PDB-C134 Photodiode Always Conducting—Why Won't It Turn Off in My IR Circuit?

I'm trying to set up a basic IR emitter and detector circuit. Here are the parts I have:

http://www.advancedphotonix.com/ap_products/pdfs/PDB-C134.pdf

http://www.digikey.com/scripts/DkSearch/dksus...ang=en&keywords=475-1454-ND&x=12&y=18&cur=USD

The photodiode is never open. It always lets current through and lights up the led I have in the circuit. I try to cover the photodiode as much as i can with everything but it never turns off. I'm new to this so I don't know how these work but I thought it only let current through if there was the IR emitter shining on it?
Thanks.

Without knowing the circuit, the voltage and the load it's about impossible to know what the problem is. The detector shouldn't even let much current through when it's on. Perhaps you have blown it with over voltage or reversed voltage. It would also help to know what you're trying to do, particularly what you are trying to drive..

The detector has 30 nA dark current, hard to even measure and only 30 uA lit, which isn't much. I can't imagine how this can be used without some sort of amplifier. Many photocells have much larger currents or even darlington outputs which can handle significant output current.

I did not notice a forward breakdown voltage in the data sheet. The max reverse voltage is 5v. Depending on your circuit you could easily be exceeding the operating limits of the detector.

I have used the transmitter detector pair QSE113/QEE113 in end of travel detectors for positioners. These are side-looking parts. The output is not a darlington but it is 25mA to 150mA max which is much more useable. There should be lots of similar parts out there.

I have no idea how to do it but I'm trying to simply tell a microcontroller if something is blocking the path in between the emitter and detector. Kind of like a trip wire. Are there any circuits you can link me to that could do this? I know microcontrollers very well but what I don't know is infrared circuit to do this.

Thank you.

The first test is to remove the photo detector Simple fault analysis first

If led still up and running when removed

If answer. Equals yes then fault elsewhere in circuit

Else

if answer equals no then
possible infra red detector fault
replace detector

if after replacement led re lights then
possible led is in the circuit round the wrong way
reverse the detector

else

cover up diode with black insulation tape
re test

end if


End if

End if

Now we learn not only how to fault find but we also learn how to program at the same time

If you still cant resolve "notice use of words chosen to move towards understanding of dns resolver for possible tuition in webservers such as apache etc "

Upload a schematic plus code if youve attached to a micro also please include a flow chart

Now you can see why cant you Method in my madness

I have no idea what the above comment means.

If I understand your request you need a simple interrupter circuit to a microcessor. I've attached a simplified version of what I have used. I hope this helps.

Are you saying you did not understand my post? If not, that is okay because you have provided me with the exact information that I need. I will order the parts very soon and your circuit will be perfect. The question I now have is resistor sizing. I will be using TTL so 5V and about .7V across the emitter which leaves 4.3V across the first resistor. If I use a 150 ohm resistor that will draw about 30mA through that diode. I'm guessing I will have to play around with this resistance a little bit to get the intensity right but is this a good resistance value to start with or do I need more current? Also, I do not know about the second resistor. Perhaps you could teach me why that is necessary and how to size it properly.

Thank you for your very helpful replies Stephen.

I have no idea what the above comment means but here is a circuit for a micro using interrupts which the questionare will then have to code in the meantime his current circuit which we dont have is continously emmiting light from an led thats attached so it appears as though the ir transmitter. Which modulates the ir led at 38khz is permanently on The questionare cannot fault find iether and your going to pass this person a schematic of a micro which uses an interrupt

Im impressed Very good

How does this person learn from this then if you could possibly explain

Could you also explain how the user who is attaching your circuit to the micro configures the port for interrupts also could you also explain what type of intterupt he should use,how the questionare should access these ports and through what registers with flow charts code etc, could you also advise as to what langauge he should use with its merrits etc and could you also explain what you dont understand about my comments please

What is the point of this site Do they come here to learn and hopefully learn new skills or do we just create dummies

And just to put the cherry on the cake just as an example the emitter reciever pair plus circuit which is now attached to the fantastic micro happens to be interrupted twice in one micro second and my circuit isnt working properly and the questioniar is none the wiser becuase he hasent learnt instructions times either or understands risk architecture
But hes going to design

Yes okie cokie


I WONDER WHY INDUSTRY IS FAILING

I WONDER WHY WE HAVE A REAL PROBLEM WITH IMPORTS

I WONDER WHY MY SO CALLED EXECUTIVE SALESMAN ARE SELLING THE WRONG COMPONENTS TO OTHER COUNTRIES AND IM RECIEVING VAST AMOUNTS OF RETURNS BECAUSE
WE HAVE SOLD THEM THE INCORRECT PART FOR THE INTENDED APPLICATION

I WONDER WHY WE HAVE NO ENGINEERS ANYMORE

hmmm not to worry

My comment on the posts refer to the comments of Mark Harrington, which seem unintelligible gibberish to me. I'm not sure they are even related to the original question. I have no idea where 38 KHz comes from. As I understand this it is a simple interrupter circuit. When the receiver "sees" the transmitter you want the output to the microprocessor to be on.

