Maker Pro
Maker Pro

Alarm reset timer

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi Everyone,

I am after some help on circuit design for a problem I have with an alarm I have put together and wonder if anyone could point me in the right direction i.e. there maybe something out there already? However, if nothing off the shelf, or adaptable I am up for building something?

I have a mains supply into a PIR controller, which switches a seperate alarm bell. The reason for the help is when the mains supply to the PIR is lost for whatever reason, when the power supply is re-established, the PIR initially switchs until it resets itself and therefore the alarm sounds, which is what I want to avoid. Both PIR and alarm are 240V mains.

Hope this makes sense

Regards

Mark

Somerset, UK
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
You could connect the alarm via a time delay relay with the delay greater than the PIR reset time. How long does the alarm sound?

Time delay relays are available or you could make a delay circuit to switch a standard relay.
 

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi,

Thanks for the reply.

It would need to be greater > 15 seconds.

I'm not bothered if there was a very short output from the alarm on power up, due to a relay latching etc, I just don't want it ringing for the initial 15 secs after initial circuit powers up!

Thanks

Mark
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Personally I would hack something onto the circuit board of the PIR device itself. It already has the low-voltage supply and the relay. It would only take a few components.

The problem is that we would have to trace out at least part of the circuit in order to know how to modify it. If you're up for it, post some good in-focus photos of both sides of the PIR board. The best illumination is natural light but not direct sunlight.
 

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi,

Thanks for the reply, I will attempt to get some photos up over the next couple of days.

Regards

Mark
 

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi,

Photos as requested. Let me know if you need any more?

Thanks again,

Regards

Mark
 

Attachments

  • PIR1.jpg
    PIR1.jpg
    147 KB · Views: 379
  • PIR2.jpg
    PIR2.jpg
    144.7 KB · Views: 337

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
That's good, thanks.

I need to know the markings on the IC on the underside at the left, marked U1.

If there's some kind of goop covering it, try cleaning it with a cotton swab dipped into detergent diluted in hot water. You may need a magnifying glass and/or a very strong light to see it.

If you can get a good photo of it, that would be great. Otherwise, just post the markings as text. There will probably be a manufacturer's logo but that's not important.
 

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi

Okay here goes,

text on the top row

LM324DG. This is probably what you are after looking on the internet quickly.

Second row

ON, assume this is the OEM, then to the RHS there is an almost indistinguishable line of text even with a high mag glass which looks like PSCB45

Thanks as always.

Regards

Mark
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Sorry for the slow response. I've been wondering how to respond.

The circuit and the PCB seem to be very well designed. The LM324 is a very common IC, as you've probably discovered. The relay is driven directly from one of its outputs, on pin 14. The main supply rail is +24V (the relay has a 24V coil).

The problem is that the output that drives the relay also feeds some other circuitry. It seems to connect to the LDR (light-dependent resistor - marked CDS on the board) and the trimpot nearby. So, intercepting the control signal to the relay coil might affect the rest of the circuitry. To be sure that the mod would work properly, I would have to reverse-engineer the whole circuit.

I can draw up a modification that MAY work, and you could try it - it would involve five resistors, two small MOSFETs, one diode and one electrolytic capacitor. Or thereabouts. Would there be room to fit that many components in the enclosure? You would need to make the circuit up on a small piece of stripboard, cut one track on the main board, and connect the stripboard to the main board with three wires.

If you can link to a user manual for the PIR unit, I may be able to understand what the LDR does, and be able to say more definitely whether a modification will work.

Alternatively, if you can spare it for a week or two, you could send it to me here in New Zealand. That's a long way away, and you would need to pay for postage both ways.

I'm sorry that a hack isn't as straightforward as I expected.

An external delay box is totally possible but would probably be more work to build than a hack. I doubt there's an off-the-shelf solution, but you could Google words like power delay relay.

Or as Duke suggested, you could use a normal time delay relay, if you can tolerate a delay in the alarm activating on a genuine alarm condition.

