AP16+ - PIR False Triggering

[Traveler]

Hello All,

I have read the post "Using a PIR Sensor" ( http://www.efx-tek.com/php/smf/index.php?topic=16.msg340#msg340 ) but searching the forum I cannot find a post about just using the PIR with the AP16+ unit. I am not trying to do anything outlandish. I just want the PIR to detect motion and then play some random audio files from the AP16+. Problem is I am getting quite a bit of false triggering. I have also referred to the Scary Terry page ( http://www.scary-terry.com/itw/pirsensor/pirsensor.htm ) and I'm using a PVC pipe to narrow the field of exposure to motion. I have the pir/pvc-pipe pointed at a blank wall but it is amazing the amount of false triggering I am getting. How do I stop that from happening? Will that bit of JonnyMac code work on the AP16+ ?

Other than the false triggering problem I've happy with what i'm hearing and how well the randomizer cycles through the SFX files.

Thank you,

Traveler

[JonnyMac]

How are you defining "false triggering?"  What are the conditions the PIR is being subjected to?  They are very tricky beasts, and I would suspect that this time of year -- when it's hot -- they are particularly twitchy. 

I'm out of town at the moment but will look at extending the debounce when I get back.  We're using the same algorithm as on the EZ-8 and have had no complaints there, so I suspect that it's conditions based.  Also, you should set the POST delay pot when using any kind of PIR or user triggering to prevent immediate re-triggering.

[Traveler]

Hello JonnyMac,

Well what I mean by "False Triggering" is the unit starts all on its own for no apparent reason. The conditions the PIR is subjected to is an air conditioned room in my house. Thermostat set to 78 degrees with a thermometer in that room registering 74 degrees. The unit it set at table height and is arranged as in the picture with the PIR inside the PVC pipe. The PIR/PVC-Pipe is pointed at basically a blank wall 6 feet away, although I have also pointed it out into the room where there are other stationary object besides the walls. (The results seem to be the same no matter where the PIR is pointed) I have sent everyone outside so there cannot and will not be any movement in the room at all.

I then turn on the unit while standing to the side with a stopwatch in hand. The unit cycles through playing a 30 second audio clip while the PIR sensor calibrates itself. I then intentionally activate the unit at which it plays another 30 second audio clip. Then I wait for the solid green status light to come on and I start the stopwatch while standing still to the side completely out of the line of sight of the PIR sensor. The results vary but I have had it spontaneously start playing an audio clip at varying time intervals. 0:22, 2:14, 3:49 are just a few examples.

Then I set the post-play delay potentiometer. That doesn't really seem to have any fine tuning capability I want and it doesn't seem to make any really difference with the problem I am having) It seems impossible to get the delay down to anything like just a few seconds. I'd be happy with a 3-5 second delay but the shortest amount of time that I've been able to set that to is 13 seconds. Anyway, I get the same results with the post-play delay potentiometer set to 13-20 seconds as I do with it set to zero. I am still getting "false triggering" (Maybe false triggering isn't the correct term? Maybe it should be spontaneously triggering).

I have setup two  of the six AP16+ units I bought like this and I am getting spontaneous activation with both. I also noticed that with the post-play delay potentiometer set the unit will not always activate when it should like when I walk in front of it when the status light is green. Several times I've walked past the PIR and it will just make the status light flash between red and green without actually playing an audio clip. If I could get the potentiometer adjusted down to just a couple of seconds that wouldn't be a problems, but it is a nuisance if the delay time is 10 seconds or more.

So what can I do, what do I need to do, to make the units less "twitchy"

Thank you for your time,

Traveler






.

[JonnyMac]

Do cap the end of the pipe facing the AP-16+ -- the regulators amps put off enough heat that it could be affecting the board.   Again, I'm out of town at the moment but will look into extending the input debounce time to counter "twitchy" PIRs.

[Traveler]

JonnyMac,

I sealed off the end with cardboard and tape and it didn't have any effect on the problem. I should also note that the photo was only showing one layout version I tried with the components. I also laid this out with the PIR/PVC-Pipe stretched as far away from the AP16+ as the cable would allow with the same results as before.

Thanks for trying to figure it out.

Traveler

[JackMan]

 I am assuming you have a Parallax PIR? You may have a bad PIR. Make sure the jumper on the PIR is set to HI. I hooked up a Parallax PIR to my AP-16+ with no choke tube and it works perfect, no false triggers after the initial PIR warm up.

I also noticed that with the post-play delay potentiometer set the unit will not always activate when it should like when I walk in front of it when the status light is green. Several times I've walked past the PIR and it will just make the status light flash between red and green without actually playing an audio clip.

I'm a little confused with this statement. The LED should flash red/green while the post delay is in progress (no audio can be triggered until the delay is over and the LED returns to solid green. If your PIR is not triggering the audio with a solid green status LED then there is something wrong with the PIR.

[Traveler]

Jackman,

Yes I am using Parallax PIRs. If I was only using one AP16+ and PIR setup I might agree with you about it just being a bad PIR, but I have two setup and I am experiencing similar false triggering with both so I think it is something more than that.

Thanks for the advice on the PIR jumper being set to high. I started with it there, then went to low to see if that was any better, and then I went back to the high jumper setting.

It was recommended to me by EFX-TEK support that I use some kind of pvc "choke" tube similar to the one found at Scary Terry, and besides that I need to be able to restrict some of the angle area the PIR sensor monitors because of the nature of the project. So doing without the restrictor tube is not an option. I need to be able to find a way to make it work. I have switched to a more open restrictor tube setup that has a much broader exposure angle area, but that doesn't seem to have made any significant difference to the problem.

