Raspberry Pi and Existing Accenta Gen 4 Alarm

Hi Shyam,

Just bear in mind the alarm will probably go off when you take the panel off : ) A pain to deal with unless you're ready with the procedure to reset (will be in the manual, and if you remember the codes!).

Also if he interfaces to the sensors in such a way that breaks the contact at any point will also set it off.

Not that he needs to as he should be able to pick up the PIR input from the alarm directly...Probably;y best to use an optoisolator as to give very little chance that anything will come across from one circuit to the other even if misconfigured.

My apologies, I should've checked stock levels

You could buy that kit, though it doesn't include 330Ω resistors. You could substitute for a similar value.

Farnell do have some 2k2 in stock here: MF25 2K2 - MULTICOMP - RESISTOR, 2.2K, 0.25W, 1% | Farnell element14

Thanks Paul

Hi Guys,

I had a few more questions before I attempted this.

1. On the breadboard, do i connect the positive to the 13v terminal on the alarm panel?  and nothing to the 0v one?
2. what pins do i connect the +/- to on the pi?
3. while looking at the connections on this example : https://raw.githubusercontent.com/sidb/PiAlarm/master/breadboard.png

     2A connects to a GPIO Pin

     3A connects to a - on the breadboard

    4A connects to a + on the breadboard

    2C and 4C are linked by a 330R

    J3 connections to a connector on the alarm

    4I is linked to the + from the alarm via a 2k2

   My question is, how comes the connection to the alarm is on J3? and not J2?  Is it because the alarm connection needs to link with the -?

Could I use a tester to see what the output is on the PI side before connecting the PI to it?

Thanks

Shyam

1) 13V to red wire '+'

2)  - to GND, + to 3.3v

3) J2 is connected to pin 3 (an unconnected pin) on the opto-isolator.

As shown on this diagram from the github page:

I'm not sure if you're aware of how the board is wired, it is similar to many plug-in breadboard layouts as indicated in the image below:

Thanks gadget.iom 

I took your earlier advise and rotated the breadboard image and compared it to the diagram...it all makes sense now!!

Good to know

It had me guessing for a few minutes.

Could I use a voltage meter on the PI side to check the voltage before connecting it to the PI?  If so, I assume the positive end of the meter would be placed where the 3v3 is, and the negative on the GPIO....  What would I connect in place of the 0v?

Scrap the above question...I got my answer from here: https://learn.sparkfun.com/tutorials/how-to-use-a-multimeter

Positive of the meter on the GPIO pin, Negative end of the meter on the 0V

That’s it for the output, if you want to measure the 3V3 output then the positive of the meter would go to the 3V3 pin but the negative of the meter still goes to 0V pin

Hope this helps

Peter

Positive of the meter on the GPIO pin, Negative end of the meter on the 0V

That’s it for the output, if you want to measure the 3V3 output then the positive of the meter would go to the 3V3 pin but the negative of the meter still goes to 0V pin

Hope this helps

Peter

Hi Guys,

I've ran in to an issue, so I've connected the above exactly the way you guys helped, but cannot get the GPIO pins to work..... 

I'm running Raspbmc on my PI and installed wiringPi and the Raspberry Pi Model B+ | Wiring Pi .  This installed successfully and I'm able to set the Pins to in or out, but everytime I read the pin I get 0.  Everytime I try to write 1 to a pin, then do a read I still get 0.

WiringPi has a built in test for the GPIO pins, that says all the pins failed....

Is there something that I'm missing? I've tried with the root user and the PI user, I've also tried RPi.GPIO and the python script on the above link, still no luck

I skimmed through the posts and don't see where you mention which pi you have. If it is the RPi2 the base address is shifted. I would work on getting the pins working with some known good circuits, simple resistor pull up and pull downs, before attempting to hook it up to the alarm. I would use something like a 10k to ground for the pull down and a 10k to 3.3 for the pull up, not at the same time.

Thanks LenReinhart, I've got a b+.  I tried to hook it up to a simple led circuit, but that didnt work either.  If I connected the LED to ground and 3.3 pin the light lit up, when i connected it to the ground and another pin and tried to assign a value of true or one to the pin, it didnt light up

The Led will draw too much current from the pi, there needs to be a resistor connected to one leg of the led, something like a 4.7k, but I use 10k for a not so bright one. Do you know about the MagPi magazine? It is free and some of the back issues will take you step by step through what you need to do to use the I/O pins. The web site is: http://www.themagpi.com/

Once you are ready to try again, I would use some pins you haven't used before.

