Fire Alarm -- Installation
When adding a horn or a strobe, does one really know how many additions have been added before?
Can a Person Just Add a Horn or Strobe?
By Douglas Krantz
In a building that has been remodeled and remodeled, when a fire drill is sounded, have you ever found that a few strobes are running slow or won't even work? Then, when troubleshooting, the supervision voltage at the end-of-line resistor shows a normal circuit?
These measurements may show the wiring to be normal, but something isn't working. What's going on?
The Box Supplying the Power Isn't the Problem
It's not the Booster Power Supply, or Signal Power Expander, or other Notification Appliance Circuit (NAC) power supply providing power to the circuit.
From the standpoint of the power supply, if too much current is being drawn, the internal safeties for the NAC outputs will turn off the circuit rather than overload the power supply. If any strobes or horns are working, the power supply isn't being overloaded.
It's the circuit itself, the wiring outside the power supply box, along with the number of devices on the circuit, that's having a problem.
Probably, over time, with each remodel or addition to the building, what has happened is a device has been added or the wiring has been changed.
The thing is, as each addition has been made, the addition or change wasn't very much. However, when including all the accumulated additions and changes, they all add up to an overloading of the circuit.
I know, the circuits usually have been designed with at least 20% headroom, so adding a single horn or strobe should not cause any kind of problem. But, when the added wiring is included, and the additions to the same circuit have been done 10 times without proper documentation on any single addition (so no one knows what is really there now), the 20% headroom that was originally designed into the circuit just isn't there anymore.
Supervision Voltage Doesn't Show the Problem
Sure, the end-of-line resistor is showing supervision voltage. But do the math; very little current is used for supervision. No matter how long the wire or how small the gage of the wire, that small current doesn't provide much of a voltage drop on the wiring.
Supervision voltage only shows that the wire has continuity. Supervision voltage does not show how much voltage will be lost in the wires when they have to carry the current needed to run the horns and strobes.
The Additions That Have Been Made
There are four ways of adding horns and strobes to a NAC circuit.
- Method 1 - Just add the extra wire and the extra device because under normal AC power conditions the circuit works now and it's just one more horn/strobe.
- Method 2 - Count the number of devices on the circuit and make a guess that the NAC circuit will handle the extra horns and strobes.
- Method 3 - Look at the original as-builts and using calculations make an educated guess that the circuit will handle the extra wire and the extra devices.
- Method 4 - To make sure the circuit performs as specified, run a full 24 hour power blackout test, and at the end of the 5 minute activation of the horns and strobes, measure the voltage at the End of Line Resistor. Compare that voltage to the required voltage listed on the installation sheets for the devices. Write the results down for future reference (that's what documentation is all about).
Results of the Four Methods
- As remodeling and additions to the building has been done, if devices have been added over time, one can count on Method 1 to have been used a number of times. Adding just one horn or strobe might in reality be adding one device on a circuit that already has lots of extra wire and three or four devices too many. When the system is running on batteries, this addition, along with all the past additions, will quite often prevent a strobe at the end of the circuit from working, or at best it may only flash slowly.
- Counting the horns and strobes works if one can make sure there aren't any hidden in rooms close by. Usually this requires turning on the zone, making noise and flashing lights to make sure all of the horns and strobes on the circuit are found. Also, as long as the history of the building
is known (and that includes knowing for sure that since the original installation nothing has been added), installing an extra single horn or strobe, along with only short wire runs will probably be OK, but it's really only a guess.
- Here, as long as the history of the building is known, and as long as the original installer made sure the as-builts were fully updated to include the wire size and wire length, and the original design included plenty of headroom to allow for the addition extra devices... adding a device or two, using the proper calculations, will usually be OK.
As far as calculations go, whereas the specifications found in the NFPA indicate that calculations need to be done, unless the system is properly tested, the calculations only show a prediction... not a surety.
The problem is that none of these methods can be counted on; they don't show whether the circuit works. None of first three methods use what is actually on the circuit. These methods also assume that no other company has had their fingers in there adding devices and wire.
- Only by using method four can the circuit be documented to actually be within the specs of the NFPA.
Method 4 Works
All this is to say that to actually make sure the circuit works, by properly testing the circuit - live - making noise and flashing lights, under adverse power conditions, Method 4
is the only method that can be counted on.
Could you expand some more on method 4 and what is meant by a 24 hour power blackout? Does that mean power interruption only to the FACP? When does the 5 minute window start? When does it end? I'm unclear about this statement: "Measure the voltage at the End of Line Resistor. Compare that voltage to the required voltage listed on the installation sheets for the devices."
Thx J M
The 24 hour blackout is meant to simulate what would happen if the power to the building is lost for a 24 hour period. During this time, the fire alarm system is operating only on the backup batteries.
For the whole time of the AC power blackout, the batteries will be losing voltage. In a 24 volt fire alarm system, the battery voltage starts out at 27 to 28 volts loses voltage slowly as its power is drained. If the batteries are drained to the point they really can't provide power anymore, they are down to about 20 volts. Below that and they will pretty much be useless.
If the batteries are down to 20 volts after 12 or 18 hours, the panel will probably go black before the 24 hour period, and therefore fail.
Now, at the end of that 24 hour period, while the power is still turned off, the question is "will the fire alarm system still work well enough to evacuate the building in case of fire?"
This is that 5 minutes of horn/strobe activation.
The 5 minutes of sounding the horns and strobes has to work in the Life Safety System (the fire alarm system) or the batteries have failed. During a power outage in real life, if a fire starts but the batteries didn't hold up, someone may be injured or die.
Because the fire alarm system is a life safety system, anything that can be considered to be connected to the fire alarm system needs to be tested during this blackout. This includes all auxiliary power supplies, horn and strobe power expanders, etc.
Now to the measuring of the voltage at the End of Line Resistor.
A properly installed fire alarm system, using Class B circuitry, has the End of Line Resistor at the farthest electrical point of the horn/strobe circuit. This is the End of the Line, so to speak.
When the horns and strobes are on during an alarm, the voltage at the Notification Appliance Circuit (NAC) terminals of the fire alarm panel is going to be about equal to the battery voltage at any time. During a blackout, the battery voltage will be slowly dropping from about 27 volts to maybe 20 volts.
This may not sound like a problem, but the wire for the horns and strobes throughout the buildings has resistance. Power will be lost in the wire itself, and the voltage at the end of the line will be far less than the voltage at the panel. Measuring the voltage at the panel doesn't tell anyone what the voltage has dropped down to at the last device on the circuit.
It's that voltage, the voltage at the last device on the circuit that shows how much voltage is available to all rest of the devices. This voltage has to be measured with the horns blaring and the strobes flashing, at the end of the 5 minutes of horns and strobes, which is at the end of the 24 hour blackout.
And this measured while the fire alarm system is still operating on batteries, before the utility power is turned back on.
The voltage measured at the End of Line Resistor should be 16 volts or higher if the manufacturer specifies at least 16 volts, or 16.5 volts if the manufacturer specifies 16.5 volts, or whatever the manufacturer of the horn or strobe specifies.
Remember, this is a Life Safety System, and you are testing to see that the system can save lives even after a 24 hour blackout.