Will this simple photocell circuit work?

I want to create a device that will detect when a certain LED lights up on my office phone; I want the device to then turn on a much, much brighter light that I will easily notice ….. This is what I'm hoping will work: I'll place the photocell (shown here: http://www.futurlec.com/Pictures/PHOTOCELL2A.jpg) above the phone's LED, in a surrounding "pocket" designed so that absolutely no ambient light could get to the photocell – only the LED's light will get to it. One lead of the photocell will go to the 9-volt battery's negative terminal (a battery of a different voltage could easily be used). The other lead from the photocell will be connected to a lead from a minilamp. The other lead from the minilamp will be connected to the battery's positive terminal …… It seems to me that this simple circuit should work. When the photocell is dark, it completely resists current, meaning that the minilamp cannot light up. When the photocell detects the LED's light, its resistance almost completely disappears, meaning that the minilamp's circuit is now complete and it lights up. Have I figured correctly? Thanks for any and all help.

Hello. Unfortunately this is unlikely to work on its own, in spite of your theory being correct in that the resistance of a Light dependent resistor (LDR) not by enough. You will need a single stage transistor amplifier for this to work. The mini lamp you mention, is it an ultra bright LED, or a type lamp? The amplifier will have an extra resistor for a LED load. I will attach a circuit when you reply. (I need to draw it on my other computer!). Cheers.

I think the Photo cell-system will work, but you'll need a different circuit. Most photo cells only lower the resistance down to a few hundred ohms, which may not be enough to let the lamp turn on. I would look for a photo cell light-detection circuit.

I doubt if the LED on the phone will be bright enough to lower the photocell resistance enough to light the secondary lamp. Colin and Mark are correct, a transistor is needed to boost the current.

Could any of you fine folks sketch out a diagram (using English words, not symbols ) what this circuit would look like. I know very little about electronics; could you give me any specifics of what type/model transistor I would need? ............. Rodney, you asked what kind of minilamp I would use; I'm figuring a plain old boring incandescent bulb like in flashlights that were common in the 70's or 80's .......... if anyone wants to email me, it's mrmud(_at_)msn.com. ................ thanks again, to all.

Hi,
Here is a circuit I made a few minutes ago. You could even use a MOSFET. Anyway, you can look on the net for the transistor connections, and although you are not too familiar with electronics, you may ultimately like to see a schematic, which I ahve attached here. It is not very difficult, and you will learn something from it and your friends here.

Try this (guessed at the resistor values -- hope it works)

BTW: the transistor is a PN2222A, the resistor on the left (yellow, violet, orange) is a 47K and the resistor on the right (yellow, violet, brown) is a 470Ω

If the LED stays on even when the phone light is off, then decrease the value of the 47K resistor. If the LED won't come on, then increase it. You could also replace it with a 500K variable resistor (e.g. trim pot) and then vary it until the LED comes on when the Phone light is on, and goes off when the phone light is off.

Hi George
That photocell is a cadmium sulphide light sensitive resistor type, by the looks of it. The resistance is 160 ohm in strong light and 160K ohm in the dark. You want it to detect an LED light, which although can be bright, does not emit much EMR becuase LED's are efficient. They run cold and do not emanate any heat with the light. The photocell may only get down to 5K ohm, say.
You can see that using, say, a 9V supply and the photocell in series with a 5K ohm current limiting resistor, the current flow will be very low, about 0.1 mA. To light a bulb you will need over 100 mA typically.
So you need a single transistor amplifier (BC548), or an Op Amp (741).
Go for it - design, build, test .....

It may work but its life will be less. The reason is LDR is meant for 5 to 10 mA current, but lamp needs large current in the order of 100 to 150 mA, so LDR heats up at this current and gets damaged. If LDR resistance is large for LED light lamp may not glow. I suggest to connect base of transistor SL 100 to LDR one end, other end of LDR to + Ve. DC supply. Emitter of SL 100 to - Ve. DC supply and collector to 6V lamp and other end of lamp to +Ve DC supply. DC supply must be minimum of 6 or 9V. This circuit may work better even if LDR is placed little away from telephone set.

Are you aware that you can use led’s for light detection?

