Re: Post #19

That is exactly what I was referring to in my earlier post.

John

 

Are you looking at the output with an oscilloscope or with a voltmeter? The output will be 1 when there is a carrier present and it will be 0 when the carrier is absent. With Vcc = 5V, and the output pulled up to Vcc, a three volt output on a voltmeter would indicate either an output that is beating up and down or an output that is floating.

 

My comment was based on the TSOP1738 and similar, which the RadioShack device is not. I assume Papabravo's comment is related to the RadioShack device specifically. For reference, however, here is the pertinent figure from the TSOP1738 datasheet:

John

 

a three volt output on a voltmeter would indicate either an output that is beating up and down or an output that is floating.

Well right now im feeding it 4.38 V and getting an output of 4.38 V with the window covered. Im guessing thats not right according to papabravo.

Okay here are two debugging scenarios (with the window of the Rx NOT convered):

1- Right now, with everything active, the reciever is placed right infront of the IR led and im getting 4.38V with a Vcc of 4.38 V fed to the reciever. It continues to read the same output with or without the IR led

2- with a pull up resistor, an input of 4.38V at the input reads a 1.22 V at the output, with or without the IR led

Both of these situations are inconclusive in my mind.

John, with my current setup for the 555 timer, here are the numbers (measure of the components themselves):
R1 = 9.7K, R2 = 14.6k, C = 0.95nF. These value lead to
Freq = 38.9Khz, Duty Cycle = 62.12 %

Does these make sense for the configuration? I dont know how i can caluclate the Signal Gap time (refered to post # 19) and i dont know how change it either to fit a specification of 20ms.

Honestly, Ive had enough of this circuit. All i wanted to do was to develop a proximity sensor using a IR led and a reciever. Can i not use just a simple IR phototransistor and an IR led to accomplish that?

Please help me!

 

If range and false receptions are not a problem, get a simple phototransistor or light dependent resistor (LDR) that are effective in the infrared. Otherwise, you need to go the dual 555 or 556 route as suggested. It is not that hard to do. There are lots of circuits out there.

Again, without knowing anything about your device, we can only guess, but we are still shooting in the dark. I read the consumer reviews on the RadioShack site, and it seems other users had similar experiences to what you are having. I realize it can be very frustrating. That is one reason I don't own RadioShack stock.

Do you have a real component supplier nearby? Where are you? What is your time frame for this project?

John

 

Okay..im a student at Virginia Tech and live in Virginia. I can order stuff online but radioshack in the closest electronic component supplier. Okay..if i do continue with the 555 timer option, i still have to deal with the signal gap time for properly modulating the IR led. If u verified the 555 timer circuit i have set-up, the only thing im not sure of is the signal gap time.

If i use the IR phototransistor and the IR led setup, i can always put them in a metal rod casing and seperate them in a way that the IR will bounce of an object and will be captured by the phototransistor, which will in turn, output a current/voltage. Should that not be an easier way.

This whole IR business started with me trying to make a sensor for a robot im trying to create. I havent made robots before so I decidede to give that a try. All the robot has to do is detect obstables and watch out for them, while moving autonomously.

Any ideas on that??

 

Yep. I had a problem several years ago where I needed to detect the position of a piece of 3/16" wire. I went the dual 555 /556 route and then discovered microcontrollers and never looked back. Specifically, I used the PIC 12F509. Whatever microcontolller you choose, it is way easier than analog.

My advice, find someone at VT who will help you do the actual burning of the chip. Start with Assembly, unless you know some other language. Download MPLab from Microchip. It's free. You may need to start with turning an LED on and off before coming back to the IR problem, but the programs will be remarkably similar.

If you want to stick with 555's let us know. I will try to find my old schematic or someone else here (Bill Marsden?) may have one ready to go.

John

 

well i would do that evetually. I still havent had any experience with an MCU before and was going to try that later.

555 timer to pulse an led at a certian frequency and then use an IR detector to detect is what I have been doing but to no use. If someone here has already done it and has a shcematic..i would love to take a look at it to see where i went wrong.

But for now, i would just stick to the IR phototransistor and IR led. BTW..do i need to pulse the LED for the IR transistor, or just have it set up as the attached picture. The picture is just a basic layout picture of the components

 

I found the 556 schematic, unfortunately it does not include values. Nevertheless, you can see how IC1B turns IC1A on and off through its reset pin. IC1A is at 38 KHz. IC1B is at the modulation frequency needed.

John

 

Nice! now i just gotta figure out what are the values of the components. Its gonna be tedious but not too hard.

Man..i gotta go hand it to you..ur are really helpful!! u know im a senior in Electrical Engineering but building circuits versus studying them theoretically teaches one far more.I'll MSG u later on when i design the IRphototransistor + IRLED circuit? You could critique it and mention any points im missing. Thanks!!

 

You already have IC1A worked out. Just figure some component values for IC1B. The target frequency is about 833 Hz. IC1B can, of course, be a separate 555.

John

 

I looked in another folder and found a schematic with component values -- sorry about that oversight. PPS stands for polystyrene capacitor. It's not really necessary. NOTE: I use a different configuration for the timing resistors than you do, so with your design, the values WILL be different.

John

 

I've been going through almost exactly this scenario trying to build an IR camera trigger. I think you've come to the correct conclusion about why it's not working. In my case I found that the output from the receiver would only go low for the 1/10 - 1/4 of a second when the IR beam hit it and then go high again. Another receiver would go high and low very quickly almost at random, which would explain your voltage reading.

I suspect you won't have much fun with the IR LED and phototransistor approach. I tried that with the camera trigger before I started working on the 38khz system, and it was horribly touchy. The receiver is easily tricked by ambient light. I would guess it would be pretty hopeless for object detection on a robot, which is going to be much more sensitive to interference than a beam-break camera trigger where the emitter and detector are pointed right at each other.

There are quite a few cheap prefab proximity detectors available for robotics. I've used Sharp's GP2D12 (see http://www.robotshop.ca/sharp-gp2d12-ir-range-sensor-10-cm-80-1.html) which is about $10 and gives you an analog output, which means you can set the proximity threshold on the controller side, instead of being dependent on the receiver's threshold.

There's lots of other options too. Check out:
http://www.robotshop.ca/infrared-light-sensors.html
http://www.sparkfun.com/commerce/categories.php?c=139
http://www.sparkfun.com/commerce/categories.php?c=84


What is going to be interpreting the signal from your proximity detector?

 

I suspect you won't have much fun with the IR LED and phototransistor approach.

The device being used here is apparently not just a phototransistor. Unfortunately, it is from RadioShack, so we do not have access to a real datasheet for it, but its description and behavior are consistent with it being a 38 KHz IR detection module similar to the TSOP1738 and similar devices.

John

 

I figured out the sensor for the robot. It works ..i would say with..80% efficiency. Ive tested the circuit and it seems to work just fine. take a look.

http://forum.allaboutcircuits.com/showthread.php?t=44994

 

That's good news. Thanks for the follow-up. John