I used a 3k3 resistor, as thats the only one i had to hand.  I'll check out the magazine.

Do you know about the V=IR formula? It is a way to calculate the relationship between voltage (V), current (I) and resistance (R). In this case it would be 3.3Volts = I X 3000 ohms or 3.3/3000 or 0.0011 amps or 1.1milliamps. So assuming I got the decimal in the right place it says your resistor should be ok. If I remember correctly the RPi can source over that ok. When too big a load is put on a pin it will try to source more current that it is capable of doing. The pins can sink current better so often an Led will be connected to 3.3 volts on one side and to one side of a resister with the other side of the resistor connected to the pin. Then to turn on the Led the pin is taken to 0 instead of 1.

I wasn't aware of that formula, but it does make sense now

I just measured the volts on each of the pins using a multimeter and only one pin (GPIO.26) measures up to 3.3v, but the WiringPI read is saying 0 for that pin....

There must be something going on like "you're using the wrong pin numbering". Wiring pi has two ways of specifying pin numbers. Double check you're using the right one.

That V=I*R formula holds for resistors. To calculate the current through your led+resistor, you need to take the voltage for the led into account So it becomes: I = (3.3 - Vled)/R = (3.3-1.8) / 3k = 0.5mA. That said: I usually use a 1k resistor, about 3 times more current and that's nicely visible, not too bright, not too weak.

The "sink better" is something "from the past". With CMOS that is no longer a real issue. There is a small difference, but especially for a led, that can be ignored. If you tie your led between 3.3V and the pin, your software will need to write a 0 to turn it on and a 1 to turn it off. That  costs brain-power, so it is better to just wire the led between the pin and GND.

Wiringpi comes with a program called "GPIO". Start with "gpio readall". It prints a nice overview. Now, look up on File:RPi P1 header.png - eLinux.org where you expect phys pin 11 and 12. Run "gpio mode 0 out ; gpio mode 1  out". Now pins 11 and 12 of the connector should be outputs. Next gpio write 0 1; gpio write 1 1 will set them both high.

I'm telling you to try two pins at once, just because you might mess up and get the one next to it. Now you should be able to see something change on the connector. while true ; do  gpio toggle 0;sleep 1 ; done is then a little program that will blink the pin (good for with the multimeter or with a led!).

I think I can see what's wrong, here. Measuring the GPIO pin voltages shows what the Raspi is writing to that pin (assuming it is configurable as an input/output pin). If you then ask the Raspi to read what's on that same pin, it will stop writing a voltage and start reading whatever voltage is coming from the outside world.

There is a danger of blowing up the Raspi because if you are shorting out one of the pins to 0v (say, emulating a switch) when you ask the Raspi to write 3.3V then there is nothing to stop the Raspi writing excessively high currents. Leonard pointed out Ohm's Law of Volts equal Amps times resistance and if there is minimal resistance then Amps goes very high (in theory). Just to be safe, I suggest that you use your 3k3 resistor in series with any push-buttons used between pin and 0v to see if that will work more safely than a full short circuit.

Have a look at these two official Raspberry Pi.org website pages which ignore my advice! However, it shows you how to use Python to read the input push-buttons and light (write to the connected pin)  a LED. There is an internal Raspi resistance that pulls up the voltage to 3.3v when reading and the push-buttons pull the pin voltage back down to 0V when operated. Using Python means that it will work safely on any Raspberry Pi model and also gives you an excellent way of writing your alarm program.

Here's the principle,

Raspberry Pi Documentation

Here's a reaction time program written in Python which reads the two push-buttons and writes to a LED.

Worksheet - Quick Reaction Game | Raspberry Pi Learning Resources

They suggest a 270 Ohm or 330 Ohm resistor in series with the LED. In my experience, your 3k3 resistor will be too high for the LED to light up.

Thanks for the replies!

The thing I don't really understand is that with the software I'm not able to set the pin to 1 (using gpio mode 0 out then gpio write 0 1), I can see that i'm using the correct pin as i can see it being toggled between IN and OUT(via "gpio readall").  So it seems its either a config/setup issue or something wrong with the PI as when I run the PinTest it shows all pins failed (expected 1 got 0 error).