Might pay to read some of these articles

I know you’re a beginner in electronics but apart from light to give you indications why not use an audible oscillator such as provided by a 555 timer for example or a simple gated 4000 cmos oscillator

See the following for more info

http://laser.physics.sunysb.edu/~tanya/report2/
or

http://www.edn.com/design/power-management/4313423/Use-an-LED-to-sense-and-emit-light

or

http://www.instructables.com/id/Light-Sensing-LEDs/step3/How-Can-A-LED-Sense-Light/

or

http://blog.makezine.com/2009/08/05/ask-make-led-as-light-sensor/

How else could you achieve this? Micro perhaps, using adc

Or using an opamp circuit window comparator circuit

Have a look at redrocks website concerned with solar power

http://www.redrok.com/led3xassm.htm#led3xc3

Good explanations on ledfs as light sensors here as well

http://en.wikipedia.org/wiki/LED_circuit#LED_as_light_sensor

And if you want some really interesting articles try this one as well Very interesting article all about using leds to sense light with very good explanations

http://www.thebox.myzen.co.uk/Workshop/LED_Sensing.html

I wonder what else you could applythe above link too

Gosh the minds starting to work here, starting to become a little adventurous

Touch sensitive pads through glass plating perhaps!!

A light sensitive code door entry system using led properties ? That’s got you thinking hasn’t it ?

To all the very helpful guys who responded here: Thanks so much ..................... I'm going to have to take a little time to sort through it all. I ordered the photocell in question from SparkFun.com. It hasn't arrived yet, but when it does I'll use the suggestions I've gotten from you all. And I'll report on the results ............ I'm looking forward to getting that bugger working!

Correct Collin and whats a photo transistor in effect practically but not quite exactly an LED used as a sensor instead hence 600mv drop or 600R re value as per your book but in reverse however most photo diode detectors are infra red !! probably wouldnt be much good to him Although they can and do respond to ambient light but for pure white light with no infra red spectrum they may not do So he might be better off using a green super bright led

Colin is right, a photo darlington transistor would probably work better (be more sensitive). Didn't think of that. Plus, it's a lower parts count

Sorry to take so long to reply. Can you show me what the circuit would look like? Bear in mind that I am even lower than a rank amateur when it comes to electronics terminology, so please explain it to me like I was a 6-yr-old.

To Steve Lawson:
That "photo" of a circuit looks like something I could understand. Can you answer a couple questions, however? I assume that that thing with all the holes is a type of circuit board I could buy. Do I just shove the stripped ends of insulated wires into the holes as shown? You refer to a PN2222A transistor - is that the object labeled "N" in the photo? And the red object – I assume that's the light that will come on when the photocell detects light coming from the telephone's LED, correct? Can I use a lamp there that is quite bright, one that will force me to notice it, that is? Thanks.

The "thing with the holes" is a "solderless breadboard" or "protoboard":

* http://www.jameco.com/webapp/wcs/stores/servlet/Product_10001_10001_20601_-1

They are really handy for prototyping circuits -- but not really intended for final assembly (i.e. the parts should, ultimately, be soldered together (or crimped). That can be done on a generic circuit board and you can actually get ones that are layed out in the same configuration as the protoboard:

* http://www.allelectronics.com/make-a-store/item/SB-300/300-POINT-SOLDERABLE-BREADBOARD/1.html

The object labeled "N" is indeed the PN2222A.

The "red object" is the LED that will indicate the state of the telephone lamp.

By "amp there that is quite bright" I assume you mean a filament lamp. This get's tricky because of the current involved -- and because this circuit doesn't insure that the transistor goes into saturation (the mode you want for driving higher currents). Up to around 100ma should be alright, and maybe a little higher, but you'll have to experiment (smoke and explosions may be eminent ;)

How about a superbright LED, or an LED strip (that draws no more than 250ma [or more if you dare -- and use a 2N2222A, instead, it has a max collector current of 600ma (but I would use no more than around 300ma -- depending on the kind of environment you plan to use this in)

OR, instead of a 2N2222A, you could try something like a TIP102. That can handle up to 8Amps (but, again, I would keep that down to 6A, unless you mount it on an adiquate heatsink -- try it out, and monitor the temperature. If it gets to hot to touch, throw a heat sink on it. If it still gets too hot, try a bigger heatsink, if that doesn't work, then use a dimmer lamp or consider a power MOSFET.

You may need to adjust the value of the 47K resistor. Also, if you DO use a filiment lamp, then remove the other resistor (the one in seriies with the LED). And replace the battery with whatever voltage the "brighter lamp" is rated for.

Whew! So many variables. I have years of smoke under my belt. That may be what it takes for you

Steve,
Thanks so much. I think I can get this done now. I'll let you know one way or the other, teacher.

