Not what I was expecting. This is the PCB from inside a pedestrian detector as used on pedestrian crossings.
It uses two fresnel lenses behind a sheet of black infrared filter material to fire out a matrix of 30 infrared dots in the same way as a disco LED moonflower light. It then measures the amount of IR reflected back from each beam and uses a PIC microcontroller to analyse any fluctuations caused by someone moving in front of the unit.
On further investigation it looks as though the pairs of 1 ohm resistors in parallel may actually be current sense resistors to regulate the current through each set of LEDs to around 1A. It also looks as though there is a resistor and capacitor in series with each microcontroller control line for the LEDs to make sure that a glitch or crash can't leave the LEDs powered continuously. Although each set of three LEDs has its own current regulator (for wider voltage-range use?) the microcontroller only uses five lines to switch them as groups of six.
I'm not sure what condition causes the relay to operate as I didn't get it to click a single time while I was playing with the unit. It may be faulty or just require a specific sequence of events to indicate a level of activity or if there's been a gap in the traffic and someone has crossed early.
The software in the PIC microcontroller must be quite complex to analyse fluctuations in ambient light and when something is causing false triggering of one beam repeatedly. I'd guess it also processes the input to make a relay trigger decision based on pedestrian activity and time.
The unit is designed to run on 24V AC or DC and the relay looks like it may be providing a volt free contact to the control system. The other relay position is probably for another function like fault indication or maybe to indicate other detected conditions like a high number of pedestrians.
The inside of the case had a barrier and sealing foam to prevent light from bouncing across internally and affecting sensitivity. The same barrier technique is used in infrared distance and obstacle sensors.
Oddly, the lower LEDs had a film of translucent residue on them that looked like it had precipitated out from the adhesive light-blocking foam or from moisture condensation. With this type of optical system the clarity of the lens is important. It wiped off easily with isopropanol restoring the clarity to the lenses. The fact it only affected a specific layer of LEDs made me wonder if it was a deliberate dulling spray for the closer range beams. But I think it was just residue.
I specifically avoid using intrusive mid-run video adverts, which means I don't earn as much from my content as other YouTubers.
If you enjoy these videos you can help support the channel by subscribing and/or with a dollar or two for coffee, cookies and random gadgets for disassembly at:-
https://www.patreon.com/bigclive
Patreon supporters get early access to videos as they are made, and also regular live streams.
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators
It uses two fresnel lenses behind a sheet of black infrared filter material to fire out a matrix of 30 infrared dots in the same way as a disco LED moonflower light. It then measures the amount of IR reflected back from each beam and uses a PIC microcontroller to analyse any fluctuations caused by someone moving in front of the unit.
On further investigation it looks as though the pairs of 1 ohm resistors in parallel may actually be current sense resistors to regulate the current through each set of LEDs to around 1A. It also looks as though there is a resistor and capacitor in series with each microcontroller control line for the LEDs to make sure that a glitch or crash can't leave the LEDs powered continuously. Although each set of three LEDs has its own current regulator (for wider voltage-range use?) the microcontroller only uses five lines to switch them as groups of six.
I'm not sure what condition causes the relay to operate as I didn't get it to click a single time while I was playing with the unit. It may be faulty or just require a specific sequence of events to indicate a level of activity or if there's been a gap in the traffic and someone has crossed early.
The software in the PIC microcontroller must be quite complex to analyse fluctuations in ambient light and when something is causing false triggering of one beam repeatedly. I'd guess it also processes the input to make a relay trigger decision based on pedestrian activity and time.
The unit is designed to run on 24V AC or DC and the relay looks like it may be providing a volt free contact to the control system. The other relay position is probably for another function like fault indication or maybe to indicate other detected conditions like a high number of pedestrians.
The inside of the case had a barrier and sealing foam to prevent light from bouncing across internally and affecting sensitivity. The same barrier technique is used in infrared distance and obstacle sensors.
