OPTEK 0PB705 Optical Switch – Reflective, Phototransistor Output Reflective Sensor

How do I wire up one of these, please - an OPTEK OPB705 optical sensor? It came without any user instructions.

It looks straightforward and yes, I've read  the datasheet but I can't get it to work. And yes, I've tried Googling but I can't find a circuit diagram featuring the thing.

I only want it to switch a LED on and off at this stage. All advice gratefully received.

Michelle O'Brien

Hi Michelle-First, it's important to limit the current to the LED.  According to the spec sheet it is rated 40mA and has a 1.7 volt forward voltage drop. If you're using a 5 volt supply try about 140 ohms in series with the LED, which will give you about 30 mA. If you're using 12 volts, 380 ohms will be about right.  This rwesistor is not critical as long as you don't exceed 40mA into the LED.For the phototransistor sensor, connect the emitter to ground and a pullup resistor between your supply voltage and the collector.  Again, this resistor is not very critical.  Try 1K to 10K ohms. If you power it up and connect a multimeter between the emitter and collector of the phototransistor you'll see the voltage move around as you move objects in it's field of view.Now- to connect the LED- since the voltage from the phototransistor is analog and won't give you a clean 0n-off indication, I'd recommend connecting a Schmitt trigger to clean up the output. You could either make one using a comparator or op amp and a few resistors, but the simplest way would be to use a CMOS hex Schmitt trigger, like a CD40106.  One of the nice things about these is that they are happy with a 5 volt supply or a 12 volt supply. Connect the collector (and pullup resistor) to one of the inputs, and connect the anode of your LED to an appropriate current limiting resistor and the other side of the resistor to the + supply, and connect the cathode of the LED to the output of the Schmitt trigger.That should get everything working.If you're eager to see it do something and don't have a Schmitt trigger on hand, just take a 1K resistor from + supply to the anode of your LED. connect the cathode of the LED to the collector of the phototransistor.  That should get it working, but you'll notice that the LED gets brighter and dimmer according to the amount of light that gets reflected into the phototransistor instead of giving a clean on-off indication.    Let us know how it goes.-Rck

OPTEK LED:cathode to supply gnd. anode to resistor, R1. other side of the resistor,R1 to supply voltage. Assume supply voltage to be 5V. Aim for 10mA through LED. Via Ohm's law R1=(5-1.6)/10=340R. (1.6 is the nominal forward voltage across the LED)If you tried the LED without a resistor R1, then I'm sorry, you're going to need a new OPB705.OPTEK TRANSISTOR:emitter to supply gnd. collector to cathode of your desired LED. anode of your desired LED to resistor R2. Other side of R2 to your supply. If we go for 10mA as well the resistor value is about the same as R1.To start out with I would use a highly reflective surface to test, bringing it up close to the OPTEK. The data sheet should give the optimal distanceIn general. The supply of the input LED need not be the same as the supply on the output side. They could be completely isolated or have a common ground or be one and the same, as I described above. Addendum:I was writing this while Rick was writing his. I think his description is better!Afterthought: You may want to add a diode like a 1N4148 (even a 1N4007 will do) across the OPTEK LED in the opposite polarity to protect the LED against a mistake in wiring.

Thank you for your words, Rick and Aubrey. But I'm sorry, I'm as much confused as before.

You are talking about two LEDS, Rick, yes? - LED-A, the one in the OPB705; and LED-B, the one  I've been attempting to use as a test of the device?

Before I burn out another OPB705, what I think I need is a sketch of a complete circuit showing

1) the OPB705 and its four pins
2) which of the pins is what (anode, cathode, emitter, collector )
3) where the resistors go
4) what is connected to 5v and what is connected to ground
5) where my test  LED, LED-B,  with its protective resistor goes and what it's connected to at each end

Could I prevail on someone for a simple sketch? I'd be truly grateful.

Michelle

Michelle you wanted a circuit, try this:
R1 limits the current through the LED of the sensor to just under 40 mA. Pin 2 of the sensor is connected to Ground (0V).Pin 4 of the sensor (the transistor collector) is connected to +5V.R2 limits the current into the BC548 base if the sensor switches hard onR3 ensures that when the sensor is dark you won't get the BC548 switching on (the dark current according to the datasheet is up to 250 nA)R4 limits the current in your LED.Once you have set this up you should see a voltage on pin 3 of the sensor (up to a few volts) when you place a reflective surface in front of it.  Otherwise there should be almost nothing on it.You will note that your reflective surface should be max 0.2 inches from the sensor.Let us know how you go.Oh and its an OPB705 not an 0PB705 (the first character is an O not a zero).  Wondered why I couldn't find anything when I first googled it....

David, thank you so much. In diagram form and with those added notes, now I do understand.

Would the shiny side of an old CD be reflective enough? I hope so. I plan to use a protractor to measure out and then paint in dark segments on the CD periphery to give a dark/reflective alternating pattern that the OPB705 (got it right this time) can pick up.
The ultimate plan is a counting circuit intended for motor control. We shall see. The most important bit for now is getting the OPB705 up and running, so thank you everyone for your help, advice and encouragement.
Michelle

The shiny side of an old CD should work fine. Bear in mind that things that are reflective (or transmissive) to visible light may behave very differently in the infrared portion of the spectrum.Please let us know how it works out!-Rick

It works brilliantly. Thank you so much Rick, Aubrey and David.

No reflective surface, the LED stays off. Reflective surface, including the shiny side of an old CD (see above) the LED lights up. Perfect.

