Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Noam]

But, I'd also be very interested in knowing why it works better than other IR-LED sources, because that is not logically apparent to me.

How do the motion sensors work? Do they emit IR, and look for reflected pulses, or do they look for radiated energy coming off an object?
Perhaps pointing an IR light source at them isn't workng, because it is a different frequency than the detector is looking to see bounced back.

If they look for changed in a detected heat pattern, perhaps mounting a hair dryer on a stand, and aiming it at the detector might create a "burst" of heat that the detector might pick up.
I'm imagine there is probably a way to then put that stand on wheels (perhaps an A/V tripod, or something low-tech like a rolling office chair), and move it slowly into the field of view of the detector, to see where it triggers?

[MD Corie]

But, I'd also be very interested in knowing why it works better than other IR-LED sources, because that is not logically apparent to me.

How do the motion sensors work? Do they emit IR, and look for reflected pulses, or do they look for radiated energy coming off an object?
Perhaps pointing an IR light source at them isn't workng, because it is a different frequency than the detector is looking to see bounced back.

That's a good question!  My understanding (which people seem to feel is generally "defective") is that the sensors are "passive" and only look at changes in the ambient IR "background" - in which case, any strong IR source ought to get a reaction out of them.

If they look for changed in a detected heat pattern, perhaps mounting a hair dryer on a stand, and aiming it at the detector might create a "burst" of heat that the detector might pick up.
I'm imagine there is probably a way to then put that stand on wheels (perhaps an A/V tripod, or something low-tech like a rolling office chair), and move it slowly into the field of view of the detector, to see where it triggers?

That hair dryer idea is intriguing.  That's one potential IR source that I hadn't thought of - and it's quick and easy to do with on-hand household items (so it meets my criteria ).  I'll give that a try and see what happens.  Thanks!

Unfortunately, it won't really test the stationary on/off scenario - because the hair dryer will have some heat-up/cool-down hysteresis, so it wouldn't work well as a "snap switch" like an IR-LED source should... but it certainly should be appropriate for the moving-source scenario.

[Noam]

...My understanding (which people seem to feel is generally "defective") is that the sensors are "passive" and only look at changes in the ambient IR "background" - in which case, any strong IR source ought to get a reaction out of them.

As long as the frequency of your IR source is in a range that it can detect.

That hair dryer idea is intriguing.  That's one potential IR source that I hadn't thought of - and it's quick and easy to do with on-hand household items (so it meets my criteria ).  I'll give that a try and see what happens.  Thanks!

Unfortunately, it won't really test the stationary on/off scenario - because the hair dryer will have some heat-up/cool-down hysteresis, so it wouldn't work well as a "snap switch" like an IR-LED source should... but it certainly should be appropriate for the moving-source scenario.

Perhaps you could block it with some heat-resistant material? Like one of those silicone baking dishes?
Another option might be to leave the heat source stationary, and rotate the motion sensor around it. Using a protractor, you *should* be able to get an idea of the field of view it has.
Then, you could mount it and estimate the edges of the field, and test them again by moving the source.

[MD Corie]

...My understanding (which people seem to feel is generally "defective") is that the sensors are "passive" and only look at changes in the ambient IR "background" - in which case, any strong IR source ought to get a reaction out of them.

As long as the frequency of your IR source is in a range that it can detect.

Another good question!  Unfortunately, I don't recall ever seeing the wavelength specs anywhere.  (Anybody know?)

... you could mount it and estimate the edges of the field, and test them again by moving the source.

That's essentially the point that I'm at in the process;  that is, I've mounted the sensors and am trying to define the areas they can actually "see" (with masks, as needed), so I can make sure there is no overlap (hopefully), and in some cases, to make them non-reactive to irrelevent motion that occurs in nearby off-premises areas.

[Noam]

That's essentially the point that I'm at in the process;  that is, I've mounted the sensors and am trying to define the areas they can actually "see" (with masks, as needed), so I can make sure there is no overlap (hopefully), and in some cases, to make them non-reactive to irrelevent motion that occurs in nearby off-premises areas.

I'm guessing that the "no overlap" is so that you don't have two sensors starting up the same macro?
I know your macro setup is already pretty complex (by AHP / CM15A standards, at least), but I suppose you could use flags to prevent double-triggering of the macros. Of course, with only 16 flags available, you might not have enough to cover all your macro scenarios.