Oddly, the lower LEDs had a film of translucent residue on them that looked like it had precipitated out from the adhesive light-blocking foam or from moisture condensation. With this type of optical system the clarity of the lens is important. It wiped off easily with isopropanol restoring the clarity to the lenses. The fact it only affected a specific layer of LEDs made me wonder if it was a deliberate dulling spray for the closer range beams. But I think it was just residue.
I specifically avoid using intrusive mid-run video adverts, which means I don't earn as much from my content as other YouTubers.
If you enjoy these videos you can help support the channel by subscribing and/or with a dollar or two for coffee, cookies and random gadgets for disassembly at:-
https://www.patreon.com/bigclive
Patreon supporters get early access to videos as they are made, and also regular live streams.
This also keeps the channel independent of YouTube's advertising algorithms allowing it to be a bit more dangerous and naughty.
#ElectronicsCreators
Let's take a look at a pedestrian detector from a pedestrian crossing and the point of this device is that it monitors for people standing in the vicinity of the crossing to tell if they've gone ahead and walked. I presume i guess probably it cancels the call if it doesn't detect that someone's uh continuing to stand there. I'd like to thank peter for sending this uh, it's huge, it's quite hard to fit under the bench. It's taken me a while to get around to it.
That's one of the reasons, but here it is, and it's very interesting. So the first thing i'm going to do is i'm going to lever out this, which is normally screwed in, and this is a infrared filter. It just looks like black plastic, but it does pass infrared, which is quite important for this. I shall put that out the way.
Then we've got two fresnel lenses in it and in the case we've got a divider and there's lots of effort is made with this black foam to make sure that no light can sneak around uh to actually get from the emitter side, because this actually fires a Multiple of beams out one lens and then detects the beams with the other lens. It focuses them into the same area. It's quite complex. I shall put that down there and show you the circuit board.
So on one side we have infrared leds. This unit is powered by 24 volts. If i were to power up now, i don't think much would really happen. I shall power up, though, because uh then then you'll know what it looks like powered up.
It goes through an initialization procedure. It's microcontroller based uh powering it up and it's drawing about 50 milliamps at 24 volts at this point now you won't be able to see this at all you can you can't really see those you can see a bit of glimmer from these leds glowing, but they Are putting out quite strong pulses of infrared light and they're actually configured there's a five by six leds, but they're wired in series multiples of three, and it is presumably strobing through them very quickly, um at very high current, because it's putting three leds across 24 volts. But with a resistor in series, well, a couple of resistors, which are one ohm each. So that's half an ohm in the series, and these are only dropping about six volts.
So that is quite a lot of current um. So it is pulsing through in stripes, with three leds at a time and there's a a reason for that. The reason it's pulsing with a high current is because they don't need to be lit all the time. They only need to be lit when the microcontroller is actually looking at the corresponding sensor, and also it keeps the dissipation down um if they're pulsing and the uh very high power means that they are putting out a very bright beam.
Even though it is just a brief pulse and that's all the microcontroller needs it's turning the leds on it's looking at what it's getting back and then it's computing based on that. So not a lot happening, but there is a little red led and if you actually sit this unit for a while it stabilizes and then that led will detect. When you have a disturbance in front it, just with a circuit board down its own, it will detect the hand approaching it and the little red led here will light um. But if you've got the lens on then from a significant distance, it will detect you going into one of those zones, because if you recall the cheap disco, moon flower lights, that use the same system, they've got a matrix of leds and then the fresnel lens. And what happens is because each of these leds has effectively already got one lens in the front. The fresnel lens, then adds an extra sort of focusing effect and what you end up with is 30 distinct beams firing out this unit, and it's very interesting never seen this done before that. It then harvests. Those beams actually focuses these onto the same areas that that is pointing at very, very interesting.
So let's look at look at the circuitry. There is quite a a bunch of words going on to this, but in reality it's got an earth wire which goes onto the case and then into the housing. It's got the uh red and black here, which are for the power supply. It can be either ac.
24, volts, ac or dc, and then i think it's yellow and blue are the output back from the relay, but the other ones aren't used they're there, because there is the provision for another relay in here. But it's only using one and i have to say i've played with this for a while. I've not managed that really to activate. But having said that, i think, there's i think, there's a calibration time.
