<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=1655021261228687&amp;ev=PageView&amp;noscript=1">

Notification Appliances

Posted by ORR Protection on Sep 23, 2020 1:06:25 PM

During the MCFP Virtual Conference series, expert Lee Kaiser, covers notification systems. In the video below, watch as Lee dives in-depth on what goes into testing notification systems. 

Video Transcript:

We're going to start talking about the noisemakers and the light flashers. The things that warn us that there's a fire in our building and tell us to evacuate. 

For notification appliances we need to do a semi-annual inspection.  Look for the location of the devices and are they obstructed, and then also are they damaged. Inspect the condition of each device. 

For testing, it goes back to that annual test, but it's a little bit different with regards to notification tests.  First off, we need to test the audible side and the visual side separately.  The sound side and the light side, and then there are additional requirements in Chapter 14 of NFPA72 for acceptance and re-acceptance testing. 

Sound Testing

I want to highlight that because it used to all be one and now they've made a distinction.  For normal yearly operational testing we just need to test the audible side to see if it functions?  Does the device make sound?  That's all that we need to do. 

There are some more things that we need to do for acceptance testing and that's we need to take a sound pressure level meter and measure the sound throughout the space. I've got a sound pressure level meter here with us.  This device is an antsy rated sound pressure level meter.  It measures sound pressure levels and decibels and it gives you that reading.  That's what we're supposed to be looking for in the code. 

It shouldn't be more than 120 decibels. That's too loud.  That's damaging to the human ear and it must be at least 15 decibels above the ambient level of sound, and so we arrange fire alarm system warning devices to make at least enough sound to be over that.  So, most office spaces, if I recall this correctly, are about 55 decibels in sound and so then when we add 15 to that we need to be at, you know, 70 decibels of sound and we would measure that during testing with this kind of device. 

If it's private mode signaling a little bit different where we're not notifying the normal population, maybe we just want to have a targeted population such as in a nursing home.  Maybe I just want to notify at the nurse's station.  That's private mode signaling.  We only need to be 10 decibels above the average ambient sound.  So, I'm going to turn this on and let you all pass this around and experiment with it.  So, while I'm talking, this is about at 75 decibels.  It jumps around a lot, but when I get quiet, it goes down to about 40-45 decibels. 

A piece of advice that I would give you is if you're an inspector out there watching technicians do their job, they may or may not have one of those, but everybody has a smartphone nowadays and the apps that we use on our smartphones to measure sound pressure levels are, while they're not an antsy device, they're not callibratable. I will tell you that it's every bit as accurate as those, so I would ask you to ease up a little bit on that.  If you're enforcing that your iPhones with the right apps are probably just as sufficient as that. 

This last bullet point is about voice messages.  In voice fire alarm systems there is a measurement of that voice that you hear over the speakers, is it distinguishable and understandable. We use voice fire alarm systems in buildings where there's a lot of public change over or buildings of public assembly to give extra messaging to the occupants about what to do when the fire alarm system goes off.  So, we do that either through pre-recorded messages that come over the speakers or live voice messages that somebody issues through a microphone to come out over the speakers. 

The code says again those messages must be distinguishable and understandable.  We can actually measure that during installation of that system and because we have so much testing equipment. I wanted to take the chance to show you this device.  This is a voice intelligibility testing meter.  It measures voice intelligibility in measurements of STI, or speech transmission index. What you do to make this work, it works a lot like a little meter. 

You take it around to each location to measure the speech transmission index, but you have to get the right sound coming out over the speakers, and so when you buy this it comes with a little USB device.  You plug it into the fire alarm panel and download these wav files that are STIPA-compliant sounds.  They get played over the speakers and then this is listening for that STIPA-compliant sound and it gives you a reading of the speech transmission index between 0 and 100.  Now, 100 is perfect and most spaces are not perfect, but there is going to be a minimal acceptable level that's usually specified for that space. 

For testing, you would have that sound coming out over the speakers and you go from room to room and in multiple locations and measure that sound.  Now, how do you test the intelligibility of the microphone?  So, the way that you do that is with this device.  This is a little speaker box that plays another STIPA-compliant sound and you hold the microphone 3 feet away from this speaker and then that produces that sound out through the speaker and then someone else is walking around with a meter and measuring the speech intelligibility.