Hey, Steve Lawson:

I appreciate the detailed picture you posted of what the breadboard setup looks like. Unfortunately, I haven't yet gotten it to work. I bought a breadboard that looks exactly like the one in your picture, and I was careful to wire it up just like you drew. So I suspect that it may be due to the components I bought at RadioShack (where the clerk knew even less than I do about electronics!). I had the photocell delivered through the mail. It's an SEN-09088 photocell with these specs: Light resistance : ~1k Ohm; Dark resistance : ~10k Ohm; Max voltage : 150V; Max power: 100mW. At the RadioShack store I bought a 5-pack of 470-Ohm, 1/2-Watt Carbon-Film Resistors; a 5-pack of 47K-Ohm, 1/2-Watt Carbon-Film Resistors; an NPN transistor marked as MPS2222A (they didn't have any marked as PN2222A, as you had specified); and a 12-Volt filament-type DC lamp. ........... I note that some here have mentioned a "trimpot." Is that jargon for something labeled as a potentiometer? ................... Anyway, I'm hoping that I'm close to success.

(never mind; I edited a mistake in the above)

Steve,

Thanks for the response. Sorry for the lack of specificity regarding the failure. What happens is this: the LED (actually, a filament-type bulb, not an LED) doesn’t light at all, regardless of whether I shine a bright light on the photocell or cover the photocell in complete darkness. (Of course, ahead of time, I tested whether the 9-volt battery would light the bulb by just using a battery + wires + bulb circuit, and it does light up that way.) I wish I could post photos of my circuit, but I don't have the equipment to do that. I can assure you, however, that it is set up exactly like you laid it out in the "picture" you posted above, with the possible exception of the "polarity" of the resistors and the transistor. What I mean by that is, if you're looking at, e.g., a 470-Ohm resistor, does it make any difference which one of the two ends you put in the hole on the right , as long as you do have its two ends inserted into the correct position/holes shown on the breadboard? That is, does flipping its orientation 180 degrees make any difference? I would think one could put them into the breadboard in either orientation. Same with the MPS2222A transistor -- if you think of the three prongs as Prong 1, Prong 2, and Prong 3, as long as Prong 2 goes in the middle breadboard hole, does it make any difference whether Prong 1 or Prong 3 goes into the left-side hole?

Your browbeating is very moderate, Steve.

I have a general question about this whole circuit setup. Let me describe what I think is going on with it. If you just set up a battery connected to the lamp by two wires, the lamp would be on all the time, of course. With the photocell added to the circuit, the lamp will never come on because the photocell always creates enough resistance that the electrical circuit cannot be completed. That is, the photocell, in that case, is just an "off switch." When you add in the transistor, somehow it detects that the photocell is producing either low resistance (i.e., when a flashlight is shined on the photocell) or high resistance (i.e., when photocell is kept in the dark). When the transistor detects a low resistance level in the photocell, it somehow "swithches" the pathway of electricity in such a way that the resistance of the photocell is completely taken out of the equation, allowing the electrical circuit to be completed. Frankly, I don't at all get what the function of resistors is.

Am I understanding even part of the matter?

AN UPDATE: I reread the posts in this thread. In particular, I have noticed, for the first time, Steve Lawson's comment that if I created a circuit like the one in his "picture," if I used a filament-type lamp instead of an LED, I must remove the 470-Ohm resistor shown in his picture. I did so. I then figured that I had to replace that resistor with a regular wire. I did so. Now, the 12-volt filament-type lamp stays on all the time, regardless of whether I flash a bright flashlight on the photocell or whether I put the photocell into complete darkness.

Colin Mitchell, you're just being a douchebag now. You say I have absolutely no idea what I am doing. HAVEN'T I MADE IT CLEAR NUMEROUS TIMES THAT I KNOW NOTHING ABOUT ELECTRONICS??!!! ......... Jesus Christ!

Now you're on a kick about "the globe." Do you not get from my responses that I have no idea what you mean with "the globe"?

Do me a favor -- don't post anymore in this thread.

Yes, the fact that it's a filament type bulb and not an LED is a HUGE distinction - sorry if I missed that (haven't had time to read all the responses).

As Colin stated, the filament bulb requires considerable more current to run, so two considerations:

* the power source has to be able to supply the amount of current required. Apparently you demonstrated that there is sufficient power, by hooking the bulb directly to the battery (though I doubt that battery [assuming it's a common 9V battery] will sustain the lamp for long--though it is a plus that the lamp will only need to glow occasionally and intermittently.
* you need further current amplification.

BUT, before we go there, if you have an LED, then use it in place of the filament lamp, for testing/troubleshooting purposes, and do the experiments I described in a previous post, and let me know the result.

