Author Topic: Linking two fire alarm panels  (Read 70621 times)

Offline monkeh

  • Newbie
  • *
  • Posts: 28
Linking two fire alarm panels
« Reply #45 on: March 02, 2007, 05:48:46 PM »
Quote from: kurnal
Thanks Wiz for your explanation. I would be grateful to see your expanation of the role of short circuit isolators, I guess that without it a single wiring fault on a loop would potentially take out  the entire loop and on a radial any device beyond the isolator? That on a conventional system a short circuit would effectively cause the thing to go into alarm mode (ie fail safe of sorts)  but on an addressible the signalling would be lost so we would get  communication faults only?

Like I have said before I only drive the things but an insight into the technical side is always useful. I do often refer to the BS (which I can understand as far as it goes)  and always  read the system manuals (which are usually double dutch)  so that I can write the  fire procedures and test instructions  for the responsible person.  But the trouble with the manuals of course is that they tend to assume full technical understanding as a starting point.
a conventional panel going into alarm would only happen if it were a short circuit = fire panel.

most panels these days need a 560ohm resistance to put the panel into fire, a short circuit would just show a fault (or on many panels, it would actually display as a short circuit fault)

on addressable systems, a short circuit will drop the power from the cables to prevent damage to the panel and to detectors.  without isolators, the entire loop would be lost (each loop connection in the panel normally has a built in isolator)  with isolators, only the section of the cables in fault between the nearest two isolators in either direction would be lost.

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #46 on: March 02, 2007, 06:02:36 PM »
Mr K. - yes, a short-circuit on an addressable loop wiring will invariably ruin the communication pulses and therefore short-circuit isolators should be installed which automatically disconnect the wiring whenever the extra current drawn by a short-circuit is detected.

In practise, the isolators either side of any short-circuit fault both operate so as to disconnect that section of the loop from the rest of the system. The remaining addressable devices still connected to the control panel should now continue operating and communicationg with the control panel.

Obviously, the greater the number of short-circuit isolators on a loop the smaller the section of the circuit that gets isolated in any fault situation. It is becoming more popular to have short-circuit isolators built into each piece of addressable equipment. However some people say that the more isolators you have, then the more chance you have of suffering a fault due to the failure of one of them!

Equipment manufacturers normally advise a the minimum number of short-circuit isolators that would be required on any loop and this figure is based on how much current is normally being drawn by the equipment connected to the loop. Obviously, the total amount of current being drawn in normal operation shouldn't cause any short-circuit isolators to operate.

 However there are also specific recommendations in respect of the number and location of short-circuit isolators in BS5839 part 1 to ensure that when they operate that the maximum section of the loop that is disconnected, is kept to well-defined limits and which generally tie-in with the traditional maximum fire zone size of 2,000m2. Also where loop-powered addressable sounders are used, there are specific recommendations in respect of such sounders being installed close to the panel and the short-circuit isolators then required.

On non-addressable systems it doesn't really matter what specific effect a single
wiring fault has on a fire zone since that fire zone should generally not be protecting an area greater than 2000m2 anyway. In real terms, a short-circuit on non-addressable zone wiring caused a fire condition on earlier BS systems but these days a full short-circuit should give a fault condition. Obviously if such a short-circuit was resistive to the amount that it passed approximately the same current as a device going in to alarm, then this would obviously give a fire condition even on a current model non-addressable panel.

One point to watch with all short-circuit isolators on any addressable loop is to ensure that they all isolate  the same wire on the loop i.e either the +ve or -ve otherwise there is no or reduced protection. For example Apollo sell both +ve and -ve short-circuit isolators and the correct ones should be selected specifically for each project. Most addressable panels have short-circuit isolators in-built on both the 'out' and the 'in' loop connections and the loop installed short-circuit isolators should work on the same 'leg' as these.