This has to go through to rule out ambient factors um or maybe that's why it's out of use. Maybe the really had failed. Although i was getting a resistance across its winding um, i should have tacked. I did stick a meter across to and then left the room and disturbed it every so off just to see if it would send a cancellation call not sure exactly what these do.
Do they just pulse that really to say the pedestrians left cancel the call, or should that really activate every time it detects movement uh, as in that led uh lighting up, maybe if something's going on the circuit or maybe the relay has physically welded inside i didn't, Should have put a resistor an led across it across the call that would have shown, but anyway here is the circuitry down. Here is a pic microcontroller. It is, and i noted this down - uh i'd noted it down. Then i've just left my notes somewhere.
One moment plays, i have my notes. The pic maker controller is a pic16c66, not a terribly modern microcontroller, and it is effectively controlling these transistors via these other buffer transistors, to actually switch these sections of leds. Then, when it's actually enabled a group of leds, it'll then probably scan the three infrared sensors and to do that. It's got chips over here, which are adg 526 akr, and they are analog multiplexers that can actually switch 16 analog inputs to one output, and that is uh probably going over that. Well, it is it's basically selecting one uh channeling, that across to the uh pic microcontroller, and it will be just putting into effect with a potential divider, so it can actually measure how much light is coming back onto this. Then it's all down to software. Something incredible must be good on this chip, because it's looking at the analog levels that are being received by these uh things, i'm going to zoom down this for you, so you can examine it closer. I may even brighten the image just a little bit.
That's better! I shall prop something under it: there we go. That's uh, that's going to make it more manageable, so you can explore it. So the microcontroller is looking at those and it's obviously storing them in memory and it's looking for significant fluctuation. But i wonder if it can also detect rogue inputs.
That say, for instance, this one is detecting little bursts of uh reflection, because there's something an environmental faction that vicinity is causing that, and maybe it can say right, i'm not going to give priority to this one um and it will have to calibrate all the time To the fact that there is some level of infrared light that will get through this lens from sunlight, so it has to compensate all the time and it will just be varying those and then looking for a predominant change when pedestrians walk in front of it and They cross into the different zones. It probably also relies on the fact that it is detecting it moving between zones, because nobody ever seems to stand still at pedestrian crossing. If you see one of these, don't stand still move about in front of it. It will probably help detect you other than that it has what you'd expect for a something like this.
The incoming supply has lots of filter, metal oxide varistors. Then it's got. What i presume is a common mode suppression, choke bridge rectifier and then the smoothing capacitor and the fact it says that it does ac and dc that would charge up to much higher voltage, let's bring in the kinked calculator. What would that charge up to with 24 volts ac 24 volts times 1.41? The peak voltage charge up to is about 33, say 34 volts, that's quite high.
Is it and it does appear to be uh driving that relay directly from that with its 24 volt winding, but so i'm guessing that it's relying the fact that it's just operated in brief pulses. Other things worthy of note here, not really a huge amount. We've got these pairs of one ohm resistors in parallel for that half ohm the transistor, that's doing the switching for the leds. They've got this distinct pattern here.
I wonder if that's just to compensate the fact that it's looking down the way and it is effectively spreading out so it uh effectively, because it's at an angle this is basically keystone correction. Then there are a couple of voltage regulators. Well, i think i've got voltage. Regulators, certainly it would make sense to have well. It needs five volts for that. It's a very low current microcontroller at four megahertz. It only draws about two milliamps, which is quite good. I wonder um if this was providing a higher voltage, then for the analog circuitry is perhaps uh processing some of the uh, the intensity, control that being detected, the variation of the resistance to send a distinct voltage back to there.
The microcontroller might not be getting the signal from here directly. It might be going to these first to actually apply a bit of pre-filtering before it comes over here, but that is it very interesting. It's i wonder, which came first, these ones these units and i've seen them around for a long time. Are the disco led moon flowers because they use the same approach.