This isn't done on an annual basis.  This is just done during acceptance and re-acceptance testing.  I also wanted to show you there is another available item. The nice thing about using that kind of meter is that it makes the testing a little less subjective. I don't know if it's happened to you or not, but the fire marshal that's going to be passing the inspection of that system may bring out someone from their office. A third party that would be the person that judges whether or not you can understand the voice messages.  It's pretty subjective frankly.  This takes a lot of the subjectiveness out of it and makes those tests more repeatable. 

Visual Testing

On the visual side, it's pretty easy.  Annual operation test is; does it flash or does it not, so we need to activate those devices and see.  Does it flash and then for acceptance testing there's a couple more things. 

We need to look at the approved drawings or the space and verify that the floorplan is still the same and if any changes have occurred, make sure it doesn't affect the layout of the visual notification. We're also looking at those approved drawings and looking at the stuff that should be labeled next to each strobe device. We're looking at the device on the wall it should have labeled on it how many feet it's set up for. If the drawings say that the device on the wall should be 115 candelas, then we go and make sure that it's 115 candela device there and that's the additional test that needs to be there for acceptance testing. 

Annually, you need to test the firearm interface within a smoke control system, elevator control system, and a few others. The intent is not to activate those systems. The intent is to see the control output or relay output from the system. Is there a change in state at that relay that would then trigger these systems?  That's all more than what's required as a minimum test of those systems. 

From a design perspective, we have some specifying engineers in the audience today, so we want to make our recommendations on how can you make these systems more serviceable over the life of the system. One of the things for fire alarms we suggest is as much as you can, specify intelligent or addressable fire alarm systems to give the user of the system more information on the system status, faster troubleshooting when there are problems, and when problems occur, give us a history of the conditions. Such as what's happened over time with that panel so we can establish patterns as to why an intermittent trouble condition may be there and gives the user more capabilities with their system. 

Second, we like to teach the location of detectors. Just like in real estate, location, location, location.  We want you to specify the right type of detection for the space and then put it in the location that you're not going to have accidental detection and cause false alarms. 

The last thing is air sampling detection. There's two reasons to think about air sampling detection and specifying it in your billings.  First off, we can use air sampling detection for hard to reach spaces where we're not going to be able to get up and service those detectors in those locations Secondly, we know that from our field experience over lots of years of air sampling detection, Vesda is one of the most robust detectors against accidental activations and false alarms.

Year over year our technicians go out for lots of emergency service calls, lots of them due to false alarms, but we don't go out for Vesda false alarms.  We may go out for other reasons on Vesda, sometimes related to issues of airflow, but not due to false alarms.  We just generally don't have those problems with Vesda detectors and I want you all to know that. 

The last thing I want to cover is the life span of your fire alarm systems, how long should it last?  First off, commercial fire alarm systems should last, you know, between 15, 20 maybe 25 years in our buildings.  The detectors don't last as long, 10 to 15 years as a rule of thumb.  They could all be extended with proper inspection, testing and maintenance, but when you have a component in your system fail, you need to think a couple of these things.  Can I buy the product that is backwards compatible with other devices that I have? 

If my panel gets struck by lightning and I need to replace it, can I buy a new panel that's still compatible with the detectors I have installed in my space?  That's the issue of backwards compatibility.  We've seen a lot of manufacturers in support of products.  What's happening in our industry is the electronics that are inside of them are not lasting as long as a raw material because chip manufacturers that are making raw components for our detectors. Our panels are not keeping those chips around as long, and so detector manufacturers, panel manufacturers are being forced to sunset products because they can't get new devices to make new panels in the same way, so they might as well  come out with a new product that's listed for the application.  That's affecting us and over the last 2 years we've seen a number of manufacturers in support for product and then have your panel changed. Do you need advanced features?  Do you need addressable devices that a intelligent fire alarm system would provide you?

Topics: Featured Article, Featured Blog, MCFP

Featured Download