Posted: Monday Apr 21st, 2008 01:16 pm |
|
1st Post |
Pgordon
Member
back to top
|
Hi all.
over the weekend I added a new PIR to my system to increase the coverage near the bottom of the stairs. It seems to work fine, I did a walk test, and it covers the area I want...
Because it is downstairs (although it covers the bottom 3rd of the staircase) I set it as zone type: InstNightAwayNC (5) and set it as YES to Entrypath, - exactly the same as other downstairs PIRs, so that in night mode it should start the entry sequence.
However, during the night, when moving around upstairs whilst armed to night mode, the PIR caused an intruder alarm. There are two troubling aspects to this, firstly, I cannot concieve how it could possibly have seen any movement on the upstairs landing, - there absolutely is no line of sight from the sensor covering the areas in which there was movement, and secondly, even if it is somehow sensing movement upstairs, why did it immediately go into full alarm rather than starting the entry delay?
Q: Are there any other settings I should be making apart from the zone type & entry setting?
Q: is is even conceivable that a PIR could pick up reflected heat signals from an opposite wall? - this is the only way I can imagine the sensor could have detected "movement"
Here's the section o fthe event log covering these events....
04/21 00:35 Mode Change #Night Mode
04/21 01:32 Alarm Type #IntruderAlarmAud (1)
04/21 01:32 Zone Activation #Stairs_PIR (12)
04/21 01:32 Dial Number #, 3
04/21 01:32 Alarm Type #IntruderAlarmAud (1)
04/21 01:32 Zone Activation #Hall_PIR (3)
04/21 01:32 Alarm Type #EntryAlert (10)
04/21 01:32 Alarm Type #Abort (20)
04/21 01:32 Sign in - User #, 2
04/21 01:32 Mode Change #Security Off
Many thanks
Paul G.
|
Posted: Monday Apr 21st, 2008 01:29 pm |
|
2nd Post |
Pgordon
Member
back to top
|
Actually, I think I may have answered (one of) my own questions...
Upon closer inspection of the PIR responses, It does appear that I have the following code assigned to the ON response of the other PIR in the same area:
If SecurityMode = NightMode Then
Start EntryDelay
End If
Which I'd obviously forgotten about!. - I'll assign the same response to this new PIR, so hopefully that will sort out the entry delay vs. instant alarm issue. That just leaves the problem of the sensor detecting movement outside its field of view in the first place... When I did the walk test when initially setting it up & positioning it, it was definately only trigerring on the bottom 3rd of the staircase, so quite how & why it was even detecting at all are a mystery to me.... - more testing required methinks...
Cheers.
Paul G.
|
Posted: Monday Apr 21st, 2008 02:41 pm |
|
3rd Post |
admin
Administrator
Joined: | Saturday Mar 3rd, 2007 |
Location: | Singapore |
Posts: | 1200 |
Status: |
Offline
|
back to top
|
I was just going to point out the entry alert action but you figured it out first
PIRs often behave unexpectedly, I have found so good luck
|
Posted: Monday Apr 21st, 2008 09:21 pm |
|
4th Post |
Pgordon
Member
back to top
|
Hmmm.... This is *darn* peculiar... what I have established pretty much beyond doubt is that the problem is not that the staircase PIR is seeing movement outside of its field of view, rather, that PIR is apparently "following" one of the upstairs PIRs - the one in the bathroom.
I've established this by doing the following:
I assigned a response to the new staircase PIR that turns on a light near the bottom of the stairs, and done a walk test in front of it. I can see the PIR activation LED and the light whilst performing the walk test. The PIR absolutely only triggers upon movement in the bottom portion of the staircase. over about half way up the stairs is outside of the view of that PIR & it does not trigger.
Turned on security check mode to hear the zones announcing as I moved around upstairs. As soon as the bathroom PIR triggers, it announces the bathroom movement, followed *immediately* by an announcement of staircase movement. It does this consistently and repeatably, - whenever the bathroom PIR triggers, without fail the system announces both zones, always bathroom first, followed by staircase.. When this happens, the staircase PIR does not *actually* trigger, and the response to turn on the light runs inconsistently - sometimes the light comes on, sometimes it does not. (I can see the light from the bathroom too). - I haven't established this for sure, but it seemed to depend on whether the system was in security check mode or not - are zone responses disabled in security check mode?
I can repeat this test with the bathroom door closed, - in this case the bathroom PIR can't "see" me moving about on the landing, and thus doesn't trigger when I walk around upstairs. In this circumstance, I can move all around the landing, right to the top of the stairs, and even get halfway down the stairs before the staircase PIR really does see me, and then behaves as I would expect.