Edit: I can probably simplify the circuit down to one resistor, one electrolytic capacitor, one transistor, and two diodes...
 
Last edited:

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I had a look for time delay relays and the ones I saw were horribly expensive.
As to Kris's suggestion that the alarm would be delayed, this could be obviated by pulling in the relay with a time delay after the power comes on and running it continuously.

The modification that Kris suggests would be much neater.
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
As to Kris's suggestion that the alarm would be delayed, this could be obviated by pulling in the relay with a time delay after the power comes on and running it continuously.
Ah yes, good thinking. Power the delayed relay unconditionally, and connect its contact in series with the contact from the PIR.
 

CDRIVE

Hauling 10' pipe on a Trek Shift3
May 8, 2012
4,960
Joined
May 8, 2012
Messages
4,960
I had a look for time delay relays and the ones I saw were horribly expensive.
As to Kris's suggestion that the alarm would be delayed, this could be obviated by pulling in the relay with a time delay after the power comes on and running it continuously.

The modification that Kris suggests would be much neater.

Ah yes, good thinking. Power the delayed relay unconditionally, and connect its contact in series with the contact from the PIR.

My Florida Cracker friend Jessie always says "Damn! Sometimes I'm so smart I can hardly stand myself"!!!! :p

Good call Duke! ;)

Chris
 

duke37

Jan 9, 2011
5,364
Joined
Jan 9, 2011
Messages
5,364
I've had another thought, that is the second this week.

With a LDR in the circuit, perhaps the detector is intended to be inhibitied in daylight. The LDR will be low resistance in daylight and it could be tried with the LDR shorted. If this works to stop the alarm on switch on, then Kris's modification could be added there.

Alert
Gold Star Award ;)

attachment.php
 

Attachments

  • Gold-Star.jpg
    Gold-Star.jpg
    12.6 KB · Views: 370
Last edited by a moderator:

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi All,

It really doesn't matter to me if its mounted in the PIR housing or remotely outside.

Happy to give anything a go, although pic taken of the PCB in the housing attached

Thanks all.

Regards
 

Attachments

  • PIR3.jpg
    PIR3.jpg
    136.1 KB · Views: 176

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
Without a lot more information on the PIR circuit, I can't suggest a reliable hack.

Here is a circuit that will close a relay after about 20 seconds.

attachment.php


This circuit is powered directly from the mains, and should all be considered LIVE. It must be enclosed and insulated so that no part of the circuit can make contact with any conductive component, especially anything that could be touched by a person.

The resistor named RELAY_COIL at the right side represents the coil of a relay. This relay will close after about 20 seconds after power is applied to the circuit. The relay must have a 24V DC coil with a resistance of about 1600 ohms (i.e. a coil current of about 15 mA).

In your application, this circuit would be connected directly to the mains supply and the contact of this relay coil would be wired in series with the contact from the PIR device. (This assumes that the alarm circuit is normally open; for security reasons, sometimes normally closed circuits are used for alarm signals.)

All of the components in this design are available from Digikey. If you want to build it, let me know, and I will post a list of links to appropriate components on the Digikey site.

There are a few features of this design that may interest other Electronics Point users.

The input bridge rectifier uses two normal diodes and two zener diodes. This provides full-wave rectification and simultaneously clamps the positive rail to the zener voltage. This clever idea actually comes from your PIR circuit; I didn't think of it myself. It avoids the separate zener and therefore saves space.

A circuit like this would normally include Schmitt trigger behaviour, so that there would be a clean, well-defined change from the OFF state to the ON state. This design doesn't have explicit positive feedback, but still has this behaviour, because as soon as current starts to flow in the relay coil, the positive rail voltage starts to drop, and this is coupled through the timing capacitor back to the input of the voltage detector (Q1), reinforcing the change from OFF to ON.

The delay is set by C3 and R4 and is about 23 seconds. This will vary somewhat depending on the Vgs threshold of Q1. R4 could be replaced by a 1M trimpot if desired.