Offline IrishFire

  • Newbie
  • *
  • Posts: 24
Linking two fire alarm panels
« Reply #47 on: March 02, 2007, 06:22:07 PM »
That in your opinion you do not like the use of radial spurs to addressable loops sytems. However you can't say it is 'incorrect' to use them

I feel I can say it's incorrect to use them..regulations are there as a guide and are not set in stone. It is my opinion that anyone who pulls in a single 2 core cable to apply to a fire loop instead of either 2 x 2 core cables or a 4 core cable is cutting corners for the sake of cost or laziness and therefore is in my opinion a "cowboy" and i'm sorry but i will find it very hard to change my mind on this one. I might like a fussy ol' git on this one but, (my wife will say yes on this one) the option of using spurs are coverd in the regs etc.. but does that mean its ok to use them? I think even the people who set the regulations would say it's not ok to use them but they need to cover all angles (maybe I'm wrong)
If it doesn't work blame the last guy

Graeme

  • Guest
Linking two fire alarm panels
« Reply #48 on: March 02, 2007, 08:33:05 PM »
a common mistake that can be made with regards to amount of devices between isolators is counting every thing on the loop as 1 device.

i.e Apollo 20 device isolator with say 18 opticals and 2 100dba loop sounders between isolators actually has 22 devices as the sounders counts as two based on the value of switch on surge current which is 1.8ma

Offline IrishFire

  • Newbie
  • *
  • Posts: 24
Linking two fire alarm panels
« Reply #49 on: March 02, 2007, 09:10:09 PM »
Quote from: Graeme
a common mistake that can be made with regards to amount of devices between isolators is counting every thing on the loop as 1 device.

i.e Apollo 20 device isolator with say 18 opticals and 2 100dba loop sounders between isolators actually has 22 devices as the sounders counts as two based on the value of switch on surge current which is 1.8ma
Happens so often it's not funny. The biggest problem I find is people trying to put too many sounders on a loop. Had one last year with 52 Apollo sounders per loop (5 loop Morley).
If it doesn't work blame the last guy

Offline kurnal

  • Hero Member
  • *****
  • Posts: 6489
    • http://www.peakland-fire-safety.co.uk
Linking two fire alarm panels
« Reply #50 on: March 03, 2007, 09:03:40 AM »
Thanks to all. May I continue with a few supplementary questions?

An I right in deducing that a short circuit isolator is a completely passive device on the loop, that does not have its own identity or address and if it operates the fault will only be dentified by the failure to communicate with the devices on the isolated part of the loop?

And is the isolator an intelligent device that can monitor the potential conditions on the isolated part of the loop  and reset itself when the short circuit appears no longer to exist or does it need to be manually reset like an MCB or replaced like a fuse?

I am having difficulty trying to visualise just how these isolators can monitor and isolate sections of a loop - thinking of Kirchoffs Law  which said that the total current flowng in a circuit (loop?) = the sum of all the currents flowing in all parts of the  circuit. I am now thinking that when you wire an alarm loop it must effectively look like a ladder with a main  conductor ( Unprotected?)running up the strings of the ladder and across the top rung (we operational chaps call these rounds) and then each rung of the ladder having a number of devices  ( ie a fire zone) with a short circuit isolator at either side of the rung (fire zone)?

Finally  I sometimes see fault messages on systems such as "Loop split on +ve line". I have always assumed this to be a wiring fault due to a defective connection but from your last point Wiz it could be something else?

Thanks for your contribution to my continuing education. Wiz- I owe you a bottle of your favourite medicine!

Graeme

  • Guest
Linking two fire alarm panels
« Reply #51 on: March 03, 2007, 10:41:32 AM »
A SCI can be incorporated into a device,detector base or stand alone.

The SCI itself has no address and when a short on a cable occurs the isolator either side of the fault open circuit,which then isolates that faulty cable run.

The panel will show a loop open circuit fault and communications faults for all the devices that have been isolated on the affected cable run.

When the short is removed the islolators automatically reset.The isloators that have been affected will normally show an amber led.

