Post about your Basic project here

Moderator: Mmiscool

User avatar
By heckler
#68077 Nice work Electroguard!

Did you ground the metal dish to the power supply gnd?

Does the dish/vegetable steamer prevent any detection from the back side so that it is only detecting from the front??

regards
dwight
User avatar
By Electroguard
#68090 To answer some questions...

The metal dish does cast a big rear sensor shadow as you'd expect, but I still wasn't able to creep up to it from behind - or any direction - without it noticing.

Obviously there is a trade-off between range and beam width.
If the sensors are at a slight angle to each other (Marty Feldman gaze) it provides a broader beam of reduced range, whereas max range is obtained when both sensors are square on to the target.
I expect max range and minimum beam width would be acheived by mounting the sensor(s) further out towards the effective focal point of the dish - but bear in mind that it's only a blanching dish, not a true parabola.

Besides, next I'm going to try using just a flat piece of metal as a reflector - which should give a wider sensor spread, reflecting everything forwards and diagonally outwards without being focused into a tight beam.

The physical range testing is a matter of connecting to the esp AP with a mobile phone, then diagonally zig-zagging away until the phone shows the sensor has lost signal, then zig-zagging sideways to discover the beam width. Sounds simple enough, but this dual sensor lash-up works so well that I found I was often moving out of esp wifi range (up to 50m) before moving out of sensor range.

So although I did try to see if earthing the electronics to the dish made any difference, I'm afraid there was no definitive results because the sensor range was already too good to be able to notice any improvement before losing esp wifi range.
That's good enough for me.

I only originally bought 5 of these sensors to evaluate, but I've already re-ordered a load more.
When the new lot arrive I plan to mount several to the same flat reflector - a couple close to the centre (as here) plus others at 3 o'clock and 9 o'clock facing diagonally out towards the side, and perhaps others to fill any perceived weak spots.

My thinking is that it may be possible to make something like a 'flower' sensor array capable of monitoring an area the size of a tennis court - if the range or sensitivity is too excessive in any particular direction then I can add metal screening if necessary... even if it makes it look a bit like a mailbox. A fun thing to try, anyway.

EDIT:
Incidentally, when time and weather permit, I'll try mounting a sensor inside an empty tin can to see if it's possible to make a very tight narrow 'spotlight' type beam specifically for aiming at a target of interest.
User avatar
By Electroguard
#68128 Some more experimentation results...

All the following tests were done with a single sensor - which for comparison has a 'bare' open-air non-reflected range of 10m.

The dish arrangement shown in the left of the first picture gave about 25m range with about 2 to 3m beam width.

Image
Mounting the sensor further out towards a perceived focal point, as shown on the right, actually reduced range by a couple of meters but gave a wider 4 or 5m beam width.

Grounding the electronics to the dish seemed to improve range slightly (about 1m).

Image
Closing up the dish petals like an iris narrowed the beam greatly (to 1m) and reduced range to about 12m.

Image
Mounting the sensor inside a shortish flat-ended metal cylinder gave a range of about 15m with a beam width of 2 or 3m.
Mounting the sensor on a flat reflector gave a forward range of at least 15m with a very wide fan-shaped detection beam about as wide as long (which is nackering to map out, I might add).