IR phototransistor reciever

Phototransistors are the wrong answer for almost anything in free space.

What are you using for a transmitter?

Cheers

Phil Hobbs

Schematic?

There has to be a range, and some idea of pointing, to be sure, but... the easy way is to use remote-control receivers and modulated (alternately on/off) transmitters. Those are immune to things like sunlight and sources that aren't modulated.

If you put a black hood over the receiver, and/or use a lens to collimate a point-source into a narrow beam, and AIM the transmitter at the receiver, that can work at such distance. The signal from a photodiode will be weaker than a logic ON/OFF unless you can perform that aim function (try viewing a white paper with your cellphone as you aim the IR at it; cellphone cameras, some of 'em, can see the IR light).

Phototransistors have small collection area, that can hurt the signal; a lens focused at the transistor might help, but will be tricky to adjust with infrared light. Since both ends will require (battery?) power, a one-sided send/receive box and a retroreflective tape patch might be convenient. So would an IR (or even just deep red) filter.

A phototransistor can read its illumination by grounding the emitter, leaving the base OPEN (unconnected) and reading current into the collector from a +v source (battery) with a resistor and milliammeter; the battery, resistor, phototransistor, and current meter are all in one series circuit.

They are designed to receive *bursts* though and won't maintain the output in the same state (low) if the modulated light signal persists. So you'd need to modulate the 38kHz (which needs to be fairly accurate- maybe a few percent, so an untrimmed 555 is probably not good enough) and then stretch the pulse with a retriggerable multivibrator at the other end if you need to transmit levels that persist for very long.

What optics do you have at either end?

Small lenses will make a big difference as will using visible light to get started so you can see how big the beam sweet spot actually is. A couple of 1cm lenses (clear marbles at a pinch would improve things).

Collimated beams should go a long way. We have done adjacent hill tops optical links with a similar setup and modest 8" SCT's. There was at one time a plan to use a similar optical comms system in some US cities.

It is still a contender for long range high bandwidth satellite comms.

Remote control receivers don't use a (bare) phototransistor, but instead special tuned infrared receiver IC, the IC looks for a signal modulated at about 38kHz and returns a solid logic level signal when it sees one.

TSSP4038 is the same sort of thing but optimised for light barrier applications - available for a couple of bucks if you don't look too hard.

As the frequiency filter is on-chip it's not very tight so a good RC oscillator (like a 555) should be able to get close enough to the sensing frequency for reliable operation if you use a reasonably good timing capacitor. (main consideration: use a temperature stable capacitor)

TSSP4038 will operate from 2.5V to 5.5V so is suitable for 3.7V lithium or

4.5V carbon-zinc, carbon-zinc batteries drop to 1V per cell at nominal full discharge so you'll need three to keep this receiver happy. unless you want tho throw them out half-used.

You could build a frequency-pass circuit like the receiver ICs have inside if you really want to. it should be possible to get even better performance as its perfomance will not be contsrained by low precicion on-chip resistors and capacitors.

Are you doing this at DC with a the IRED continously on? You will need a lot a gain at the receiver end and drift free DC gain is hard. This is why almost all schemes use pulsed sources and AC amplification at the detector. As a bonus the setup then ignores slowly varying ambient light sources.

piglet

Not sure it's directly applicable to OP's problem but there was this kinda cool circuit for increasing phototransistor tolerance to ambient light:

Interesting find! One old-school solution to that problem was to operate the phototransistor into an inductor.

piglet

On 2022-01-09 20:40, piglet wrote: [...]

They want me to switch off my adblocker. Sorry, but no.

Jeroen Belleman

No loss. It's truly lame.

I wonder why some people design elaborate lowpass filtered negative feedback loops to accomplish AC coupling. There are easier ways.

Here's a way to use the commercial recievers:

U2-B output (with a pullup) should be what you want.

This is a really nice circuit for garage door safety beam, minimum wires and designed so that DIY and numb-skulled installers can't easily break anything or easily compromise the safety. Well, they

*can* do something like this:

He wants to build a beam-break detector using a phototransistor for whatever reason. I think there could be reasons to want to modulate the transmitter with a linear signal vs. pulses but then you have to have a way such that the phototransistor when connected as shown doesn't respond to low frequency; how do you "AC couple" it when it has a collector load as shown so that it don't pull down into saturation when you turn on the room lights.

I don't think a photodiode has quite the same problem you can just bung it into a TIA that doesn't have low-frequency gain, but that wasn't what OP say he wanted to use.

For a beam-break across 2 feet IDK why you want a separate detector and receiver anyway, why not put them in the same box and use a lil parabolic mirror on the other side. FM a pilot tone and bounce it across and back closing the PLL loop around that, when the phase detector goes out of lock the beam is broken.

I think a missing-pulse detector could work also just driving with pulses but then you gotta worry about how many missing pulses is an actual break vs. maybe a glitch.

Digital seems like the right way to transmit data using an IR LED and detector but I'm not enamored of it for this one; it's a beam-break detector not a garage door opener. I think a missing-pulse detector will be fiddly..but a "DC" beam is also not the right answer.

With an FM loop so long as your detector doesn't saturate due to ambient light/low frequency you can just amplify the crap out of your carrier on receive and do a PLL-kind of thing, if your loop goes out of lock the beam is broken.

There's also integrated cookbook solutions to the detect-only-modulation function. Consider the venerable NE567 or LM567... discussed here

I've seen it used successfully, to make a sensor blind to ambient light, but electronic ballasts and LED illumination and backlit monitors... we're in an age where our ambient illumination comes with a lot of baggage.