Dollar store window / door alarm - Hacking question

[caphair]

Hi guys this alarm uses a reed switch to turn on a siren when a door or window is opened.

I'm wondering how I could tweak it so even after the magnet lines back up (door / window closed) the alarm keeps sounding for a certain amount of time around 5-10 seconds.

Here's the circuit I couldn't find anything on the ic used

[caphair]

Re: Dollar store window / door alarm - Hacking question

Anyone? Can I simply solder in a capacitor across the piezoelectric alarm terminals or?

[joshnz]

Re: Dollar store window / door alarm - Hacking question

could put a high value cap across reed switch.

[caphair]

Re: Dollar store window / door alarm - Hacking question

It would have to be across the reed switch? How would that work

[joshnz]

Re: Dollar store window / door alarm - Hacking question

The switch opens when magnet is away letting voltage rise which then is buffered to drive the piezoelectric drive transformer. so it will slow voltage rise taking longer for alarm to activate. unless reed switch is normally closed type.

[caphair]

Re: Dollar store window / door alarm - Hacking question

Ah I want it to delay when the magnet is placed near the unit so if the door is closed the alarm still sounds for a bit.

Don't want an activation delay

[joshnz]

Re: Dollar store window / door alarm - Hacking question

sorry my bad.
whats the number on the 8 pin chip.
Also can you tell me what 1 2 4 5 in attachment are connected to.

[caphair]

Re: Dollar store window / door alarm - Hacking question

1 and 2 connect to one end of the power on toggle switch which the + of battery connects
4 goes to one end of the reed switch (which is normally closed operation)
5 seems to go to the other end of the power toggle switch

Number on the 8 pin ic is YL3710

[budm]

Re: Dollar store window / door alarm - Hacking question

Pin 5 of the IC looks to be the Trigger pin. Check the DCV at pin 5 of the IC with ref to pin 8 of the IC to see if it goes HIGH when the alarm is triggered and LOW when alarm is not ON, or another way around.
That IC boosts the Battery Voltage to higher level for driving the Piezoelectric transducer.
https://www.badcaps.net/forum/attach...2&d=1473604683

[caphair]

Re: Dollar store window / door alarm - Hacking question

Pin 5 of the IC looks to be the Trigger pin. Check the DCV at pin 5 of the IC with ref to pin 8 of the IC to see if it goes HIGH when the alarm is triggered and LOW when alarm is not ON, or another way around.
That IC boosts the Battery Voltage to higher level for driving the Piezoelectric transducer.
https://www.badcaps.net/forum/attach...2&d=1473604683

+ probe on pin 5 and - probe on pin 8 reads 0v when not triggered. Move magnet to trigger alarm the voltage jumps only to .70v

Probing + on pin 8 and - to ground/negative of battery reads battery voltage 4.8v when not triggered. When magnet moved and triggered on voltage slowly drops off pin 8

[budm]

Re: Dollar store window / door alarm - Hacking question

'Probing + on pin 8 and - to ground/negative of battery reads battery voltage 4.8v when not triggered. When magnet moved and triggered on voltage slowly drops off pin 8' Tahat does not make sense since in the picture is marked pin 8 with (-) the yellow wire is soldered to, so that (-) marked is NOT the negative of the batteries? Unless the picture from post 7 is marked wrong.
https://www.badcaps.net/forum/attach...2&d=1473604683

[caphair]

Re: Dollar store window / door alarm - Hacking question

Oh oops that picture confused me I used the 5 mark as meaning it was the 5 pin of ic.

Ok so remeasuring: pin 5 goes high when alarm triggered. Low when not

[budm]

Re: Dollar store window / door alarm - Hacking question

So the pin 5 of the IC which is connected to the middle pin of the Transistor it goes high when the alarm is triggered, how high?

[caphair]

Re: Dollar store window / door alarm - Hacking question

Went high to battery voltage 4.8v

[caphair]

Re: Dollar store window / door alarm - Hacking question

Anyone?

[eccerr0r]

Re: Dollar store window / door alarm - Hacking question

NECRO ALERT!

There are two 1M resistors, one has a capacitor connected in parallel. Call the one in parallel with capacitor R1 and the other one R2.

The two options are:
1. Increase R2 to a higher resistance.
2. Add a resistor in parallel to R2.

Adding a resistor in parallel to R2 is easier but will reduce battery life. A resistor lower than 470K resistor will probably make the alarm never shut off.

Likewise increasing R1 past about 4.7M ohms and the alarm will never shut off.

Personally though I'd like hacking things like this, but this I'd not hack to do what you want to do, very dicey to get it to behave the way you want it to behave. I'd start over from scratch and make the whole thing anew.

[caphair]

Re: Dollar store window / door alarm - Hacking question

NECRO ALERT!

There are two 1M resistors, one has a capacitor connected in parallel. Call the one in parallel with capacitor R1 and the other one R2.

The two options are:
1. Increase R2 to a higher resistance.
2. Add a resistor in parallel to R2.

Adding a resistor in parallel to R2 is easier but will reduce battery life. A resistor lower than 470K resistor will probably make the alarm never shut off.

Likewise increasing R1 past about 4.7M ohms and the alarm will never shut off.

Personally though I'd like hacking things like this, but this I'd not hack to do what you want to do, very dicey to get it to behave the way you want it to behave. I'd start over from scratch and make the whole thing anew.

What about changing the capacitor for a different value? Any effect?
I'm guessing the resistors and capacitor create a time constant?

[eccerr0r]

Re: Dollar store window / door alarm - Hacking question

You can do that too, but you need to make sure you use very low leakage capacitors. The main reason for not suggesting it is because I don't know what's the current capacitance.

[budm]

Re: Dollar store window / door alarm - Hacking question

YL3710 Schematic.

[eccerr0r]

Re: Dollar store window / door alarm - Hacking question

Actually I think you may need to put in a bigger capacitor there after all. I've never bought one of those units so I don't know what is behavior is like, but likely it's close to instant on/off implying that capacitor to being in the single digit nanofarads or lower. Adding several hundred nF to 1 uF of capacitance will keep the alarm on longer and would also take longer to shut off, of which you would decrease the resistance of R2 to speed up shutoff (and increase power consumption.)

Again a low leakage capacitor is mandatory, most electrolytics will fail.