The only change I made was to substitute a BC547 NPN transistor for the BC548 recommended by David in his hugely helpful circuit diagram, and that simply because I did not have a BC548 to hand. It worked just as well.

This has made my day. The OPB705 combined with that circuit is a great piece of kit and EEWeb is a great site for help, advice and encouragement. Thanks again, everyone.

Michelle

Great result, well done Michelle, your thanks makes this often thankless task very worthwhile.Bc547/8 - in this instance no problem, the minor differences will not affect this circuit.Have fun with your further experimentation (with motors I think?)Cheers // David

If I feed the product/output of this device/circuit into a counting circuit will it require debouncing?

Michelle

Michelle probably not, but you'll have to try it and see.  If your black / reflective edges are nice and sharply defined, probably not.  If you do get multiple counts (ie per black area) you could try putting a 0.1uF or larger capacitor across the 10K resistor (ie between the BC547 base and ground.  If that does not help you'll have to adopt stronger measures, but let's cross that bridge when we get to it.  You could try things like feeding into a schmitt trigger input gate, for example, but we'll advise you on this if you need it.

To ensure the edges of the black areas are as sharply defined as possible against the parts of the old CD that have to be left shiny for the OPB705 to work (please remember, this is kitchen table electronics) I've substituted black electrical insulating tape for black marker pen.
To get things even more spot on, I used a circular (all 360 degrees) protractor to mark out exactly where the black tape has to go.The really difficult part has been to eliminate as much as possible any trace of wobble when the CD spins. And thereby to keep the head of the OPB705 at constantly just the right distance from the spinning periphery at each stage of the CD's revolution.Easier said than done. But so far so good, I'm glad to say.Michelle

Good thinking.  Sounds like you're having fun, Michelle

It occurs to me that if I carefully position a second OPB705 in this set up and feed the results I obtain from the two of them into an Arduino (or whatever) I shall get:not only a lot LEDs flickering on and off, but also I could get a recurring ON ON/OFF ON/OFF OFF/ON OFF pattern (or vice versa) - ie. the start of a quadrature rotary encoder for controlling the direction and speed of an "ordinary" dc motor.
Now there's something to tell the Marines. Trouble is, like me, they  follow this stuff on the internet and already know. But you have to admit it's all very interesting.
Michelle

Michelle if you google "Quadrature decoder" or "Hardware quadrature decoder" you will find some ideas.  You have the choice of doing this completely in software - which I am sure has been done before and you'd probably find some code somewhere - or using a hardware decoder which will present you with two neat outputs - count indicating a movement, and direction indicating which way it's going.  There are a few relatively simple hardware circuits that will do this but if software is your "thing" then in the scheme of things it's not going to use much resources.

If I wanted to feed the on/off signals from the OPB705 into, for example, a 555 timer or use them as the clock pulse for something like a 4029 counter where would I start the Vout signal line in that circuit you've given me?

Michelle

Michelle sorry,only just saw your above,  if you reply I get an email alert, if you put a new comment I don't.  (Since it's your post you'll get alerted to any comment).OK, as a start I would try the following
The 0.01 uF capacitor (10 nF) should give some protection against glitches.For very fast transitions (eg if your CD is on a fast motor shaft) you may be able do reduce it to 1nF or take it out altogether.If that does not work (it should but it's possible that the output will not trigger whatever you are driving) then you could modify the original circuit like thisYou could also try the 10nF capacitor from the junction of R2/R3 to ground on the above circuit if you find you're getting false triggering or extra counts.See how you go with that and let us know how you go.  I'll try and keep an eye on this post....

My apologies for ignorance about the posting protocols for EEWeb, David.

When Practical Electronics magazine in the UK shut down its own  hosted on-line community last year and EEWeb kindly agreed to give us PE waifs a new on-line home I grabbed at the chance without properly understanding the rules here.

So, when I hit "reply" here you get an email alert but if I hit "add comment" instead you don't, right?

That apart, why more Practical Electronics readers don't participate on EEWeb's specially created forum for them I don't understand. The excellent advice and consideration I personally have received here - not least from you - have been a big encouragement to press on with a pursuit that on a bad day can appear dauntingly difficult.

I enjoy the challenge of electronics. But the hard work of understanding - well, sometimes it stretches the grey matter to creaking point. But I suppose that's the point.

Michelle

Hi Michelle.  I used to get PE many years ago and loved it, there days I subscribe to Elektor and a local Aussie electronics mag.  I should get PE again, it seems like they have some good projects in there.Comment / reply....  If you think this one's bad, you should try the old EE times.... But yes, replies alert me but comments (unless it's my thread) don't.  I am in the habit of looking up older ongoing threads but it does not happen every day.If you look up this thread, replies come in Grey background underneath the comment being replied to (then alternate white/grey). You replied this time, and I got an email alert! Hence the quick response.  Being on opposite sides of the earth does not help and delays it sometimes.It's always a pleasure responding to readers like you who appreciate it.  The first circuit I posted was just off the top of my head (though I have played with these things before) so I probably got almost as much kick out of it (when it worked ok) as you did :-)As you're a lady I won't ask how old you are.  I'm 62 and without electronics (and the odd word puzzle in the paper) to keep my grey matter ticking over I think I'd be a vegetable by now.  So keep it up!  As I say, it's a pleasure to assist anyone who has a genuine question (even if it seems a bit dumb) so don't be afraid to ask anything.Cheers // David