ring sitting on gold contacts, sealed in a plastic
case.
Sometimes we could identify a switch as bad, but if we took it out, it would
start working just fine again.  After a lot of troubleshooting, I ended up
sitting at various sites with an oscilliscope across the loop.  I would
watch this for hours (no storage!) and after a while, I'd start seeing the
voltage across the loop getting noisy and rising erratically.

After a bit of research, I modified the processor boards for about 10x the
loop current that was originally designed.  BINGO. Problem solved.  I fed
this back to litton, and they incorporated my fix into all later production.
Apparently we weren't the only ones having problems!

So the moral of this story, is that while a microamp might work on the
bench, in real life it may have all sorts of problems.  A closed switch
needs some small amount of current to stay closed.  An open switch can get
it's wetting current in a small pulse, like from a capacitor.
You don't really care (?) if the switch goes open later I'm guessing,
especially if it has to sit for hours undisturbed to get into that state.



--
Jealousy is a disease, love is a healthy condition. The immature mind often
confuses one for the other, or assumes the greater the love, the greater the
jealousy. In fact they are almost incompatible; both at once produce
unbearable turmoil.
Jubal Harshaw, "Stranger in a Strange Land"

2007\03\13@215437 by Denny Esterline

picon face
>
> Hi:
>
> If I used a Normally Open vibration switch as a PIC input it could be
> connected between Vcc and the input.  A high value resistor between the
> input and ground would pull down the input and I think the current used
> would be minimal.
>
> But if a Normally Closed vibration switch is used is there a circuit
> that would not draw current while waiting for vibration?  A coin cell
> will be powering this circuit.  This type (2 ball) is much more
> sensitive to vibration than the spring over rod type that's N.O. See:
> http://www.pacificsites.com/~brooke/Sensors.shtml#Vibration
>
> Have Fun,
>
> Brooke Clarke


http://www.signalquest.com/product_selection.htm

I've used dozens of thier SQ-SEN-200 with a 5.1 Mohm pull up resistor and a
12F629 that sleeps most of the time, years on a pair of AA batteries.

-Denny

2007\03\13@221410 by Gerhard Fiedler

picon face
David VanHorn wrote:

>> But depending on your timing requirements, you may be able to power the
>> switch only once in a while (say every 50ms for a few tens of
>> microseconds) by powering it from a PIC port.
>
> It's even more fun than that.. There's a thing called "wetting current"
> that you rarely see specified anymore. Below this current level, the
> switch isn't guaranteed to close, or to stay closed!

The required wetting current depends a lot on the switch. Gold is good,
mercury (like in reed contacts) is better (I think).

In my experience, the wetting current required for closing is higher than
the one required for staying closed. So you may get by with applying a
higher current for a few milliseconds, then reduce it to a lower current.
(This may or may not work well with switches that stay closed for a looong
time...)

Gerhard

2007\03\13@222541 by David VanHorn

picon face
>
> The required wetting current depends a lot on the switch. Gold is good,
> mercury (like in reed contacts) is better (I think).


True. The mercury was used to get to zero wetting current, since it's a
liquid metal.

In my experience, the wetting current required for closing is higher than
> the one required for staying closed. So you may get by with applying a
> higher current for a few milliseconds, then reduce it to a lower current.
> (This may or may not work well with switches that stay closed for a looong
> time...)


That's what I'm thinking. But I have no data to support it.

--
Jealousy is a disease, love is a healthy condition. The immature mind often
confuses one for the other, or assumes the greater the love, the greater the
jealousy. In fact they are almost incompatible; both at once produce
unbearable turmoil.
Jubal Harshaw, "Stranger in a Strange Land"

2007\03\13@223551 by Jinx

face picon face
part 1 579 bytes content-type:text/plain; (decoded 7bit)

> An open switch can get its wetting current in a small pulse, like
> from a capacitor

Denny, I have an application with a N.O. mercury-free tilt switch
(rolling ball). This works fine. The low value pull-down makes
the circuit pretty immune to false triggering. The only time it has
gone off in error was when an RT was used right next to it. I've
not tried it as N.C. but the current consumption should be low
enough for a coin cell. A 2032 would be ~ 200,000uAh, so best
case might be 14 years with N.C. and a mostly sleeping PIC





part 2 2238 bytes content-type:image/gif; (decode)


part 3 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2007\03\14@085915 by Gerhard Fiedler

picon face
David VanHorn wrote:

>> In my experience, the wetting current required for closing is higher
>> than the one required for staying closed. So you may get by with
>> applying a higher current for a few milliseconds, then reduce it to a
>> lower current. (This may or may not work well with switches that stay
>> closed for a looong time...)
>
> That's what I'm thinking. But I have no data to support it.

Come to think of it... this is about a vibration switch. I don't know how
these are built, mechanically. If the vibration reaches the contact
surfaces, this may reduce the wetting current requirements quite a bit.

Wetting current is mainly needed to overcome/remove/prevent the oxide film
over the contact surfaces. Any even slight vibration probably helps a lot
with this.

Gerhard

2007\03\14@111109 by David VanHorn

picon face
>
>
> Come to think of it... this is about a vibration switch. I don't know how
> these are built, mechanically. If the vibration reaches the contact
> surfaces, this may reduce the wetting current requirements quite a bit.
>
> Wetting current is mainly needed to overcome/remove/prevent the oxide film
> over the contact surfaces. Any even slight vibration probably helps a lot
> with this.


The switches that I had so much trouble with, were vibration switches,
gold-on-gold, and sealed.

2007\03\16@013651 by Denny Esterline

picon face
> > An open switch can get its wetting current in a small pulse, like
> > from a capacitor
>
> Denny, I have an application with a N.O. mercury-free tilt switch
> (rolling ball). This works fine. The low value pull-down makes
> the circuit pretty immune to false triggering. The only time it has
> gone off in error was when an RT was used right next to it. I've
> not tried it as N.C. but the current consumption should be low
> enough for a coin cell. A 2032 would be ~ 200,000uAh, so best
> case might be 14 years with N.C. and a mostly sleeping PIC
>
>

In my application I couldn't be sure of the orientation of the product
at rest. Thus I couldn't be sure if the ball-in-tube switch would be open
or closed most of the time. So a low value resistor was out of the question.

-Denny

2007\03\16@030137 by Jinx

face picon face
> In my application I couldn't be sure of the orientation of the
> product at rest. Thus I couldn't be sure if the ball-in-tube
> switch would be open or closed most of the time. So a low
> value resistor was out of the question

The pull-down's low value is only a concern if it's actually got
a voltage across it. In my configuration the current through the
pull-down is limited by the 3M3, which is in effect the load on
the battery

Even if you didn't use the external interrupt pin as such, an
indeterminate state change could still be detected by interrupt-
on-change

More... (looser matching)
- Last day of these posts
- In 2007 , 2008 only
- Today
- New search...