I would not use a 150 ohm resistor. Unless the transmitter and emitter are very far apart there is no need for it and there is no reason to drive the transmitter near it's design limits. 470 ohmgives you about 10 mA which will work fine. 1k would probably even work. If in your application the receiver doesn't turn on strongly I'd increase the receiver and LED resistors, rather than reduce the transmitter resistance.

The microprocessor input should draw so little current that it has no effect. If the microprocessor input has an optional internal pull up I would turn it off.

The second resistor, connected to QSE113, pulls the microprocessor input to ground when the receiver is off. A pull down resistor with the receiver tied to Vcc means that the input to the microprocessor is virtually ground when the receiver is blocked. The input is pulled up when the receiver is on. This resistor could be much smaller, 1k or 2k would be good. Even 5k would probably work fine. I picked 470 so there was only one resistor value in the circuit.

The LED and resistor to indicate the output are optional and the LED could probably be driven with much less current so 1k or 2k would work here as well. Even with 2k the LED should be quite visible indoors.

In general I see no reason to drive parts near their limit when it isn't necessary.

- Stephen the Obdurate

Normally to get any distance on an ir transmitter reciever pair you modulate the ir transmitter with 38khz You can find out more about this if you read up on sony ir codes and phillips ir codes The other reason is so that any other ir source in the vicinity doesent interfere with you ir reciever

And you want cut out your comments as well regarding giberish We dont exchange comments like this on this site

If you dont understand what some one has written or why thier comments are such then ask but dont insult people in here most of us are proffesionals and proffesionals dont insult other people Use face book if you want to do that possibly more your class

The distance between the emitter and receiver is roughly 6in. Should I consider modulating it?

Ill re explain what ive done for the benefit What ive tried to do is show this person how to fault find on a very simple and is such a simple circuit logically so that without this even realising what ive done He will read if else end if comments with instructions as to what to under certain tests which he may not use right at this moment but later when he does he need to write software for any micro or application This will subconciously register at the back of his mind making next task hopefully more understanding without to much of a struggle

Any more funny comments want to make or can you now possibey behave like a proffesional adult

I could care less about Mark, but his comments seem to me to be off topic or wrong headed and seem to me to just confuse things. I don't see any specific information, shch as circuits or part numbers. I also don't see how personal rants about executive salesmen are helpful or germane.

38 KHz is not a magic frequency and will not increase the output power of the transmitter (that would violate conservation of energy). Using modulated signals with IR transmitters that transmit data has to do with communication issues and preventing the receiver from responding to spurious signals from other sources.

I assume here that there are no other IR sources to contend with.

IR encoders and decoders means more parts and probably microprocessors and software on each end, beyond and possibly far beyond what it takes for a microprocessor to recognize a simple input.

I am not like Noel Coward's Critics. I know how it's done, I have seen it done and have even done it myself. I started designing hardware and writing software for closed loop industrial hydraulic motion controls in the late 70's, starting with the 8080 (1/4 k ram and 4k prom - wow) and the 8048.

Multiplexing and demultiplexing the photocell adds hardware and a bit of software, making this more complicated, even assuming you are good at microprocessor programming. This is WAY beyond making a simple photocell. I have been programming for years and wouldn't consider it in this application, that is based on what I know about the application, which is almost nothing. I'm assuming that this is a simple blocked or clear detector, not recognizing pulses, as you would if you were looking at the teeth of a rotating sprocket..

The distance IS a big problem. The QSE113 data sheet shows the transmitter has a half angle of 50 degrees (the angle within which the output will be at least half the maximum brightness), making it easy to aim but giving it a small range.

The light distribution along with the coupling between the QSE113 and QEE113 are shown on page 4 of the data sheet. As you can see the couplling drops off quickly and is about zero at 6" with the transmitter pulsed at 100 ma with a 0.1% duty cycle.

Unfortunately the circuit I drew will not work at 6".

Pulsing the transmitter will not solve this and makes the project much more complex.

you could use an IR LED with a much narrower half angle (emission angle) but that would make it much harder to aim. The LED on the output could help with aiming. You could point the transmitter based on the LED brightness. You might also consider a visible transmitter and receiver which would make aiming easier but would only work if there wasn't a lot of ambient light or if you shielded the receiver. The circuit I drew would still work.

You could even use a laser. Small visible lasers are pretty cheap and it'd be easy to aim. Lasers will be much brighter than any ambient light so it wouldn't be a problem.

I haven't looked for transmitters with a narrow beam, lasers or visible receivers. I be happy to look if I know what you want, or I could tell you what I think about any parts you find.

Quote. "I couldnt care less about mark harringtons comments but unfortunatley the circuit I drew will not work at more than 6 inches " , ok, nothing to be concerened about at all is it! ! You dont design for the company ive just visited with all the returns comming back do you by any chance? Just wondered, simple question curiosity more than anything else Hope you dont mind me asking

Dont think i need to comment do you I think youve just realised that before you insult other people you need to look at yourself first hmmm

Would you care for some webpage links showing you how to build an ir transmitter reciever pair hmmm?

