Fire Alarm Systems & Automation

[Bruce Cloutier]

We get a not so insignificant number of support calls where there are problems integrating the fire alarm system. I envision that the individual at the other end just has a pair of wires coming from somewhere and is guessing as to how to connect them. We rarely can get any information as to what fire system is at the other end let alone any specs as to the signal. Can any of you guys point to any standard for this? Or, at least, some rules of thumb?

In some cases I think they assumed a dry contact output and so they tap our power supply. Then when they test we reboot. In other cases everything appears to work even with a test but there are false triggers. I can imagine there being short pulses from the fire system confirming active interconnect but not reporting disaster. For all we know sometimes they might actually have a data connection.

Pretty much the problem goes away I think when they finally fetch someone knowledgeable at their end. Then, of course, we get no feedback.

To be fair, percentage wise, this is not a norm. But, naturally, we want to help but don't think that we should guess either.

[Steve Guttag]

When we get the option (most of the time...probably upwards of 95% of the time), we specify dry contacts that CLOSE on fire alarm. We request one set per auditorium. We HAVE received one set for the booth and have had the automations network such that one will set all off. On bigger complexes though, we've had zoned systems where if there is a fire down at #16, they will want that area evacuated first before dumping more people, out of say theatre 1. But that is up to the Fire Alarm company and, again, if they provide one pair of dry contacts, then they get to control precisely how they want it zoned off.

There are some fire systems were they want the opposite logic...open on Fire, we can work with that but it does mean that when things malfunction, it will tend to trigger the Fire sequence...which is why I don't prefer it. You can't work on things without keeping in mind that lifting a wire or whatnot will set off the fire sequence.

On the most rare, they provide control wires at a specified voltage (to work with an opt isolated input). I can't remember when we were asked to work with that but it has happened. By and large, getting the dry contacts has been a reasonable enough request that is what we get. And, well into the 90% of the time, it is CLOSE on fire. Normally, the fire company will put an "ice cube" type relay to feed us so they don't really care if they send us an NC or NO contact.

[Bruce Cloutier]

So it is a mix with maybe CLOSE on fire as a majority? I suppose that there is less of an issue when bringing a new theater up for the first time to where you are part of the design. It seems that down the road the new people involved aren't privy to the original plans and so aren't so sure what they have.

In this market we rarely have much for inputs save this fire alarm signal. Our inputs are opto-isolated so you basically have to illuminate an LED. There is a resistor in series with the LED and then a zener limiting the voltage across that and, finally, another series resistor. It ends up that it takes about 2.5V to engage the input. I wouldn't think parasitic voltages would be enough to trigger it. I guess if the wiring is long enough and poorly routed. The latest is an old theater we are told.

My only experience with fire system people was with an IEEE committee some decades ago in working up a standard for integrating building systems (pre-BACnet). Reporting a fire condition was one thing. They also had wanted a way to always be certain that the system was ready to respond to such a situation. From a wiring perspective there being a way to tell if the signal hadn't been disconnected. So NC and, yeah, pulling a wire causes potential panic. With a protocol some 2-way ongoing exchange to where they could report a trouble situation.

Trying to remotely diagnose without an EE at the other end is difficult. We don't want to just add to the guessing game.

[Steve Guttag]

It isn't a mix under one roof...it is consistent under one roof. Since we request close on fire, we generally get what we ask. I really think it is 95 or higher, for us, close on fire. The balance is open on fire. It all depends on how insistent the fire alarm people are in their belief on what is best/correct.

I would say, if you are using optos (which everyone does), you should have the power available to cover your opto such that a dry contact merely completes (or breaks) the circuit and not require external power.

It has been mostly retrofits that hadn't had a fire alarm system tied to the projection booth before where we get asked to let one signal trigger all cinemas. As you say, when things can be designed in, things are much more predictable.

[Dave Macaulay]

Generally we request a dry close on evac dry contact at each projector, and on new builds that's not a problem. This is easy with newer alarms, they just set an address for the evac alarm at the panel, and any number of modules can trigger on that address signal. On digital conversions with older FA panels not so easy and a single contact is not unusual, but it can be daisy chained through the site. We either use one Jnior supply as the power, or add another somewhere to power them all.
Years ago on a 35mm 6 screen install we were assured we had a dry contact, and connected it without testing. FA test day came a few days before opening ... and it was 120VAC on evac. Replacing 6 smoked TA10 automations was not a fun job. We could only get unprogrammed ones on such short notice and had to do all the setup on site.

[Matt Fields]

Like Steve, we prefer and almost always have the relay close on Fire Alarm with one relay wire to each auditorium. With a GDC server I will use the modbus server connection to the JNIOR to have the GDC listen for the JNIOR Digital Input 1 to become active. The GDC will then perform its Fire Alarm Macro. On a Dolby IMS we will use the JNIOR cinema program to perform the macro. Instead of messing with the IMS/Doremi trigger cue (which works but to me is not intuitive), I will setup the IMS/Doremi as a device in the JNIOR and send it a command to pause the server.

We generally do four things on fire alarm: Raise auditorium lights, turn off projector light source, mute sound processor, pause server.

See attached pic for how to wire digital input one on the JNIOR and use the JNIOR power supply to provide current. Another tip would be, when setting up the JNIOR registry in the cinema program, make both the fire alarm macro input and fire alarm reset inputs the same input (in this case 1). If you do not, the JNIOR will do the fire alarm macro the first time, but will not do it again until the reset input has been triggered.

[Bruce Cloutier]

The site that got me to asking has confirmed wiring for either NO/NC dry-contact fire alarm. We are not sure which at the moment. But the current theory relating to their claim of false triggers on the fire alarm input we believe might be electrical noise. They are using the same small power supply to power output isolation relays as they do the input and, we are thinking, there are no flyback diodes on their external relay coils. So while we determine if the back EMF from these relays is being properly dealt with, we suggested use of a separate power supply for inputs vs output circuits.

BTW, without those diodes the resulting arcing in our signal relays will do them in pretty quickly.