Photodiode for an IR LASER

Thanks Dave,

Can you suggest a place to get these silicon photodiodes?

John,

Thanks for the reply. As you mentioned none of these seem optimized for 780nm. Are photodiodes optimized for 780nm hard to come by? Should I be looking for a different IR LASER diode the will be more compatible with the available IR photodiodes?

I'm going to be using this breakbeam outdoors in very bright light, so I think I'll need the daylight filters.

For the IR LASER diode the Wavelength @ Peak Emission (min/max) is

770-799nM

Thanks.

John,

Do you think there is a better IR laser diode that is around 880nm so I don't have to take the extra step of adding my own filter?

My plan was to just run the laser as is, do you think it would be a good idea to modulate the laser?

My whole set up was going to be:

1. IR laser diode (focused on 2)
2. IR photo diode (connected to 3)
3. High Speed Analog Input Module (weedtech.com Model:WTPCT-M) connected to a PC.

Then I would write a simple program to monitor the photodiode continuously and report every time the IR laser diode is not pointed on the photodiode (whenever something is obstructing the beam).

What do you think?

I have no idea what sensitive area you need or what frequency response, but diodes with a visible blocking filter built in are easy to use and quite low cost. Here are a couple examples available from Digikey: PD481PI

PDB-C158F

BPW34FA

If you use a detector in an T1 3/4 LED package it may be enough of a lens to ignore some interference from the side: PDB-C142F

These are really optimized for 880 nm LED sources, so using the unfiltered version would slightly increase the sensitivity to 780 nm, but would let in a lot of interfering daylight. BPW34 is representative of many unfiltered silicon photo diodes:

I don't know of any silicon diodes that come with filters optimized for 780 nm, but the ones centered on 880 nm usually pass about 50 to

70% at 780 nm. You can certainly use an unfiltered diode and add your own long pass filter that passes more than the prefiltered diodes do.

An example of a filter glass that would pass 780 nm efficiently:

Are you modulating the laser at some frequency?

What kind of collection optics do you imagine? There are some plastic Fresnel lenses that are made with IR long pass filter material (I.E Poly IR6).

I think you should start with a diode filtered for 880 and see how much signal you can pull out. A 1/2" RG715 filter from Edmond Optical costs about $14. You can build a bit of amplifier for that.

Yes, a very good idea. This eliminates any need for DC accuracy, low frequency noise or stability in the detector amplifier. At the very least, you amplify a signal that passes through a band pass filter at about the right frequency, eliminating the static background illumination and the 120 Hz flicker caused by line powered lights. At the best, you make the band pass filter a lock in amplifier driven by the modulating signal, so that it amplifies not only the correct frequency, but when synchronously demodulated, only the correct phase.

This will pull a badly corrupted signal out of a large amount of background noise.

Have you thought about eye safety, and what happens if someone stares into the beam? You may need to expand it to a significant beam diameter to keep much of it entering the eye's pupil. Or you hang a sign over the laser that says, "Don't stare into the laser with the remaining, good eye."

Here is where I see the need for a lens that collects collimated beam over as large a area as possible and concentrates it onto the diode. You can add a tube in front of the lens to help block stray light.

You don't say how far 2 is from 1. Is this an inch or a mile?

If you have signal problems, the bandpass filter and rectifier or lock in amplifier goes between detector and analog input.

Do not stare into the laser with the remaining, good eye. IR lasers are more dangerous than visible ones, because they do not trigger the blink reflex.

Silicon photodiodes work fine for that wavelenght. They are likely to be your best choice and readily available.

?

You have to modulate the beam, the smaller the bandwidth or your reciever the better. Lasers are not the best choice for transmitter,very hard to keep aligned at that distace, any reasonable IR emitter with a lens will do.

...

I wonder, how do you point it? You can't see the spot. Telescopic IR camera?

Thanks, Rich

Thanks for the reply John.

I searched around for the filter you were talking about on edmondoptics.com but I couldn't find the one you mentioned, can you give me a link? You also mentioned an amplifier, is this something I will definitely need? If so, how do I go about building one?

Can you explain what you mean by "DC accuracy"? How do I go about modulating the laser? Is there a step by step guide for this online somewhere?

at

Can you help me select the correct band pass filter to filter out "eliminating the static background illumination and the 120 Hz flicker caused by line powered lights?" What would be the correct frequency?

Okay, if this is the best way to do it, then this is what I want to do. I want this to work as well as it can. Although, how do I go about making the band pass filter a lock in amplifier driven by the modulating signal, so that it amplifies not only the correct frequency, but when synchronously demodulated, only the correct phase? Is there a step by step guide somewhere that explains how to do this?

I am definitely concerned with eye safety, but I've had very little experience with lasers so I'm not sure what precautions to take. If I expand the beam diameter (how do I do this?) to keep much of it from entering the eye's pupil, how do I refocus it back down to the diameter of the IR receiver, or can I just leave it as a large diameter? And by the way, how large of a diameter would I need to increase the beam to keep much of the laser from entering the eye's pupil? Also, won't this cause diffusion of the beam and prevent me from projecting the beam 650feet (or 200m)?

This sounds like a great idea, can you help me select the correct lens?

650feet (or 200m)

Sorry, for all the questions. Thank you so much for your help.

Can you explain, how I go about modulating the beam? Is this something that I have to tweak on the diode driver?

Why will an IR emitter with a lens be easier to align at 650feet than an IR laser diode?

Rich,

That is a really good question. I guess I'll try and get the laser and photodiode at the exact same elevation, then use a level and a compass to try and dial them in. Then, I'll just monitor the photodiode with a PC to see when its reading the most IR energy.