Might also be helpfull Although I do understan you possibly wont take any notice of my comments because they are in your oppinion not worth taking notice of but I do find from expeiance about 43years and more Im no 50years of age that is does really help if youve actually done some fault finding, field service in numerous fields and also done some programming and built a few projects See that way you realise what can go wrong more importantly that there are definate steps to learning how to design making sure of course that before you submit a design you consider if is of proffesional quality or at least works

Otherwise you end up with rescessions such the current which cause companies to fold left right and center Equally if you happen to be a salesman or a director It always pays to look after your staff and to listen on accasions plus if course pay people on time and see to it that they recieve decent training Then they can invest

OBVIOUSLY BARING THE ABOVE POINTS IN MIND, You tend to land up with stable economics and a decent company to work for plus people who dont mind working for you

I know very unimportant to you

Your ok now are you


Dont you just hate it when that happens hmmm say no more

Finally whilst chatting about design and software please see example below for proteus proffesional a cad software package aimed at people whishing to design and run this file in the simulator whilst refering to the datasheet and ask Why should people have to pay 25 pounds stirling a year plus purchase the package when the software written by a designer doesent work properly

Go ahead This is a typical example of why companies end up bankrupt all becuase someone wont listen to the consumer or thinks thier products always work doesent and fails to communicate the consumer ongoing whilst this is not a bad package Can you imagine what the outcome would be if you were incapable of building testing and producing this on paper see below Version 5.01 Still waiting for answers Now up to Year 5

Not acceptable is it www labcenter.co.uk

see below

The cheapest laser I can find is over $5. Since there will be 3 more pairs that puts a simple cheap project over $20 which is not what I am looking for in this case. I could use a visible light pair with a narrow range. The project will be under some things, not fully blocking light, but should do a significant job at blocking most light.

I think it would be best if I just quickly describe exactly what I am trying to do. The goal of this project is to read which gear a 5 speed manual transmission is in. 4 pairs of emitters and detectors can sent input information to a microprocessor which can decide which gear it is in (1,2,3,4,5,R,N).

I hope this helps. I found a guy who did it using IR emitters and detectors but I lost the link and cannot find it anymore I forgot to bookmark it and windows restarted my computer for updating without my permission. Blasted windows!

I hope you can help me out Stephen although I am pretty sure, because of how helpful you've already been, you will.

Thanks.

I do have some questions. I assume you are shensing the shifter outside the gearbox. I'd think you could do that with less than a 2" separation if it the sensors were near where the shifter enters the gearbox.

There are of course lots of shift patterns. I once drove an old fire truck with a 5-speed. 1st and rev were in the middle.

Making a guess at the shift pattern with arrows for the transmitters and receivers I'd imagine the sensor layout would be:
V V V
> R 2 4 >
neutral
> 1 3 5 >
V V V

You could even eliminate the center vertical sensor in the above picture and have four sensors to recognize six shift positions including reverse as well as neutral.

With a 1" or 2" range, the circuit I drew might even work, especially if you used narrower angle transmitter. However cross talk between sensors could be a problem. swapping some transmitters and receivers would help:

V /
> R 2 4 >
neutral
< 1 3 5 <
/ V

If you had a microprocessor with enough I/O and little to do it could even turn one transmitter on and read one receiver, completely eliminating crosstalk.

I'm curious what the microprocessor is doing with the shift information. It could even be possible with some very simple logic to output the actual shift position to a microprocessor, display or whatever.

There are bunches of microprocessors and lots of argument as to which is the best. I favor the Atmel 8051 family, mostly just because I'm used to it and I have lots of software already written. I also like using through hole parts for prototyping and Atmel has lots.

Let me know If I can help.

I'd be happy to give you my direct be-mail but I don't know how to do that without it becoming public and I already get enough spam (How do they know I need Viagra?)

- Stephen

The programming is easy. It just looks at the patters. Using 4 pairs it will produce enough patters to use simple if if else else statements. That's not what I need help with. It's getting the input to the microprocessor. You're right, I could try and get it really close to the ball joint but it's not a short throw shifter so I might be able to do 3". I have the microprocessor already. As long as I can get the inputs I can handle the rest no problem. I've just never worked with an emitter/receiver pair like this before so I need help setting that part up.

Again, I'm making wild guesses but here goes. My old car has a gearbox with shifting rods that extend outside the back of the transmission and are protected with a sheet metal cover. This cover already has a limit switch that is activiated by a shifting rod when the car is in reverse. In this case I'd try drilling holes in the cover and mounting transmitters and receivers side by side so the light bounced off the shaft into the receiver. This would have a very short sensing distance and the mounting would be easy. Just a thought ...

I wrote my last comment before I read your last comment.

Basically the circuit I have drawn will work. It's just a matter of finding a suitable IR transmitter with sufficient power, which means a small half angle. There can be a cross talk problem, depending on how the sensors are laid out, thaqt is how close they are. A narrow angle receiver will help. If you pulse individual transmitters you can increase the power some, but then the rerceivers have to be synchronized which makes things much more complicated so it's probably not worth it.