A father / daugther project :-) simple door monitor

[Markus Zingg]

Hi Group

My dougther showes interst in my electronic hobby. She's 10 years old,
and we setteled on creating a little monitor device for her, showing
the open/close state of several doors and windows (we would like to
end up with a total of 16 switches) in our home. I have tons of 0,2mm
copper wire here from old transformers and we intend to use this for
the wireing between the monitor appliance and the switches. I also do
have plenty of LEDs and other stuff (including the hex inverter
schmitt triggers mentioned later) and obviousely I would like to use
what I have here already.

I figure that the partially long thin wires from the monitoring
appliance to the switch and back will see a varying drop in the
voltage, so I figured that driving the leds this way directly is a bad
idea. I therefore thought to use hex inverter schmitt triggers to have
a clean signal to drive the leds and also to be able to define
individually per door/window wether open should light the led or vice
versa.

I started to capture a little schema which you can see here:

http://mypage.bluewin.ch/mzingg/doormon.htm

The idea should be to input the wire from a switch into a gate (hex
inverter schmitt trigger) and by jumpering the output of a given gate
apropriately either drive another gate (to invert the signal once
more) or directly drive the led. The pulldown resistors at the input
of the second gate are meant to avoid a floting input in case the
"user" defines to use the inverted signal alas not using the second
gate.

JP1 shall get the returning wires from the switches, JP2 is here to
have enough connectors to power the wires to the switches.

I think this should work, but I don't have enough experience with
stuff like this and thought I ask before I burn more time. Then, I
also would like to know if this works so as I do not start to explain
all this to her ending up with something not working....

TIA

Markus

 

[Arlet]

I figure that the partially long thin wires from the monitoring
appliance to the switch and back will see a varying drop in the
voltage, so I figured that driving the leds this way directly is a bad
idea.

For teaching purposes it wouldn't be such a bad idea to start with
simple switches, a long thin wire, and a series resistor, and see what
happens. If you have a volt/ohmmeter, you can even measure the
resistance and/or voltage drop.

 

[Rich Grise]

end up with a total of 16 switches) in our home. I have tons of 0,2mm
copper wire here from old transformers and we intend to use this for
the wireing between the monitor appliance and the switches. I also do
have plenty of LEDs and other stuff (including the hex inverter
schmitt triggers mentioned later) and obviousely I would like to use
what I have here already.

I figure that the partially long thin wires from the monitoring
appliance to the switch and back will see a varying drop in the
voltage, so I figured that driving the leds this way directly is a bad
idea.

Not necessarily - LEDs are currnet operated - as long as your wire is
good for 10-20 mA, then all you have to do is adjust the current-limiting
series resistor to accomodate the series drop in the wire.

http://mypage.bluewin.ch/mzingg/doormon.htm

Looks pretty good - AWFULLY good for somebody who claims to be a beginner! ;-)

But one piece of advice - there should be pullups or pulldowns on inputs
that might float: a floating CMOS input screws things up - indeterminate
state, linear region, power consumption, oscillation, etc, etc, etc...

Cheers!
Rich

 

[Bill Bowden]

Hi Group

My dougther showes interst in my electronic hobby. She's 10 years old,
and we setteled on creating a little monitor device for her, showing
the open/close state of several doors and windows (we would like to
end up with a total of 16 switches) in our home. I have tons of 0,2mm
copper wire here from old transformers and we intend to use this for
the wireing between the monitor appliance and the switches. I also do
have plenty of LEDs and other stuff (including the hex inverter
schmitt triggers mentioned later) and obviousely I would like to use
what I have here already.

I figure that the partially long thin wires from the monitoring
appliance to the switch and back will see a varying drop in the
voltage, so I figured that driving the leds this way directly is a bad
idea. I therefore thought to use hex inverter schmitt triggers to have
a clean signal to drive the leds and also to be able to define
individually per door/window wether open should light the led or vice
versa.

I started to capture a little schema which you can see here:

http://mypage.bluewin.ch/mzingg/doormon.htm

The idea should be to input the wire from a switch into a gate (hex
inverter schmitt trigger) and by jumpering the output of a given gate
apropriately either drive another gate (to invert the signal once
more) or directly drive the led. The pulldown resistors at the input
of the second gate are meant to avoid a floting input in case the
"user" defines to use the inverted signal alas not using the second
gate.

If you hardwire the output of one inverter to the input of the second,
you will have both high and low signals available for the LED. Then
just use one jumper to select the desired state, inverted or
non-inverted. That will reduce your parts count somewhat.

Also be sure every input to the 74HCT14 is terminated with some
resistance to ground to avoid static and oscillation problems. 100K
resistors should work.

-Bill

 

[Markus Zingg]

Hi Bill

If you hardwire the output of one inverter to the input of the second,
you will have both high and low signals available for the LED. Then
just use one jumper to select the desired state, inverted or
non-inverted. That will reduce your parts count somewhat.