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #52 on: March 04, 2007, 11:12:14 PM »
Quote from: Graeme
a common mistake that can be made with regards to amount of devices between isolators is counting every thing on the loop as 1 device.

i.e Apollo 20 device isolator with say 18 opticals and 2 100dba loop sounders between isolators actually has 22 devices as the sounders counts as two based on the value of switch on surge current which is 1.8ma
Yes, the figure of 20 is arbitary. It all depends on the currents being drawn by the devices between the isolators. With the original loop-powered beam detectors from Apollo I found it best never to install more than 1 or 2 between isolators. I've noticed that the latest versions are supplied with integral short-circuit isolators.

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #53 on: March 04, 2007, 11:21:14 PM »
Quote from: IrishFire
The biggest problem I find is people trying to put too many sounders on a loop. Had one last year with 52 Apollo sounders per loop (5 loop Morley).
I've noticed that many newer addressable panels are providing higher loop current capabilities. A few manufacturers are currently advertising 50 loop powered sounders as a possibility. So what seems impossible now will be a reality in a few years!

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #54 on: March 04, 2007, 11:49:20 PM »
Quote from: IrishFire
That in your opinion you do not like the use of radial spurs to addressable loops sytems. However you can't say it is 'incorrect' to use them

I feel I can say it's incorrect to use them..regulations are there as a guide and are not set in stone. It is my opinion that anyone who pulls in a single 2 core cable to apply to a fire loop instead of either 2 x 2 core cables or a 4 core cable is cutting corners for the sake of cost or laziness and therefore is in my opinion a "cowboy" and i'm sorry but i will find it very hard to change my mind on this one. I might like a fussy ol' git on this one but, (my wife will say yes on this one) the option of using spurs are coverd in the regs etc.. but does that mean its ok to use them? I think even the people who set the regulations would say it's not ok to use them but they need to cover all angles (maybe I'm wrong)
Once again I will state that my only problem has been that previous comments could be interpreted as confirmation that spurs were not allowed under the BS 5839 recommendations that most of us work to, when this is absolutely not the case, and furthermore that anyone who did so, was not necessarily incompetent or a cowboy.

Here is a scenario where, I'm sure you will agree, using a spur is the best option:

A customer decides to upgrade his existing non-addressable system to an addressable system. One of the existing non-addressable zones consists of a dozen smoke detectors in an area where the installation of new cabling is prohibitively expensive. The existing cable is in perfect condition and meets all current standards. To install additional cable will mean spending tens of thousands of pounds  and waiting months for permission to do so from National Heritage.
Therefore this original non-addressable zone is converted into a spur off the new addressable system and suitably protected by short-circuit isolators. No additional cable needs to be installed and the benfits of addressable detection are provided in this area. This is a solution acceptable within the recommendations of BS5839 Part 1 2002.

Although you may not like them, and I agree with some of your reasons for doing so, it is important that readers of this forum are aware that spurs on addressable systems are perfectly acceptable within the specific recommendations of BS 5839 previously mentioned.

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #55 on: March 05, 2007, 12:42:30 AM »
Quote from: kurnal
Thanks to all. May I continue with a few supplementary questions?

An I right in deducing that a short circuit isolator is a completely passive device on the loop, that does not have its own identity or address and if it operates the fault will only be dentified by the failure to communicate with the devices on the isolated part of the loop?

And is the isolator an intelligent device that can monitor the potential conditions on the isolated part of the loop  and reset itself when the short circuit appears no longer to exist or does it need to be manually reset like an MCB or replaced like a fuse?

I am having difficulty trying to visualise just how these isolators can monitor and isolate sections of a loop - thinking of Kirchoffs Law  which said that the total current flowng in a circuit (loop?) = the sum of all the currents flowing in all parts of the  circuit. I am now thinking that when you wire an alarm loop it must effectively look like a ladder with a main  conductor ( Unprotected?)running up the strings of the ladder and across the top rung (we operational chaps call these rounds) and then each rung of the ladder having a number of devices  ( ie a fire zone) with a short circuit isolator at either side of the rung (fire zone)?