They use the leds, the clearly five millimeter leds, with the same, focusing through a fresnel lens, just to create a splat of lots of beams of light out of the front of it. That's what this thing's doing, but an infrared, but it's actually detecting the bounce light from whatever it hits coming back very clever, very complex. The circuit board itself has just single-sided uh construction, quite a nice design of circuit board. The only thing that's on this side is that the led, which sticks down underneath the unit and just lights when it detects movement in the vicinity very interesting.
The surface mount here, the through-hole components, it's just a very interesting design. I really wonder what went into the software, because the software is ultimately what is going to make this unit quite special. It's going to determine its ability to differentiate between um, random ambient fluctuations or disturbances, a piece of litter blowing in front and disturbing one of the sensors continually. It may just learn that and then just learn to ignore it briefly until it's gone very clever, very interesting, not what i was expecting inside at all well worth opening up.
So i'd like to thank peter again for sending that sorry for taking so long to get round to opening it up and exploring it, but it was actually well worth it. It's a fascinating device.
I have previously pronounced the ‘s’ Fresnel, perhaps we need another pole like the one for grease 😂. Although Monsieur Fresnel is not around to respond since he died in 1827, pity. I will have to remember when I next say the word that the ‘s’ is silent 🇫🇷
Lol, I was literally looking at one of these about an hour ago. The traffic lights were taking ages to change and I wondered if it was working.
Somehow, I feel that you would probably assemble a pelican crossing in your house at somepoint, just because of the fun of having such a thing that most people just take for granted… 😛
This is a no cost too great design, the antithesis of just about everything coming out of China. Those Schaffner filters alone are not cheap. Built to work in tough environments and made to last…
So, it sounds like this is basically a somewhat upgraded, but very large and clunky, PIR detector. I was thinking it might be able to detect pedestrians in the crosswalk, and extend the light until the crosswalk actually clears, but it doesn't seem this is anywhere near that sophisticated.
I think it could be part of a puffin crossing. There is no fixed time and no flashing green man. Traffic remains stopped until the sensor detects no more pedestrians on the crossing.
Nice device, high quality and interesting pattern of LEDs and photodiodes. Seemingly irregular, but this irregularity almost certainly serves some purpose.
That was very interesting! I've seen these things next to pedestrian lights in NL and always wondered what it was. I think there's also sonar ones!!
I've been working on projectors and cameras for over 30 years, and it seems I've been pronouncing that wrong all this time ? Is the 's' silent in Fresnel ? Learn something new every day 🙂
I remember that we had a unit just like this (or highly similar at least) in the late 70s as part of a very high-end home alarm system.
The chips might not have been a part of it, might all have been done with discrete parts though…
All the crossings where I live appear to have "press the button" to activate the crossing.
There are some traffic lights that use a microwave head to detect traffic approaching the lights to activate the change of traffic lights.
Somewhere, I have a microwave Doppler head that is used for the traffic lights.
If I can find the microwave Doppler head, I'll see if it is worth sending for "investigation".
K Watt.
That makes me want to replace the emitter LEDs with alternating red, yellow, green, blue, then set it up with its lens and film it at high speed.
If that circuit board was designed in 1997 and those chips made in 2001 (which matches the look of that overall design) then it definitely predates LED moonflowers as 5mm colored LEDs which project didn't exist back then – the best you had was some dim indicators. If that particular unit saw any real use then those LEDs are most likely shot – even pulsed, IR LEDs don't last that long, especially if you run high current through them.
Very interesting discrete implementation of this idea, I love the non-square grid layout of the emitter and sensor diodes, presumably it's designed to be placed at a very specific height and angle to the crossing.
Those remind me of all the railway crossing protection equipment in Japan, those sorts of IR units are common along with simple beam break, LiDAR, radar and visual cameras designed to detect pedestrians, cars and trucks in the crossing when the gates are down. If they detect something they trigger warning lights well before the crossing and approaching trains immediately stop or continue slowly. With how often I see news stories of trains hitting high centered trucks on US railway crossing I don't understand why this technology is never implemented there, although I have seen a couple of those on UK rail crossings but I don't believe there is any requirement.
Wonder how it does differentiate between a person/car/pigeon/rubbish. I can't see them sticking in a object recognition chip into something so low cost/low power