Silly me! Of course, that's a very good idea. I did not liked the two
jumpers and your aproach not only saves me parts, it makes the whole
thing less error prone also with regard to invalid jumper settings
which are now actually not possible.

Also be sure every input to the 74HCT14 is terminated with some
resistance to ground to avoid static and oscillation problems. 100K
resistors should work.

Will do. Thanks again.

Markus

 

[ehsjr]

Hi Group

My dougther showes interst in my electronic hobby. She's 10 years old,
and we setteled on creating a little monitor device for her, showing
the open/close state of several doors and windows (we would like to
end up with a total of 16 switches) in our home. I have tons of 0,2mm
copper wire here from old transformers and we intend to use this for
the wireing between the monitor appliance and the switches. I also do
have plenty of LEDs and other stuff (including the hex inverter
schmitt triggers mentioned later) and obviousely I would like to use
what I have here already.

I figure that the partially long thin wires from the monitoring
appliance to the switch and back will see a varying drop in the
voltage, so I figured that driving the leds this way directly is a bad
idea. I therefore thought to use hex inverter schmitt triggers to have
a clean signal to drive the leds and also to be able to define
individually per door/window wether open should light the led or vice
versa.

I started to capture a little schema which you can see here:

http://mypage.bluewin.ch/mzingg/doormon.htm

The idea should be to input the wire from a switch into a gate (hex
inverter schmitt trigger) and by jumpering the output of a given gate
apropriately either drive another gate (to invert the signal once
more) or directly drive the led. The pulldown resistors at the input
of the second gate are meant to avoid a floting input in case the
"user" defines to use the inverted signal alas not using the second
gate.

JP1 shall get the returning wires from the switches, JP2 is here to
have enough connectors to power the wires to the switches.

I think this should work, but I don't have enough experience with
stuff like this and thought I ask before I burn more time. Then, I
also would like to know if this works so as I do not start to explain
all this to her ending up with something not working....

TIA

Markus

I think you are making it far more complex than it
needs to be. The voltage drop in the wires to the
doors/windows should be such a small percentage
of the total drop due to the limiting resistor
that it can be ignored. You'll have about 17 ohms
in 100 feet of wire. For example, if you run 12 volts
out to the door/window switches, and use 1K limiting
resistors for the LEDs, you'll get about 10 mA. That
translates to a voltage drop in 100 feet of wire of
..017 volts. You may have more variation in the combination
of the limiting resistance tolerance and the Vf variation
in the LEDS, than variation caused by the different lengths
of .2 mm wire.

Aside from that, if you do decide to go with the more
complex circuit, I don't think the idea of making
each LED independently switchable to indicate closed
or open status is good. I think they should be consistent;
that is, if led lit = open that should apply to all
leds or if led off = open that should apply to all leds.

Ed

 

[Markus Zingg]

Hi Ed

I think you are making it far more complex than it
needs to be. The voltage drop in the wires to the
doors/windows should be such a small percentage
of the total drop due to the limiting resistor
that it can be ignored. You'll have about 17 ohms
in 100 feet of wire. For example, if you run 12 volts
out to the door/window switches, and use 1K limiting
resistors for the LEDs, you'll get about 10 mA. That
translates to a voltage drop in 100 feet of wire of
.017 volts. You may have more variation in the combination
of the limiting resistance tolerance and the Vf variation
in the LEDS, than variation caused by the different lengths
of .2 mm wire.

Thanks for pointing this out. Some wires are likely to be in the 300
feets (or even more) range. If I understand you right, this would
tripple the voltage variation to about half a volt versus say nearly
nothing with shorter wires with switches located more "nearby". I just
intend to end up with an appliance where the connections can be easily
interchanged later, alas where I don't have to adjust resistors to
have a similar birghness of the LEDs.

Aside from that, if you do decide to go with the more
complex circuit, I don't think the idea of making
each LED independently switchable to indicate closed
or open status is good. I think they should be consistent;
that is, if led lit = open that should apply to all
leds or if led off = open that should apply to all leds.

You have a point here. However, the reason to have it this way comes
from the fact that we most likely will have different switches for
windows and/or doors based on their different construction. We will
make simple switches ourselfes and it's handy being able to implement
them as closers or openers so as they fit a given door/window
construction best.

So, with the jumpers we actually have a tool in our hands to make the
display consistant with the logical state of the doors/windows
independant of the kind of switches used.

Markus

 

[Rheilly Phoull]

Hi Ed


Thanks for pointing this out. Some wires are likely to be in the 300
feets (or even more) range. If I understand you right, this would
tripple the voltage variation to about half a volt versus say nearly
nothing with shorter wires with switches located more "nearby". I just
intend to end up with an appliance where the connections can be easily
interchanged later, alas where I don't have to adjust resistors to
have a similar birghness of the LEDs.