[MD Corie]

I'm guessing that the "no overlap" is so that you don't have two sensors starting up the same macro?

Actually, I'm wanting no overlap so that the system can determine which camera to turn on, based on where the motion is detected.  (Specifically, it is to determine which individual macro should be triggered, which, in turn, activates the appropriate camera - among other actions).

... I know your macro setup is already pretty complex (by AHP / CM15A standards, at least), but I suppose you could use flags to prevent double-triggering of the macros. Of course, with only 16 flags available, you might not have enough to cover all your macro scenarios.

The response macros are triggered by individual H-U codes from each sensor, and are prevented from re-triggering by conditions set for the monitored H-U codes of lights, each of which is controlled by the respective macro.  So there is no problem of double-triggering the sensor-response macros.

[HA Man]

I hesitate to post suggestions because I am no expert and only have some experience to offer.  Also, I think what I have to suggest may be obvious anyway, but please bear with me because my suggestion has raised some questions in my mind that maybe someone can answer.

I have observed that my sensors seem to trigger easily from cars and trucks on a nearby road, but only when it is relatively dark.  My thinking is the sensors detect the headlights or maybe even the running lights of the cars and trucks, so I was going to suggest trying an automotive light bulb to trigger the sensors.

But then I realized something that seems contradictory, and so it turns my suggestion into a question instead.  The contradiction is that even though my sensors seem to react well to the traffic on the nearby road, they often do not respond to cars in my own driveway, which runs nearly parallel to the road, but about 30' closer to the sensors.  I do not understand this.  Can anyone explain why this happens?  It seems like lights in the driveway, being closer to the sensors, would appear more intense and also give more horizontal movement, which is what triggers the sensors as I understand things.

Anyway, I have two questions:

Do the motion sensors actually react to automotive lights, or are they triggering on something else?

Why do the sensors trigger much better on cars and trucks that are on the road, which is further away, than they do for cars and trucks in my driveway, which is closer to the sensors?  My driveway averages about 35' away from the two motion sensors, and the road is at least another 30' beyond my driveway.  I believe even the driveway is beyond the specified range of the sensors anyway, and the road is almost twice as far, so things do not make sense to me.

PS:  I really do not want the sensors to detect the traffic on the road, but if they detect traffic in my driveway, that would be OK.  Any suggestions how I could get the sensors to give that desired result?  Thanks!

[MD Corie]

Anyway, I have two questions:

Do the motion sensors actually react to automotive lights, or are they triggering on something else?

Concensus says that I don't know much either, but I'll pass along my experience regarding this, FWIW:

I don't think they respond to the automotive lights... at least, mine do not seem particularly responsive to light bulbs.  I once tried to find ways to detect cars as they were approaching my house.  In that process, as far as I could tell, the sensors respond to the cars even less at night when their lights are on.  So, my guess is that the sensors are oblivious to car lights.

Why do the sensors trigger much better on cars and trucks that are on the road, which is further away, than they do for cars and trucks in my driveway, which is closer to the sensors?  My driveway averages about 35' away from the two motion sensors, and the road is at least another 30' beyond my driveway.  I believe even the driveway is beyond the specified range of the sensors anyway, and the road is almost twice as far, so things do not make sense to me.

Unfortunately, I suspect that is due to one of the quirks of the motion sensors.  I have seen cases where nothing at all beyond 25' would cause a sensor to trigger... but I've also seen cases where sensors trigger fairly reliably on things that are 100'-150' away.  Apparently, there is considerable variability in the response of the individual sensors.  So, if you have other sensors, you might try swapping them to see if it changes the result, or maybe changing their aiming angle (if possible) - although, I don't know why they would miss nearby stuff if they pick up stuff that is farther away.

PS:  I really do not want the sensors to detect the traffic on the road, but if they detect traffic in my driveway, that would be OK.  Any suggestions how I could get the sensors to give that desired result?  Thanks!

Are you using the X10 floodlight sensors?  If so, try adjusting the sensitivity ("range") setting, or change the aiming angle (downwards).  If you're using other sensors, see if there is any comparable adjustment(s) that you can tweek.  One extreme option - that probably isn't practical - might be to put up some sort of wall or obstable between your driveway and the road... although others here have claimed that the sensors can "see" around obstacles!  How that works is beyond my understanding, though.