I'm a little confused with this statement. The LED should flash red/green while the post delay is in progress (no audio can be triggered until the delay is over and the LED returns to solid green. If your PIR is not triggering the audio with a solid green status LED then there is something wrong with the PIR.

I'm confused too, but I don't know how you can come to the conclusion that it must be the PIR. What changed in the equation was the additional use of the post-play delay potentiometer. The PIR activates normally when I walk in front of it otherwise with no delay set. I might add this doesn't happen all the time, but its happened more than twice when I had a post delay set. As I mentioned before I just can't seem to be able to adjust the post-play delay potentiometer the way I would like to. If it really is a necessity to use it with the PIR I'd like to know how to get it down to just a few seconds delay. I just can't get it down below 13 seconds and that is too long of a delay for the way I want things to work.

Thanks for your input JackMan. Every bit helps,

Traveler

[JonnyMac]

The resolution on the POST play pot is a little coarse because we need to be able to do fairly-long delays.  The START button and PIR input connect to the same circuit, so if your board behaves fine when pressing the START button (and PIR disconnected) I'm going to go with Jack on the bad PIR.  They're low-cost devices and Parallax has had troubles with them in the past.

[JackMan]

It was recommended to me by EFX-TEK support that I use some kind of pvc "choke" tube similar to the one found at Scary Terry, and besides that I need to be able to restrict some of the angle area the PIR sensor monitors because of the nature of the project. So doing without the restrictor tube is not an option.

I wasn't suggesting that you need to use the PIR without any type of choke tube, I was merely trying to tell you that mine works when the PIR is completely exposed which makes it even touchier. As far as the post play setting goes, yes, it does seem that the lowest setting is about 15 seconds (or 0 sec. if turned fully counter clockwise). Typically if you need a post play delay it is usually around 60 seconds, not sure why you are looking for just a 3-5 second post delay. Considering you are having false triggers with two PIR's, and two AP-16+'s are you sure they are connected properly? While unlikely, it's not entirely out of the question that you have two bad PIR's. All I can say for sure is that my PIR and AP-16+ works fine. Do you have a Prop-1 or some other device that you can check the PIR's with?

[Traveler]

This maybe a case where there were 2 different things causing the false triggering problem I have been encountering. After both JackMan and JonnyMac expressed their opinion that the problem was a faulty PIR sensor this morning I swapped out a third one I had with the one setup that was particularly twitchy. After allowing time for the sensor to adjust I intentionally triggered it and then started my stopwatch after the audio had stopped playing. When I reached 10 minutes (plus) of it not spontaneously triggering I swapped the PIRs between the 2 units. The second unit went a similar 10 minutes without any false triggering. Just to be sure that I had removed a bad PIR I reinstalled the suspect one and sure enough I started getting false trigger when I installed it on either AP16+.

So that explained why one unit might start on it's own, but not both. With the known bad PIR removed from the mix I started switching the 2 other PIRs between the 2 AP16+ boards. With both of them turned on at the same time this morning neither one was spontaneously triggering. I swapped them back and forth several times and still no problem. That had me really puzzled but then the central air came on and low and behold one of them triggered on its own. I turned off the AC and the false triggering stopped. Turn it on and it started again. So there does also seem to be an environmental element contributing to the triggering problems. JonnyMac had mentioned that might be the cause, but with the bad PIR also causing troubles it was hard to diagnose that.

Since it may not always be possible to turn off an air conditioner or a furnace are there any ways to reduce that effect on a PIR? Would extending the input debounce time counter the effects of air flow on the PIR enough to prevent that particular twitchy behavior?

Thanks again for helping me work through the problem(s) I have encountered with false triggering.

Traveler

[JonnyMac]

Not likely; we face these difficult environmental things all the time -- the key is to change the environment.  PIRs work from a change in IR (heat), you have to be able to control that.  Now, these a DIRT CHEAP PIRs; one possible fix is to change to a commercial unit; they'll cost a bit more but are designed to deal with more obnoxious environment.  Some of them are tunable.

[Traveler]

Now, these a DIRT CHEAP PIRs; one possible fix is to change to a commercial unit; they'll cost a bit more but are designed to deal with more obnoxious environment.  Some of them are tunable.

JonnyMac,

I'll try to work with the Parallax PIR, but are there any commercial units that EFX-TEK has tested with the AP16+. Or are there any suggestions from any users that have tried something other than the Parallax?

Thanks,

Traveler

[JonnyMac]

No, as you can imagine there are too many possibilities for us to burden ourselves with.  We only carry the Parallax PIR as a convenience for our customers who want a low-cost PIR.

On a related note about environment... I'm out in the desert at a friend's home and just noticed the installation of a PIR that is part of the alarm system (see photo).  Note that it is not up high where heat gets trapped; even commercial PIRs are sensitive to the environment, though less sensitive than cheapies.

[Traveler]

. . . We only carry the Parallax PIR as a convenience for our customers who want a low-cost PIR . . . Note that it is not up high where heat gets trapped; even commercial PIRs are sensitive to the environment, though less sensitive than cheapies.

OK Thanks for the advice. My original plans for this were to contain the AP16+ and motion sensor in a base unit that wouldn't be more than 18" high anyway, so the height issue shouldn't be a problem.

Traveler

[JackMan]

Glad to hear that you found the problem with the PIR's. Going back to your post play delay issue, I don't know your reason for needing just a 3 - 5 second post delay but one way to get that would be to edit your .WAV files and add 3 - 5 seconds of silence to the end of them.