You have a point here. However, the reason to have it this way comes
from the fact that we most likely will have different switches for
windows and/or doors based on their different construction. We will
make simple switches ourselfes and it's handy being able to implement
them as closers or openers so as they fit a given door/window
construction best.

So, with the jumpers we actually have a tool in our hands to make the
display consistant with the logical state of the doors/windows
independant of the kind of switches used.

Markus

Why not run the wiring and experiment with the various combinations of
current control ??
For my money the simple resistor and LED's should be OK.

 

[Rich Grise]

intend to end up with an appliance where the connections can be easily
interchanged later, alas where I don't have to adjust resistors to
have a similar birghness of the LEDs.

Then just associate a particular resistor with each LED - the wire
length isn't going to change, is it?

Or, if you want to get fancy-schmancy, you could use current drivers
with voltage compliance all the way from almost Vcc to a volt or
so above ground:


+ Vcc +Vcc
| |
| |
[R1] [LED]
| |
| |
| b / c
+--------------| NPN
| < e
| |
[1N4148] |
K | |
| [R2] 68R for ~10 mA
[1N4148] |
K | |
| |
GND GND

R1 is non-critical - just keep the diodes and e-b junction conducting
through the dynamic range, which shouldn't be very far, for LEDs.

This way, the voltage compliance at the collector makes up for differences
in the wire resistance.

You also might want to watch your power dissipation if you're running a
high Vcc, but at 10 mA it shouldn't be too much of a problem. (the
transistor dissipates the heat that the dropping resistor would, the
difference is that the transistor is constant-current regardless of
Vcc, up to a point. )

Have fun!
Rich

 

[Bill Bowden]

intend to end up with an appliance where the connections can be easily
interchanged later, alas where I don't have to adjust resistors to
have a similar birghness of the LEDs.

Then just associate a particular resistor with each LED - the wire
length isn't going to change, is it?

Or, if you want to get fancy-schmancy, you could use current drivers
with voltage compliance all the way from almost Vcc to a volt or
so above ground:


+ Vcc +Vcc
| |
| |
[R1] [LED]
| |
| |
| b / c
+--------------| NPN
| < e
| |
[1N4148] |
K | |
| [R2] 68R for ~10 mA
[1N4148] |
K | |
| |
GND GND

R1 is non-critical - just keep the diodes and e-b junction conducting
through the dynamic range, which shouldn't be very far, for LEDs.

This way, the voltage compliance at the collector makes up for differences
in the wire resistance.

You also might want to watch your power dissipation if you're running a
high Vcc, but at 10 mA it shouldn't be too much of a problem. (the
transistor dissipates the heat that the dropping resistor would, the
difference is that the transistor is constant-current regardless of
Vcc, up to a point. )

Have fun!
Rich

Or, you could substitute a single red LED in place of the 2 diodes and
eliminate one more part.

-Bill

 

[ehsjr]

Hi Ed




Thanks for pointing this out. Some wires are likely to be in the 300
feets (or even more) range. If I understand you right, this would
tripple the voltage variation to about half a volt versus say nearly
nothing with shorter wires with switches located more "nearby". I just
intend to end up with an appliance where the connections can be easily
interchanged later, alas where I don't have to adjust resistors to
have a similar birghness of the LEDs.




You have a point here. However, the reason to have it this way comes
from the fact that we most likely will have different switches for
windows and/or doors based on their different construction. We will
make simple switches ourselfes and it's handy being able to implement
them as closers or openers so as they fit a given door/window
construction best.

So, with the jumpers we actually have a tool in our hands to make the
display consistant with the logical state of the doors/windows
independant of the kind of switches used.

Markus

Aha - your last paragraph explains. Nice idea. As to the
length of the wire - yes, running 300 feet with a 10 mA
LED current would give you about a 1/2 volt drop. That may,
or may not, cause discernable difference. It will certainly
be minor. With 300 feet of wire at 17 ohms per 100 feet,
with a 1K limiting resistor, you'll have 1051 ohms resistance.
At 12 volts, that will translate to ~.0097 mA through the LED,
figuring 1.8 Vf. If the wires were very short, the resistance
would be negligible, so you'd have 1000 ohms resistance.
That translates to .0102 mA - a difference of .0005 mA.

Getting back to the homemade switches: you can make either
kind of switch, and mix normally open and normally closed,
and still make lit = open (or lit = closed) with either a
normally closed or normally open switch and your jumper idea,
without the need for the inverter. You just change where the
LED plugs in to the header. Drawn below is lit = open

+12 -------[1K]---+---[LED]---+---Gnd
| |
| / |
+---o o----+ Normally Open

When door opens, switch opens, removing short across LED


+12 -------[1K]---+ +---[LED]---Gnd
| |
| |
+---o--o----+ Normally Closed

When door opens, switch closes, completing circuit to LED.

You can also do it vice-versa where lit = closed - just
use the same diagram and swap the switches.

Ed