Still have false trigger issues with the Parallax PIR sensor....

deno

Well, I bought 2 more of the above sensors from Parallax, on the indication that others were having better results with a new design?· No such luck on this end with either sensor.· Both will continuesly produce false triggers.· With no heat source around, and after hours of 'on' time, the units will trigger randomly.· It seems the only way I can produce satisfactory results is to put the device in the refrigerator.· When I open the door I will get a trigger. (This is with a meter attached to the output and the test leads routed out the side of the door so I can monitor the device while the door is closed)· I guess they are only effective in much lower temps then 70 to 80 degree range.

I am running the PIR sensor with 5 volts...has anyone had better results with 3.3 volts?

Any thoughts on this...?

Deno

allanlane5

"after hours of 'on' time the unit will trigger randomly" is not usually considered the same as "continuously produce false triggers". If you want to solve a problem, defining it well is important.

I've found issues with other sensors, where putting in a 10 Kohm pull-up reduced their tendency to "false trigger". Now, the "other sensors" here are usually IR-Decoder chips, which "pull low" when they detect a signal.

I haven't used the PIR sensors myself, so I don't know if they're "active Low" or what kind of signal they put out when triggered. Can you do some "noise" filtering in your code? Is there a "bouncy" signal? Is their signal "active low" or "active high"?

Ugha

I've had no problems with the PIR sensors from parallax... are you giving them a few seconds to calibrate to the ambient
room temp?

Are you following the instructions listed in the documentation to the letter?

Allanlane5 has a good point, it is often helpful to use a pullup/pulldown on sensors to avoid false triggering.

allanlane5

Do you have the "jumper" in the "H" position? If so, how many checks do you make that the signal is "High" before your program uses the signal?

DosManDan

I've used the PIR sensors in Halloween props and had no problem with detection. I didn't do anything special except for move the jumper on the PIR. I didn't use resistors, but that does seem like a good way to avoid false input. Definitely let them warm up for a minute or so, after that, it's pretty stable.

Dan

deno

I have the jumper in the 'low' position, as I want a 'high' pulse produced each time there is a detection.· Previously, I said the unit produces false triggers when there are no trigger sources around.· The units(both of them)·'continuously produce false triggers at random intervals' meaning that some times (15 to 20 seconds) the output will be low indicating 'no' dectection, then for a period of time the units will trigger 'high' and 'low'· repeatly indicating several detections of 'something?' when there is NO 'something to detect.· There is no motion of any heat source, as I am outside the room looking thru a window.

A pull up resistor for a high going pulse won't work, and a pull down resistor still produces the above results.

To use the stamp to void out a certain number of false hits will defeat the purpose of alarm detection, as the stamp may void out the one 'detection' that is important.

DosManDan, exactly how do you have your PIR sensor wired up for your Halloween props, and where do you live, meaning, how cold is it outside where you live during Halloween?

Like I said, units work good inside a refrigerator.

Deno

LoopyByteloose

Well, if they work well in the 'low going to high on trigger' only, you could add a bit of logic in the form of an inverter. One with hystersis might even be better.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:

Do you want a quickly operational black box solution or the knowledge included therein?······

···················· Tropically,····· G. Herzog [noparse][[/noparse]·黃鶴 ]·in Taiwan

Ugha

deno: I live in Virginia where its sweltering hot in the summer and freezing in the winter... PIR seems to work either way when
given a proper 30-second to a minute "warmup" time.

During the first minute of operation it should be as absolutely still as possible in the sensor's field of "vision".

Also remember that this thing can see a WIDE range... pretty much if you can see the white of the lens, then it can detect
you... you may need to narrow the angle some way (the parallax site for the PIR has an outside link that has an interesting
suggestion or two)

Please note that I've only used the PIR in short-term applications so my information may be incorrect.

Have you tried the sensor with the jumper in the L position? Does it make a difference at all?

deno

Well...I think I got to the bottom of my false triggering of the PIR sensor from Parallax.· It seems if you 'hang' to much load on the output of the device, it will false trigger.· Not only was I interfacing the output to the Basic Stamp, I was also lighting a bright 'blue' LED for a visual trigger situation.· Remove the LED or exchanging it for a lesser current draw LED, or increasing the value of the current limiting resistor will allow the device to work properly.· Thought this might be of some interest to others.

deno

PJAllen

Did you have the LED/s connected to the Ping when you were "testing" it in the refrigerator?

deno

Yes, I did have the same LED config hooked to the PIR sensor.·Maybe the colder temp changed the·devices operational habits? ·I could't locate the maximum current draw allowed on the output of the PIR sensor in the data provided.· As soon as I pull the LED out of the breadboard, and only interface with the Stamp, I get the expected results from the device.

Now, I use the trigger (output) straight to the Stamp input, and to the base of a NPN, which inturns lights the LED when there is movement.

Just wanted to clear the air...

Deno

allanlane5

Thank you for the update to clear the air, that was very kind of you. I'm sure it would have occurred to very few people that you might have hung an LED (with resistor) across the signal pin. It's nice to know this is possible, with judicious selection of LED and resistor. Or better yet, as you have done with a transistor.

Dr. Resonance

Sometimes, if the run from PIR to the Stamp is more than a few feet, you may have to use a shielded cable for a 10-30 ft run.· This will catch and ground out any electromagnetic waves produced by (example) an air conditioning motor (high current) turn-on that will produce a EM pulse.· As this pulse passes your unshielded cable, a potential is produced in the wires (Maxwell's equations) which can interfere and produce erractic results.

Dr. Resonance

·