Finally  I sometimes see fault messages on systems such as "Loop split on +ve line". I have always assumed this to be a wiring fault due to a defective connection but from your last point Wiz it could be something else?

Thanks for your contribution to my continuing education. Wiz- I owe you a bottle of your favourite medicine!
Professor K. Please save the medicine for the next potting shed midnight meeting of the old codgers club. If Matron finds it in my room I'm in big trouble after all the palaver with the pavlova that Lucky 'borrowed' from the kitchen and hid in my bedside cabinet.

Graeme has suitably answered and confirmed some of your queries on the short-circuit isolator question.

I think your ladder analogy is probably not correct. Of course, I could have misunderstood it since I have a fear of heights! Instead, Just think of a two wire loop of cable where all your addressable devices are connected across the two wires. At suitable positions around the loop a number of short-circuit isolators are also connected across the loop. If a short-circuit occurs at any point of the loop then the isolators nearest (either side) the fault will automatically disconnect the loop at these two isolators. I.e. the section of cable with the short-circuit is now isolated from the control panel. The control panel now communicates along each of the ends of the loop still connected to the panel. Only the addressable devices on the section of the loop now isolated by the isolators are no longer functioning correctly. If the short-circuit is removed then the isolators automatically reset.

With respect to the other part of your question concerning our good friend Kirchoff (he has now been moved to room 4 on the first floor - next to the lift) I think I can understand your concerns. I believe your question asks why the current flowing through the isolator nearest the panel is not so great, even under non-short circuit fault conditions,  that it doesn't automatically isolate?. This is because the short-circuit isolators don't actually measure the current flowing through the circuit. They are in fact looking for the normal operating voltage of the loop to drop quite drastically as an indication of a fault. I believe the voltage needs to drop to about 15 Volts for an isolator to operate. This voltage reduction will happen whenever there is a short-circuit on the loop because the maximum current that can be drawn by the loop is limited by the loop driver card circuitry and a consequence of this limting is a reduction in loop voltage during a short-circuit. Obviously a low resistance short-circuit will try to draw a lot of current.
My understanding is that this monitoring of loop voltage is carried out by the isolators  a number of times a second and once the isolator nearest the short-circuit fault automatically disconnects the faulty section of the circuit, then the  operating voltage will return to normal on the non-faulty sections.

I hope the above makes sense, (isolators measure voltage not current) if not please let me know and I'll try and find a better explanation.

Actually, this thread has gone way off target and I think this may have clouded the issue of the original promise of payment for a solution to the problem of connecting two fire panels together over a pair of wires. I need the money to pay for an escape attempt from the home. I understand that a tunnel is being dug from the sluice room in the East wing and I want to be included.

Offline kurnal

  • Hero Member
  • *****
  • Posts: 6489
    • http://www.peakland-fire-safety.co.uk
Linking two fire alarm panels
« Reply #56 on: March 05, 2007, 08:00:06 AM »
Thanks Wiz I am happy now, It all makes sense now Dr Kirchoff has been put back to bed in the isolation wing of the care ohm for the bewildered.

But dont be too quick to throw away my ladder- you may need it to climb down into the tunnel. And my advice before you start digging is pick a sluiceroom on the ground floor of the building.

Graeme

  • Guest
Linking two fire alarm panels
« Reply #57 on: March 05, 2007, 08:01:41 PM »
Quote from: Wiz
Quote from: IrishFire
That in your opinion you do not like the use of radial spurs to addressable loops sytems. However you can't say it is 'incorrect' to use them

I feel I can say it's incorrect to use them..regulations are there as a guide and are not set in stone. It is my opinion that anyone who pulls in a single 2 core cable to apply to a fire loop instead of either 2 x 2 core cables or a 4 core cable is cutting corners for the sake of cost or laziness and therefore is in my opinion a "cowboy" and i'm sorry but i will find it very hard to change my mind on this one. I might like a fussy ol' git on this one but, (my wife will say yes on this one) the option of using spurs are coverd in the regs etc.. but does that mean its ok to use them? I think even the people who set the regulations would say it's not ok to use them but they need to cover all angles (maybe I'm wrong)
Once again I will state that my only problem has been that previous comments could be interpreted as confirmation that spurs were not allowed under the BS 5839 recommendations that most of us work to, when this is absolutely not the case, and furthermore that anyone who did so, was not necessarily incompetent or a cowboy.