If anyone has any ideas on how to do this, I'd be very interested.

About 2 screens down the list, glass reference number RG715.

Larger ones are further down.

Measuring steady intensity requires DC stability (constant gain, constant offset, with low level of low frequency noise). The noise level of opamps rises as the frequency falls.

Modulating the source at a constant frequency allows you to pass the signal through a bandpass filter you need constant gain at that frequency, and low noise only in that band. DC drift doesn't matter at all, and the only noise you deal with is that in the pass band. Then you rectify the amplified AC to restore the intensity information. If you can do synchronous rectification (reversing the sign of the gain each time the laser changes from off to on and vice versa)_you have a lock in amplifier that rejects noise that gets through the band pass filter but is not synchronous with the laser modulation. (If the noise is random, some gets amplified with one polarity and some gets amplified with the other polarity, so a bit of averaging after the demodulation averages it out to near zero.)

This approach may need an adjustable phase shifter before the demodulation to correct for the phase shift through the system for maximum output. A true lock in amplifier would demodulate two signals with 90 degree phase shift and take the square root of the sum of the squares of the two DC results to find the magnitude of the modulated signal, regardless of its phase. But that is probably way more capability than you need, here.

at

You want as high a frequency that does not degrade the signal with either the laser or photo diode response. A few to a few tens of kilohertz is common. The bandpass filter is usually just an active filter made with the opamps that amplify the photo diode signal.

A nice little filter design program is available free from Texas Instruments:

Here is another one from Microchip, though I haven't used it, yet.

Do you know how to make an amplifier that switches from a positive gain to a negative gain using a cmos switch? There are several ways, including just have two amplifiers in parallel, one inverting, and one with the same gain but non inverting, and selecting back and forth between them with a SPDT CMOS switch. you get 3 of these in a CD4053.

It can also be done with a single opamp and a SPDT switch like one of the 4 in a CD4066. (which you can combine to make the double throw switch, above.)

You need the modulation pulse signal from the laser available at the photo diode location to make this work. So it might not be practical if they are widely separated, without running a long cable.

A pair of lenses (a small concave one to fan the beam out and a larger convex one to collimate it back to a parallel but larger beam is one way. There is still some danger, if the beam bounces off a shiny surface that happens to refocus it, but simply intercepting the wider beam is a lot less dangerous then looking directly into the concentrated one. It is also easier to get the wider beam somewhere on the detector. A single convex focusing lens can be used at the detector to gather enough of the beam to let it operate. If the beam is several times the diameter of the collection lens, there should still be enough signal for a clean recovery, and a lot less problems with beam aiming. Put the visible blocking filter (if any)right against the detector to minimize the need for a big filter.

How expensive is your lawyer? The calculations are pretty complicated. I'll see if I can find the fat document that defines the safety standards. But it depends mainly on the power of the laser. If its power is low enough the raw beam may be safe.

You will have a pretty fat beam at 650 feet, even if you try to make the beam as small as you can. Spreading it properly will just make it fat all the way over, instead of cone shaped. You may also need a circularizing lens at the laser to change the fan shaped beam from the die into a more round beam. Homework:

Adding a cylinder lens to the collimation optics can really improve the beam collimation.

You may need a stand in visible laser in the same package as the IR laser to test the optics.

You don't need a high quality image, just gathering, so a plastic Fresnel is probably the lowest cost, large but short focal length lens you can get.

They sell them the size of large screen televisions, but I think a few inches in diameter might do.

Doesn't fit in a mouse, then. ;-)

You simple turn the laser on and off with a square wave.

It won't produce a nice a beam, but if you make a multi watt array of LEDs a foot across, and use narrow angle LEDs (and maybe a Fresnel lens in front of that) you can throw a lot of light that you don't align so much as point in the general direction and flood the area.

And being a big fat beam that can't be focused to a point, it is no danger to a tiny pupil.

Here is an example of a cheap, high output, narrow angle LED that is a perfect fit to the filtered photo diodes.

John,

Thanks so much for all your help.

As I think more about my project, I can't use a large array of LEDs. I'm planning on have a small array of breakbeams stacked on top of eachother, maybe a couple inches apart. I need the beams to be very small, because I can't have them accidently projecting into one of the other photodiodes. Also, I'd prefer the beam to be a perfect cylinder all the way from the laser to the the receiver. So maybe I should get a cylinder lens as you mentioned above, can you suggest a good source for one of these cylinder lenses?

Okay, great, I'll get one of these.

Yeah, I'm not sure if it will be feasible to run a long cable. So I guess I'll just try and start out simple and see if it works. Then I'll add more complex filters and modulation as it becomes necessary.

I think I'll start out with this:

1. IR Laser Diode w/ collimating lens
   formatting link
   No.LM4).
2. 880nM filter.

----650 feet------

1. Visible light filter attached to the front of a black ABS tube.
2. IR photodiode
   formatting link
   ) inside black ABS tube. I'll start out with no modulation and see how much distance I can get out of this setup. What do you think?

This is a really good idea, I hope I can incorporate a visible laser into the system fairly simply.

As far as safety, maybe I just need to get a lower powered laser. Do you know what the highest power is that is safe if it accidentally enters the eye? The laser I want to use is less than 5mW.

This is why I think I might need to change my whole approach. It just seems like its too dangerous.

What are my options for doing this with LEDs? It can't be an array of LEDs it has to be a single IR LED that is focused similar to a laser beam. I know they use them in outdoor security devices like this one:

Model:AX-350/650 MKIII

I've looked inside of one of these things and they use LEDs, I just don't know enough about electrical engineering to copy their setup.

Edmund Optics sells them, but they are spendy. Knowing what focal length lens you need is another problem.