Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Alarm system from scratch

Status
Not open for further replies.

zachtheterrible

Active Member
I do believe that there is someone that has been trying to get in my garage. The garage door lock looks like it has been fiddled with, and im pretty sure i heard some noises in there a while ago. Of course this gives me an excuse to build an alarm system :lol:

I know how to make the sensors, so that is no problem (although i might ask a few questions). Since it would be impractical to run wires throught the house, I was thinking of using the existing mains to transmit signals.

My basic requirement is that i be able to tell which alarm was triggered.
 
I would use 433mHz transmitter and receiver from Qkits, about $10 US each. Mouser has 433 mHz antenna for about $3. I got 1000 feet range using a parabolic reflector on the transmitter, I think 150 feet without the reflector would be possible. If range were a problem, a good AM radio with 433 mHz down converter would give plenty of range.

A 555 oscillators modulating the transmitter with different audio frequencies could be seperated at the receiver with a PIC or just use bandpass filters.

If you were to use 3 transmitters, hetrodyning at the receiver would be a problem when more than one was active.
 
The more that I think about it, it would be so much simpler to just use mains to transmit the signal instead of wireless. Can I use a 555 to transmit different audio tones over the mains without blowing it up :lol: ?
 
Sure, use a small value coupling cap to attenuate 60 (or 50) Hz and put a resistor in series to limit transient current (10K would be good). A 0.27 uF cap will have 589 ohms reactance at 1000 Hz and 9824 Ohms at 60 Hz.
 
Thanx russ, this sounds like it is going to be quite a bit simpler than I thought.

Question though: How would I transmit the signals over the mains using a 555 without blowing out the 555? Would I connect the signal to neutral and the ground off the 555 to mains ground? And then connect the signal in of my receiver circuit to neutral and ground of the reciver to mains ground?

Sure, use a small value coupling cap to attenuate 60 (or 50) Hz and put a resistor in series to limit transient current (10K would be good). A 0.27 uF cap will have 589 ohms reactance at 1000 Hz and 9824 Ohms at 60 Hz.

Your are talking about the receiver part of the alarm I presume?) I think that I will use a bandpass filter at this part so that I can actually design it. and its cheaper :lol:
 
If you want a smoke bomb conect it directly. (I once had fun whith an broked 555 timer i coneted to an 20V 10A mains transformer.First nothing and the sudenly POP and it made an about 20cm ball of smoke)

you wod need an prety good bypas filter since you have an 9V modulated signal and 220V mains signal.the mains signal must be completly filterd out!

and what about an electrcic load on the power line?It wod simply golp up your signal.

You may have seen those mains LAN acses points.They work on an radio signal that uses the mains to travel

Do be careful whith mains!
 
electro, you have raised a good question about I'll need some very good filtering. There is no electrical potential between neutral and ground, correct? So if I inject the signal between those two, no electrical load will eat up the signal, and I won't have to worry about frying anything. Am I missing something here??

Another question that I need to know is: How far apart should the audio signals be in terms of frequency? Is 1000hz 2000hz and 3000hz good enough?
 
zachtheterrible said:
electro, you have raised a good question about I'll need some very good filtering. There is no electrical potential between neutral and ground, correct?

Not true! - there is a potential between neutral and ground, dependent on the voltage drop across the current paths - usually it's a few volts.

Mains communication is usually done either ground/neutral or ground/live, neutral is obviously the safer option - but you must still treat it just as you would live, because the wiring may be incorrect!.
 
In the US, (in my experience anyway) neutral and ground are connected together at the main junction box, so you should measure a short circuit between them.

The power main is low impedance, so you should expect the signal to be attenuated by the isolation means, and require amplification at the receiving end. A state variable active filter will provide high Q, so seperating the frequencies will not be a problem. If you are using DSP that would be even easier.
 
What I really need to know is: Is it impossible or something to do this by mains? From what you said Russ, it would be impossible to do it using ground and neutral. EDIT: Confirmed, there is .05 ohms resistance between neutral and ground.

I came across a home automation system called X10 that would send out its signal during the zero crossing point. I think this is the best way to do it because there will be no interference during this point: I wont have to worry about any kind of active filter. So how do I do only send my signal during the zero crossing point?

I made a block diagram of what my system will basically look like.
 

Attachments

  • block_diagram.gif
    block_diagram.gif
    7.3 KB · Views: 1,589
The usual method is to use a low frequency RF carrier, which is what X10 does - and all the other mains transmission systems (both audio and data) that I've seen.
 
Zach
Nigel is right on target. Search for X-10 information. I have seen many schematics on the rf to mains interconnections. X-10 transmission over mains is restricted to the mains Zero-crossing interval as that is the "quiet time" best suited for data transmission. Your simple system could also use DTMF based transceivers. This would keep a single frequency detection (noise) from falsely triggering an alarm, and allow a unique sensor ID to be employed by the transmitted DTMF digit .
Just a few thoughts
Dialtone
 
Okay, so I'm definitely going to have to send my signal during zero cross.