Here is a scenario where, I'm sure you will agree, using a spur is the best option:

A customer decides to upgrade his existing non-addressable system to an addressable system. One of the existing non-addressable zones consists of a dozen smoke detectors in an area where the installation of new cabling is prohibitively expensive. The existing cable is in perfect condition and meets all current standards. To install additional cable will mean spending tens of thousands of pounds  and waiting months for permission to do so from National Heritage.
Therefore this original non-addressable zone is converted into a spur off the new addressable system and suitably protected by short-circuit isolators. No additional cable needs to be installed and the benfits of addressable detection are provided in this area. This is a solution acceptable within the recommendations of BS5839 Part 1 2002.

Although you may not like them, and I agree with some of your reasons for doing so, it is important that readers of this forum are aware that spurs on addressable systems are perfectly acceptable within the specific recommendations of BS 5839 previously mentioned.
you could also use a ZMU as another option.

Offline Wiz

  • Hero Member
  • *****
  • Posts: 1591
Linking two fire alarm panels
« Reply #58 on: March 06, 2007, 10:37:58 AM »
Quote from: Graeme
Quote from: Wiz
Quote from: IrishFire
That in your opinion you do not like the use of radial spurs to addressable loops sytems. However you can't say it is 'incorrect' to use them

I feel I can say it's incorrect to use them..regulations are there as a guide and are not set in stone. It is my opinion that anyone who pulls in a single 2 core cable to apply to a fire loop instead of either 2 x 2 core cables or a 4 core cable is cutting corners for the sake of cost or laziness and therefore is in my opinion a "cowboy" and i'm sorry but i will find it very hard to change my mind on this one. I might like a fussy ol' git on this one but, (my wife will say yes on this one) the option of using spurs are coverd in the regs etc.. but does that mean its ok to use them? I think even the people who set the regulations would say it's not ok to use them but they need to cover all angles (maybe I'm wrong)
Once again I will state that my only problem has been that previous comments could be interpreted as confirmation that spurs were not allowed under the BS 5839 recommendations that most of us work to, when this is absolutely not the case, and furthermore that anyone who did so, was not necessarily incompetent or a cowboy.

Here is a scenario where, I'm sure you will agree, using a spur is the best option:

A customer decides to upgrade his existing non-addressable system to an addressable system. One of the existing non-addressable zones consists of a dozen smoke detectors in an area where the installation of new cabling is prohibitively expensive. The existing cable is in perfect condition and meets all current standards. To install additional cable will mean spending tens of thousands of pounds  and waiting months for permission to do so from National Heritage.
Therefore this original non-addressable zone is converted into a spur off the new addressable system and suitably protected by short-circuit isolators. No additional cable needs to be installed and the benfits of addressable detection are provided in this area. This is a solution acceptable within the recommendations of BS5839 Part 1 2002.

Although you may not like them, and I agree with some of your reasons for doing so, it is important that readers of this forum are aware that spurs on addressable systems are perfectly acceptable within the specific recommendations of BS 5839 previously mentioned.
you could also use a ZMU as another option.
Yes Graeme, we could, but this would give us no advantages over the existing  non-addressable. We want to give the customer the benefits of an addressable system in the area in question but on an existing radial cable that we can't use as part of a loop. So we use this cable as a spur off the addressable loop, fit addressable detectors etc.. We are left with all the advantages of addressable and still comply with BS 5839 Part 1 2002

Graeme

  • Guest
Linking two fire alarm panels
« Reply #59 on: March 06, 2007, 05:37:16 PM »
very true Wiz,just thinking that if the customer was a tight ar5e.

g