The converse does not occur, - triggering the staircase PIR announces only the staircase movement as I would expect, - it does not also announce any other zone.
What on earth could cause this behaviour? - this never used to happen before, triggering the bathroom PIR has previously had no effect on other zones. I've not made any changes to the bathroom zone type or responses as part of this excercise, but now for some reason *every* time the bathroom zone triggers, the staircase zone apparently follows it.
The bathroom PIR is set to zone type NightModeDisable (18), and has no responses assigned to either the zone on or off event. Both sensors use DEOL resistors. The two *are* on adjacent zones though, - bathroom is zone 11, staircase is zone 12, could there be some kind of interference going on? - I'll double check the wiring tomorrow, as I could imagine that a stray strand of wire could bridge the zones & possibly cause this behaviour, but to be honest, I was pretty careful when I did the wiring, and I don't think I'll find anything wrong there. - Plus surely if I'd mistakenly bridged the zones then surely triggering either zone should trigger both? - so I should have heard the bathroom zone announced when activating the staircase zone?
Can you think of any other possible causes of this behaviour? For the time being, I've temporarily changed the zone type of the staircase PIR to SwitchNC(9), which hopefully will prevent it causing an entry delay every time the bathroom is used in night mode, but this isn't a proper fix of the problem.
Cheers.
Paul G.
|
Posted: Tuesday Apr 22nd, 2008 02:31 pm |
|
5th Post |
admin
Administrator
Joined: | Saturday Mar 3rd, 2007 |
Location: | Singapore |
Posts: | 1200 |
Status: |
Offline
|
back to top
|
It would be worthwhile to test if Comfort itself is causing this behaviour
Remove the two PIRs from the zones and set to NO EOL
Connect two wires in the zones instead and simulate the bathroom and staircase PIRs by removing the wires, see if the same thing happens
if not then you can focus on the PIRs or the connections
By the way. good to have met you at the NEC last week
|
Posted: Tuesday Apr 22nd, 2008 03:32 pm |
|
6th Post |
Pgordon
Member
back to top
|
Thanks Chiu. It might be a little while before I have a chance to do this testing. - this sort of work is normally reserved for weekends; when I have more time available, and am less likely to upset the neighbours with alarms going off! - however, I'm away from home this coming weekend, so it may well be nearly 2 weeks before I can fit this in.
The change to zone type SwitchNC(9) has, for now, stopped the problem. - It means that new PIR effectively isn't part of the security system, but that isn't a big problem since I installed it mainly to increase the sensitivity of a lighting response. - I'd still like to get to the bottom if it though!
Regarding setting the zone(s) to no EOL I remember this is a jumper on the main board, but I can't remember if it is set individually per zone, or if a single shunt sets EOL or NO EOL for zones 1 through 8 in tandem... I guess I need it to be set per zone so I don't have an untoward effect on the other sensors?
Shame the show at the NEC was quite small this year... :-( But if anything comes out of it, - you getting together with the Sensible Heat guys to come up with a Hometronic UCM would make it all worthwhile! :-))
Cheers.
Paul G.
|
Posted: Wednesday Apr 23rd, 2008 05:41 am |
|
7th Post |
admin
Administrator
Joined: | Saturday Mar 3rd, 2007 |
Location: | Singapore |
Posts: | 1200 |
Status: |
Offline
|
back to top
|
The EOL settings are per zone. There is a 3 position shunt at each zone for you to set it
|
Posted: Wednesday Apr 23rd, 2008 06:44 am |
|
8th Post |
Pgordon
Member
back to top
|
Ah! - there's a thought, - since the zone was previously unused, I would imagine it's likely that it was set to NO EOL by default?... - I definately didn't think to check/change the EOL shunt when I installed the new PIR (which *does* have both EOL resistors fitted), so potentially the sensor might be operating in NO EOL shunt mode, but with EOL resistors fitted... - do you think this could (would) cause the odd behaviour I'm observing?
Cheers.
Paul G.
|
Posted: Wednesday Apr 23rd, 2008 06:54 am |
|
9th Post |
Pgordon
Member
back to top
|
UPDATE: I just changed the EOL shunt for the new zone, went into security check mode, and now the bathroom does *not* announce both zones, so that would *appear* to have been the problem... :-)
Well I never would have thought of that! - there's one lesson learned that won't happen again! - REMEMBER TO CHECK THE EOL SHUNTS!
Thanks for the yip-off Chiu. I think it's problem solved.
Paul G.
|
Posted: Wednesday Apr 23rd, 2008 06:55 am |
|
10th Post |
Pgordon
Member
back to top
|
Pgordon wrote:
Thanks for the yip-off Chiu. I think it's problem solved.
That should *of course* read "TIP-off"....
|
|