Simple switching circuit

[[email protected]]

I have designed and built a circuit that consists of an infrared
emitter & detector along with a power transistor. The idea is that
when the beam of infrared light is broken, I want a small 2.5 Amp - 12
volt pump to turn on. I have it working with one problem... when the
beam is partially blocked (very small window), the output voltage is
below 12 volts, and I'm afraid that it could burn up the pump motor.

What I need is either 12 volts or 0 volts. I'm thinking I need
something between the detector and the base of the power transistor to
cause the power transistor to turn completely on, or not at all. Any
help would be appreciated.

The circuit can be viewed at: http://lgman.com/circuit1.jpg

Thanks,
Mick

 

[Jan Panteltje]

I have designed and built a circuit that consists of an infrared
emitter & detector along with a power transistor. The idea is that
when the beam of infrared light is broken, I want a small 2.5 Amp - 12
volt pump to turn on. I have it working with one problem... when the
beam is partially blocked (very small window), the output voltage is
below 12 volts, and I'm afraid that it could burn up the pump motor.

What I need is either 12 volts or 0 volts. I'm thinking I need
something between the detector and the base of the power transistor to
cause the power transistor to turn completely on, or not at all. Any
help would be appreciated.

The circuit can be viewed at: http://lgman.com/circuit1.jpg

Thanks,
Mick

Problem with circuits like this is, that if you make it more sensitive
it will also start reacting to other light (environment).
Perhaps you could add some posivive feedback, to make it flip over.

------------------------------------------------------
| | | |
R1 3k3 R3 10k e |
| |-----R4----- b T2 |
| 10k c c PNP |
|----R2------- b T1 | |
| | e NPN | 100 c
photo | | |---R7----b T3
transistor ----- | -----R5 ------| e
| | see text | |
| | | |
| | R6 load
| | | 2k2 |
/// /// /// ///

When light falls on the photo transistor, then T1 will conduct,
and so will T2.
T2 will make T1 conduct a bit more via R5, R5 should be a high
value, maybe 10 to 100x R2..
The idea is: Schmitt trigger
http://en.wikipedia.org/wiki/Schmitt_trigger

 

[John Devereux]

I have designed and built a circuit that consists of an infrared
emitter & detector along with a power transistor. The idea is that
when the beam of infrared light is broken, I want a small 2.5 Amp - 12
volt pump to turn on. I have it working with one problem... when the
beam is partially blocked (very small window), the output voltage is
below 12 volts, and I'm afraid that it could burn up the pump motor.

Or the transistor, more likely.

What I need is either 12 volts or 0 volts. I'm thinking I need
something between the detector and the base of the power transistor to
cause the power transistor to turn completely on, or not at all. Any
help would be appreciated.

Hi Mick,

Your thinking is correct - you need a comparator with hysteresis.

You should also have a reverse connected power diode accross the pump
terminals. Otherwise the transistor is likely to be destroyed by the
turn-off spike from the motor inductance.

Someone here may sketch up a circuit for you, but you could also
consider a relay as a quick solution. (This will need the diode
accross its coil, too).

 

[Fred Bloggs]

I have designed and built a circuit that consists of an infrared
emitter & detector along with a power transistor. The idea is that
when the beam of infrared light is broken, I want a small 2.5 Amp - 12
volt pump to turn on. I have it working with one problem... when the
beam is partially blocked (very small window), the output voltage is
below 12 volts, and I'm afraid that it could burn up the pump motor.

That is not a good circuit, the transistor does not saturate and will
end up dissipating quite a bit of power. That 3.3K base transistor is
not nearly enough drive for a 2.5A load, the output will current limit
and the motor will not spin up. Assuming a motor resistance of about 1R,
then a typical HFE of 100 makes the transistor look like 33 ohms in
series with 12V-VBE for a stalled motor voltage of 11.2/34=0.3V and
transistor power dissipation of 3.6W. That is not going to work very
well. Your test load on the schematic is not representative of anything.
You need to figure out a way to inject about 100mA into that base
terminal before you do anything else. This will be left as an exercise.

 

[Jan Panteltje]

Problem with circuits like this is, that if you make it more sensitive
it will also start reacting to other light (environment).
Perhaps you could add some posivive feedback, to make it flip over.

------------------------------------------------------
| | | |
R1 3k3 R3 10k e |
| |-----R4----- b T2 |
| 10k c c PNP |
|----R2------- b T1 | |
| | e NPN | 100 c
photo | | |---R7----b T3
transistor ----- | -----R5 ------| e
| | see text | |
| | | |
| | R6 load
| | | 2k2 |
/// /// /// ///

When light falls on the photo transistor, then T1 will conduct,
and so will T2.
T2 will make T1 conduct a bit more via R5, R5 should be a high
value, maybe 10 to 100x R2..

PS
You may want to add 1k from base T1 to ground.
This will put the switch point at about 7 V ( (1 / 11) * 0.7)

 

[[email protected]]

Hi all,

Thanks for the suggestions. I've added a Schmitt Trigger, increased
the current at the base of the output transistor, and added a diode to
protect the output transistor. Any comments?

Thanks,
Mick

 

[[email protected]]

Whoops, set the new design at http://lgman.com/circuit3.jpg

 

[Fred Bloggs]

Whoops, set the new design at http://lgman.com/circuit3.jpg

That's not nearly enough drive, and the switching levels are off. You
need something BIG. Most switching applications using transistors place
the load in the collector circuit, something like this, and an LM555 is
the most convenient way to realize a Schmitt with maximum hysteresis:
View in a fixed-width font such as Courier.

..
..
..
.. +12V
.. |
.. .-------+---------+---------------+------+---------.
.. | | | | | |
.. | | | | | |
.. | | | | [10,1/2W] |
.. | |/ | 555 47,000u|+ | |<
.. v ~~ | ---+--- === +-------| D45V10
.. - |> | | | | |\
.. | | | out|----. | [15R,10W] |
.. | | | | | | | |
.. | +-+---|trig | [120,1W] | | +----> To motor
.. | | | | | | | |/ |
.. | | '---|thresh | +-----------| D44V10 -
.. | | | | | | |> ^ 3A
.. [Rc] [3.3k] | | [220] | | | diode
.. | | ------- | | | |
.. | | | | | | |
.. | | | | | | |
.. '-------+---------+--------+------+------+---------'
.. |
.. ---
.. ///

 

[Fred Bloggs]

Whoops, set the new design at http://lgman.com/circuit3.jpg

This one configures the 555 a little bit better:
View in a fixed-width font such as Courier.

..
..
..
.. +12V
.. |
.. .-----+-----------+---------------+------+---------.
.. | | | | | |
.. | | | | | |
.. | | .--+ | [10,1/2W] |
.. | |/ 4| 8| 555 47,000u|+ | |<
.. v ~~ | ----+--- === +-------| D45V10
.. - |> |rst V+ |3 | | |\
.. | | 6| out|----. | [15R,10W] |
.. | +----+-|thresh | | | | |
.. | | |2| | [120,1W] | | +----> To motor
.. | | '-|trig |7 | | |/ |
.. | | 5| dis|----+-----------| D44V10 -
.. | | .-|ctl | | | |> ^ 3A
.. | | 0.1| | gnd | | | | | diode
.. | | === -------- | | | |
.. [Rc] [3.3k] | |1 [220] | | |
.. | | | | | | | |
.. '-----+----+------+--------+------+------+---------'
.. |
.. ---
.. ///