It may be that you're just running into the extreme quirkiness of the motion sensors, though... so there may not be any straighforward explanation... nor any easy fix.

[IPS]

MD Curie
When I said, in my last post that, " I covered the sensor with a Hard black plastic"'. I meant that I put apiece of hard black plastic on the sensor like a sun visor. I think that has limited the exposure from the sunlight.

HA Man

"Why do the sensors trigger much better on cars and trucks that are on the road, which is further away, than they do for cars and trucks in my driveway, which is closer to the sensors?  My driveway averages about 35' away from the two motion sensors, and the road is at least another 30' beyond my driveway.  I believe even the driveway is beyond the specified range of the sensors anyway, and the road is almost twice as far, so things do not make sense to me".


I think the reason trucks/ cars on the road are able to trigger more easily is because, may be your sensor is at right angle to the road. Sensors react better when motion is across than towards them.

IPS

[Brian H]

They do detect better when the motion is across their field horizontally than directly towards them.
http://kbase.x10.com/wiki/Optimizing_Motion_Sensor_Detection

[HA Man]

I think the reason trucks/ cars on the road are able to trigger more easily is because, may be your sensor is at right angle to the road. Sensors react better when motion is across than towards them.

IPS

AND

They do detect better when the motion is across their field horizontally than directly towards them.
http://kbase.x10.com/wiki/Optimizing_Motion_Sensor_Detection

Sorry if I wasn't clear about it, but my driveway runs nearly parallel to the road, and across the front (so to speak) of my house.  I have a motion sensor on each corner of the house.  The two motion sensors that are on the front corners are partially exposed to the driveway and to the road.  One reacts a little more to the road traffic than the other one does, but they respond to the traffic about the same overall.

Are you using the X10 floodlight sensors?  If so, try adjusting the sensitivity ("range") setting, or change the aiming angle (downwards).  If you're using other sensors, see if there is any comparable adjustment(s) that you can tweek.  One extreme option - that probably isn't practical - might be to put up some sort of wall or obstable between your driveway and the road... although others here have claimed that the sensors can "see" around obstacles!  How that works is beyond my understanding, though.

I do not think it would be practical to put up a wall between the road and the driveway, and the zoning people probably would not allow it anyway.

I am using the X10 PR511 sensors.  I have tried adjusting the range control, but I am not sure what it really does.  It does not seem to eliminate the road traffic without eliminating ALL triggers.

By the way, I can not figure out how to have the motion sensors trigger at all times but turn on the floodlights only at night.  They have a photocell in them for dusk/dawn operation but I do not see how to set them so the floodlights will turn on only at night when they are needed, at least not without disabling the motion sensors during the day too.  If there really is no way to set this, it seems to be a glaring omission because it makes them waste all that energy during the day when it the lights are not needed.  Am I missing something?

[dave w]

  If there really is no way to set this, it seems to be a glaring omission because it makes them waste all that energy during the day when it the lights are not needed.  Am I missing something?

No, sadly you are not.
Do you have AHP for home automation?

[HA Man]

  If there really is no way to set this, it seems to be a glaring omission because it makes them waste all that energy during the day when it the lights are not needed.  Am I missing something?

No, sadly you are not.
Do you have AHP for home automation?

Yes.  But I gave up on it because it seems to be squirrelly.  Is there something in AHP to suppress the floodlights?  I thought the floodlight operation was all local to the PR511s themselves. (?)

[Dan Lawrence]

  If there really is no way to set this, it seems to be a glaring omission because it makes them waste all that energy during the day when it the lights are not needed.  Am I missing something?

No, sadly you are not.
Do you have AHP for home automation?

Yes.  But I gave up on it because it seems to be squirrelly.  Is there something in AHP to suppress the floodlights?  I thought the floodlight operation was all local to the PR511s themselves. (?)

What version of AHP did you have?   Also, the PRS511 Module in AHP only turns it on or off, it does not support timers. I use a Pro Wall Switch Appliance module in AHP to control the PRS511. 

[HA Dave]

Just put a screw-in photocell controller between the floodlight bulb and it's holder. The power will come on, but the bulb won't be turned on unless it is dark. Or if you're compelled to use X10... use SocketRockets (and a complex series of macros).