I'm not sure whether using small diodes in the bridge is a good idea; they could be susceptible to damage from spikes on the mains. The standard arrangement, using 1N400x diodes and a separate zener with an electrolytic across it, would be harder to damage because the 1N400x diodes can handle large brief spikes and the electrolytic would help to protect the zener. This circuit relies on the fusible input resistor to absorb the energy from any mains spikes. I would like to hear what other forum members think of this matter.
 

Attachments

  • Power-up delay.gif
    Power-up delay.gif
    9.1 KB · Views: 489

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi Kris,

Yes I'm keen to give this a go, so would appreciate the links at Digikey when you get the chance. I'm away for a few days so no rush.

Thanks

Mark
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
OK. I've updated the schematic to include some more information, and some component value changes that I hope will improve its handling of mains voltage spikes.

attachment.php


Here's a list of components on Digikey. I'm not sure where you are. If you're not in America, you may find Digikey's shipping charges are a bit crazy, and you may want to look at Farnell (now Element 14 I think) or RS Components, or some other supplier.

R1 (1k 0.5W fusible) http://www.digikey.com/product-detail/en/NFR25H0001001JR500/PPC1.0KBCT-ND/614242
R2,3 (220k high-voltage) http://www.digikey.com/product-detail/en/RNV14FAL220K/RNV14FAL220KCT-ND/2665274
R4 (1M trimpot) http://www.digikey.com/product-detail/en/CT6EW105/CT6EW105-ND/738336
R5 (22k) http://www.digikey.com/product-detail/en/RNMF14FTC22K0/S22KCACT-ND/2617471
C1 (0.22 uF X2) http://www.digikey.com/product-detail/en/R463I322000M2M/399-5902-ND/2571337 or http://www.digikey.com/product-detail/en/B32922C3224M/495-2320-ND/778982
C2 (100 uF 35V) http://www.digikey.com/product-detail/en/UVR1V101MED/493-1081-ND/588822
C3 (22 uF 35V) http://www.digikey.com/product-detail/en/UPW1V220MDD/493-1853-ND/589594
D1,2,7 (1N914) http://www.digikey.com/product-detail/en/1N914ATR/1N914ACT-ND/1626125
D3,4 (1N4749A) http://www.digikey.com/product-detail/en/1N4749A/1N4749AFS-ND/977295
D5,6 (1N5245B) http://www.digikey.com/product-detail/en/1N5245BTR/1N5245BFSCT-ND/1992220
Q1,2 (2N7000 or BS170) http://www.digikey.com/product-detail/en/2N7000TA/2N7000TACT-ND/3042479 or http://www.digikey.com/product-detail/en/BS170/BS170-ND/244280
Relay (24V 2880R coil) http://www.digikey.com/product-detail/en/G5NB-1A-E DC24/Z2773-ND/1731473 (SPST) or http://www.digikey.com/product-detail/en/V23105A5305A201/PB388-ND/291024 (DPDT)

According to my quick mental additions, the total component cost is USD 7.32.
You'll also need some stripboard to mount the components on, and a mains plug and some kind of connector for the relay contact(s).

The Digikey pages have links to the data sheets so you can get the pin allocation of the MOSFETs.
 

Attachments

  • Power-on delay.gif
    Power-on delay.gif
    15.5 KB · Views: 612
Last edited:

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Thanks Kris,

This will take me a while to get everything together.

I'm in the UK, so as you suggest Farnell/RS will be my fist stop.

Regards
 

KrisBlueNZ

Sadly passed away in 2015
Nov 28, 2011
8,393
Joined
Nov 28, 2011
Messages
8,393
OK, cool.

You may want to add your location to your profile. It will appear in your posts. This is helpful to us, because it tells us which site we should refer you to, when suggesting components.
 

caferacer

Aug 11, 2013
17
Joined
Aug 11, 2013
Messages
17
Hi Kris,

All on order.

Is there any zero volt testing that should be done on the circuit prior to initial power up?

Regards
 
Top