1. The unwanted noise would have to be the right frequency (1khz, 2khz, or 3khz) in order to trip the alarm. Is that a legitimate concern? Or is it very unlikely?

2. DTMF is the dialtones on a phone, which is basically the same thing that I'm planning on doing: Sending audio tones through the mains. Why then should I use DTMF instead of audio tones generated by the 555? Using the 555 would allow me to have a unique alarm ID: alarm 1=1khz, alarm 2=2khz, alarm 3=3khz. I don't really care what the frequencies are btw, those are just examples.

3. Is there any advantage in using RF frequencies instead of audio?

4. What sort of device should I use in order to connect the 555 to mains during zero cross? It would obviously need to be able to handle 200 something volts in its off state.
 
well meaby an HV trasistor (PNP) that has the base conected to the mains (big resistor ufcurse or the transistor is fryed)emiter to ground and colector to the osclilator.

But this way your circuit wod be live so if you toced it you wod be zaped and the circuit fryed.
 
Ive been struggling with the concept of what exactly the voltages and waves and where everything connects in mains, so i can't really figure out how to draw a schematic of a zero cross detector.

Its my understanding that there is somehow a 240 volt RMS potential, but that is divided in half, leaving a 120vRMS positive swing and a 120vRMS negative swing, with respect to a center tap on the secondary of the transformer. Is this what appears in mains? If this is the case, then I suppose that I could inductively couple the gate of two FETs (not sure what type yet) to the mains line. This would take away the need for a bulky transformer or very high wattage resistor. They would be a P type and N type. I forget if they need to be in enhancement mode or depletion mode. The idea being that they would only conduct during zero crossing.

Would this work?
 
Volts RMS is a nice way to measure AC because it is easy to calculate power - calculate it just like DC. To get amplitude you need to multiply by square root 2. You then multiply by 2 to get peak to peak voltage.

Just do a google search for zero-crossing detection circuits and you should come up with a ton.
 
zachtheterrible said:
Okay, so I'm definitely going to have to send my signal during zero cross.

I fail to see why?, haven't you ever seen mains borne intercom systems?, they transmit continuous audio via the mains.

1. The unwanted noise would have to be the right frequency (1khz, 2khz, or 3khz) in order to trip the alarm. Is that a legitimate concern? Or is it very unlikely?

For anything like this you don't want to just send a tone to trigger the alarm, you want digitally encoded data, which the receiver checks is the correct value to trigger the alarm.

2. DTMF is the dialtones on a phone, which is basically the same thing that I'm planning on doing: Sending audio tones through the mains. Why then should I use DTMF instead of audio tones generated by the 555? Using the 555 would allow me to have a unique alarm ID: alarm 1=1khz, alarm 2=2khz, alarm 3=3khz. I don't really care what the frequencies are btw, those are just examples.

DTMF frequencies are specially selected to not be harmonically related, random audio frequencies (or any audio frequencies at all!) aren't a reliable solution.

3. Is there any advantage in using RF frequencies instead of audio?

Yes, huge ones!, which is why it's done that way - probably the most important is that the mains wiring will represent a low impedance to your audio signals (shorting them out), it will represent a far higher imedance at low RF frequencies, allowing them to travel along the wires.

4. What sort of device should I use in order to connect the 555 to mains during zero cross? It would obviously need to be able to handle 200 something volts in its off state.

Check the MicroChip application note for X10 using PIC's, that gives a suitable interface for mains communication.
 
I just got finished looking at those application notes like you said Nigel, and wow, that looks rather complicated. Then i looked for prices, and they are about the same as 433mhz transmitters. I think I'm going to go with 433mhz transmitters actually because: 1. they are so much simpler, 2. because i dont have to mess with dangerous mains voltage 3. because as you all know, im fascinated with wireless and radio :lol:

Transmitter: **broken link removed**
Receiver: **broken link removed**

So, a couple more questions (i know this is getting annoying):

1. Does anybody foresee any problems with using DTMF? I think I might use this because then I won't have to do any filtering or anything, and I won't have to worry about any harmonic interference.

2. Should I really worry about buying an antenna? I dont see why a 36cm whip antenna wouldn't work just as well. Hey, itll save me $10 :roll:

Sorry to be such a fickle undeciding pain :x
 
I looked at Microchip's application notes for the X-10 receiver and transmitter. It says that since the devices are intended to have a small form factor, a transformerless power supply is used. Is it a requirement to use the transformerless supply for technical reasons, or simply for size? Could I build a system with a normal linear power supply w/ transformer? Just wondering.
 
bonxer said:
I looked at Microchip's application notes for the X-10 receiver and transmitter. It says that since the devices are intended to have a small form factor, a transformerless power supply is used. Is it a requirement to use the transformerless supply for technical reasons, or simply for size? Could I build a system with a normal linear power supply w/ transformer? Just wondering.

Yes, home built projects usually use a transformer, the transformerless design is only for manufacturing cost reasons (cheapness!).
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top