Your other questions:

* Resistors don't have polarity (i.e. doesn't matter which way they are connected)
* Transistors _do_ have polarity -- connecting them backwards will, typically, cause them to just not work, but in a few cases they will fail, sometimes quietly, sometimes with smoke, popping and other such pyrotechnic display (in your case, hooking it up backwards should cause no damage, just no happiness . Just make sure the _half moon_ shape on the top of the transistor is oriented the same was as in my diagram.

Actually, the transistor is a _current amplifier_. It has three terminals. One is called the _emitter_, one the _base_ and one the _collector_. A current traveling through the transistor from the base to the emitter [or base-emitter junction] will control a current traveling form the collector to the emitter (as long as the transistor is running within it's _active region_ -- consider that a preview of the complexities involved).

So, the photocell is controlling a smaller current through the base-emitter junction which will cause the transistor to attempt to control a larger current passing from the collector to the emitter. I say, _attempt_ because, if the _load_ (e.g. the lamp), has too low of a resistance (or impedance), then the transistor won't "open up" enough to allow that much current to flow. The ability of the transistor to amplify current is called the _current gain_ or _beta_ of the transistor, and for a small signal transistor, like the PN2222A, has a range of 35 to 300, depending on the voltage and current that the device is dealing with, but Colin's estimate of 100 is probably realistic.

Colin's idea of putting a 10Ω resistor in series with the lamp might work, but not for the reason he stated. If it does work, it's because the 10Ω resistor is limiting the surge current that is caused by the cold filament (the filaments resistance increases substantially when it heats to the _white hot_ level that provides illumination). By limiting this surge current, it might give the lamp a chance to heat up, even though the transistor is, essentially, keeping the current, going into it's collector, at a fairly constant value, determined by the formula:

Ic = ß*Ib where Ic is the _Collector Current_, ß is the _Beta_, and Ib is the _Base Current_

But, my prediction is that it won't make a difference because by limiting the current, the filament can never get hot enough.

Getting the _function_ of the resistors, takes study, over months if not years, so you might have to crack a book or two

Yes, that would be a useful parameter: the current that the lamp requires to illuminate properly. Do you have a way to test this?

And, I must confess, I don't remember that I quoted before--so many responses to sift through--but if I advised you to remove the 470Ω resistor, I might have led you astray. That transistor may not be capable of handling the current required by the filament lamp, and it may have gone bad (i.e. the collector-emitter junction may now be shorted).

The best course would be to replace the transistor, and see if you can get it working with an LED using the suggestions I have made in other posts. Then we can work from there.

Steve Lawson,
You have been very helpful; I appreciate all of it. You know that "picture" of a circuit you posted above (labeled as "one month ago")? Would it be possible for you do do another one like that, but showing the modifications you have recently proposed? And can I ask you to be as specific as possible regarding the LED you first spoke of? When I went to RadioShack to buy the components, they had quite a few LEDs, and without knowing what specs to look for, I just got confused, so I cought a 12-volt filament lamp instead. Let me say that I am perfectly willing to but new components for everything -- that's no problem. (I do still have 2 SEN-09088 photocells with these specs: Light resistance : ~1k Ohm; Dark resistance : ~10k Ohm; Max voltage : 150V; Max power: 100mW; 4 470-Ohm, 1/2-Watt Carbon-Film Resistors; 4 47K-Ohm, 1/2-Watt Carbon-Film Resistors; and a 9-volt battery.

If you are willing to still advise me, Steve, let me express my thanks in advance.

Steve Lawson,
Please note my message to you just above.

The LED you need is just a plain ol' Red LED, rated to run at no more than 20-30ma. Make sure it isn't one designed to run on, say 5V or 12V. It should have a forward voltage of around 1.7 to 2.2 volts (anything below 2.2V should do it - in other words, that's a way to distinguish between a plain (discrete) LED and one with internal current limiting).

So, based on the data you provided [Light resistance : ~1k; Dark resistance : ~10k], give the attached a try.

But, "light resistance" is a qualitative parameter and it's hard to tell from that if the light on your phone will be bright enough to trigger the circuit I offered, below. That's why I said you may need to _tweak_ the value of the resistor that goes from the Transistor base (the center wire) and ground (the negative side of the battery).

As for the 12V lamp, lets get the circuit working with an LED first. Driving a filament lamp is a bit more involved and adds enough extra variables to make this too complex to achieve with someone with the lack of experience the you have -- no offense.