10 Ultrasonic on Mega 2560, one cable to connect all

Hi,
I am planning stair lightnings with ultrasonic sensor HC-SRO4 with Mega 2560

Logic would be simple that when distance is less than particular range says less than 120 CM then relay light will be turned on vice versa.

Programming is simple, however question arises that is it possible to do one cable to run 10 ultrasonic in one cable, this means than there would be 20 wires for echo and Trig. For power I am planning another cable carrying +5 Volts.

I am concerned about noise which could interfere among different devices and length which is approx 9 Meters.

Please share experience with long and multiple cables carrying different signals.

Why do you need 10 sensors?
Why do you need ultrasonics?

First thoughts are you would just need 1-2 PIR sensors mounted midway.

Yes, but I you can make interesting effects when you know where the persion is on the stair. e.g add a ghost comming down the stairs etc.

But I advise against ultrasonics. Even when you do not hear them, permanent ultrasonic noise kills your audio perseption - just go ask a dentist (the turbine of the drill emits ultrasonic).

@Slumpert
Yes, 10 Ultrasonic

@zwieblum, I didnt get you, do you mean emitting ultrasonic sensor will effect human ear in long run ?

A 9 metre cable is a bit of an ask.

The recommendation here would be to place one or two Pro Minis (or Nanos) adjacent to the ultrasonic modules. Two Nanos would look after five sensors each so there would be no trouble about enough input pins. A Mega 2560 is then unnecessary; your master controller can then be another Nano.

The two Nanos can then do all the fussy stuff controlling the HC-SRO4 sensors and would simply relay the information back over serial links. They can be "chained" so that one passes the information to the next, then on to the "master".

You do have concerns still passing information over nine metres. And power. Consider using RS-522 modules - balanced communication over two wires - and send the power as 12 V with a switchmode "buck" regulator at the Nanos. Total, four wires or use "cat 5".

I say "Nanos" as easier to type. Pro Minis do not include the USB interface, superfluous except for re-programming. Not having it connected makes connecting to the serial pins easier.

saleemsadruddin:
@zwieblum, I didnt get you, do you mean emitting ultrasonic sensor will effect human ear in long run ?

More urban mythology I fancy. In any case, the sound levels and directions involved with these sensors would be very low.

So many ultrasound sensors in close proximity means you also have to worry about cross talk between sensors. There must be enough time in between pinging two sensors to not have them hear an echo of another.

They're also pretty bad at detecting clothed people. IR break beam sensors may work better.

saleemsadruddin:
@zwieblum, I didnt get you, do you mean emitting ultrasonic sensor will effect human ear in long run ?

Yes, it will ... you'll get deaf. But it's always a question dose. And if you have pets, you could end up with quite unhappy pets.

You could use pressure pads on the treads - digital outputs and don’t need power.
If you have a lot of ultrasonic sensors , then they may have problems setting each other off .
I’d be looking at IR beam sensors if the pads won’t work.

The dB output of one these hobby units is undocumented, but being that max current is 20 mA @ 5 volts this would be equivalent to a 1/10 watt speaker. Not something I would suggest sticking in your ear all day, but beyond that direct range doubtful there is enough energy to wear on your ears. Now commercial units running in the 10s-100s of watts is a different matter.

It’s like looking into a IR laser, just because you can’t see it doesn’t mean it won’t hurt you.

IR Beam is not possible as I have access to only 1 side of stair i.e next to wall. Other side of stair is open which means I cannot do anything over there.

How about using IR sensor, which would work best in my situation?

Use IR proximity sensors, such as the Sharp GP2y0d805z0f (has a few versions with different range - analog output) or the LV53L0X (IR time of flight sensor; I2C output).

I ordered LV53L0X lets see outcome.

Anyone can suggest good way for human presence?

PIR sensors work great for detecting humans.

PIR also detects your dog or cat. It's detecting any moving thing that's warmer than ambient. PIR also only detects humans (or squirrels, or a passing car) when they're moving.

How accurately do you want it to be? Filter out other non-human warm bodies? Detect only moving humans or also stationery ones?

Hi,
Problem is that is detecting good. But scenario is different.

Lights on when someone enter room - (good with PIR)
Light to remain on when someone in room (not good with PIR if person goes sleep or stationary for long time)

Hi,
would appreciate comment on above.

If you do a search on this forum you'll quickly come across suggestions (believe me, you're not the first that is looking at this) and realise why this is in the Pretty Much Impossible category.

The fact that there are no readily available sensors for "human present in room" is a major clue to that extent as well.

I loved the original Star Trek episode where they kicked everyone off the ship and then used sound sensors to detect heartbeats to find out if someone was hiding in a Star Ship...

Even in a futuristic Sci-FI series, detecting a human in a room was quite a challenge.