Response levels of smoke and heat, what are they ???

[Benzerari]

Analogue values are not defined in a British Standards. The digital numbers an addressable device generates are representations of the various conditions that that device can find itself in. These digital numbers can be different from manufacturer to manufacturer, and from range to range. That an analogue value of 60 in Company 'A's range might represent the point that should represent a fire condition, the equivalent point in Company's 'B's range might be an analogue value, of say, 45. It is up to the control equipment to be designed to respond correctly to the various analogue values.

I though it must be some reference to what the analogue addressable systems are designed respond to, for smoke detectors readings for example it must be some reference in percentage of obcusration or number of particules in a square mm:

Just as an example:

From: 1%      To: 9% ---------------> Faults condition or ( No reading received )
From: 10%    To: 40% --------------> Normal condition
From: 41%    To: 49% --------------> Prealarm condition
From: 50%    To: 55% --------------> Fire condition

etc...

Am I making sense ?

I trust the following is an answer to your question.


I think you will find that response levels to smoke, heat etc. are determined by various Standards, for example, the Apollo XP95 ionisation specification mentions sensitivity as being to EN54 Pt 7 1984; (BS 5445 Pt 7 1984).

Wiz; I need either a link or a PDF copy of the doc stated above; I want to know the response levels to smoke, heat determined by these various standard...

Any help would be appreciated

[Wiz]

Benzerari,
I do not have copies of the documents you request. You could buy from BS or search your local large library for copies.
I would suggest talking to a detector manufacturer e.g Apollo, firstly to ascertain exactly which documents contain the information you are searching for.

[Benzerari]

Benzerari,
I do not have copies of the documents you request. You could buy from BS or search your local large library for copies.
I would suggest talking to a detector manufacturer e.g Apollo, firstly to ascertain exactly which documents contain the information you are searching for.

Thanks mate

[John Webb]

It's a while ago since I worked with these standards, but as far as I recall for the smoke detectors only:
BS 5445 Part 7 1984: For testing detectors to be used in commercial premises, detectors are graded according to their response times to the standard fires
BS 5446 Part 1 1990: For testing domestic detectors on a 'go/no-go' basis to a slightly different range of fires.

Both BS use a 3m high room of a set size.

BS 5445 uses Test fires 2: blocks of wood on hotplate; 3: Suspended cotton wicks; 4 - PU Foam; 5: n-Heptane
as defined in Part 7.

BS 5446 uses fires: slow-burning wood crib fire; fast-burning wood crib; petrol; PU Foam (less than BS5445)

BS 5445 has certain measuring instruments and the smoke rise must be with certain limits as measured by these instruments.
Class A detectors: Ion Chamber change <1.5, optical change <0.5 dB/m, <15 deg C heat rise
Class B detectors: ..     ..            ..        < 3       ..       ..        <1           , <30 ..
Class C detectors:                                < 6                          <2             <60

(The ion chamber change is a relative reading - the BS gives the details.)

In BS 5446 detectors of either type must operate within 10sec of the external optical density being 0.5dB/m for the wood fires and 0.8dB/m for the petrol and PU foam.

A summary comparing these two standards for smoke detectors was published by the then Home Office FRDG in 1991 or 1992 - Cath Reynolds, now in charge of CLG's fire research, was the author. I was the author of the Fire Research Station report (we were contracted to do the work for the HO) on which the summary was based.

Hope this helps.

[Benzerari]

It's a while ago since I worked with these standards, but as far as I recall for the smoke detectors only:
BS 5445 Part 7 1984: For testing detectors to be used in commercial premises, detectors are graded according to their response times to the standard fires
BS 5446 Part 1 1990: For testing domestic detectors on a 'go/no-go' basis to a slightly different range of fires.

Both BS use a 3m high room of a set size.

BS 5445 uses Test fires 2: blocks of wood on hotplate; 3: Suspended cotton wicks; 4 - PU Foam; 5: n-Heptane
as defined in Part 7.

BS 5446 uses fires: slow-burning wood crib fire; fast-burning wood crib; petrol; PU Foam (less than BS5445)

BS 5445 has certain measuring instruments and the smoke rise must be with certain limits as measured by these instruments.
Class A detectors: Ion Chamber change <1.5, optical change <0.5 dB/m, <15 deg C heat rise
Class B detectors: ..     ..            ..        < 3       ..       ..        <1           , <30 ..
Class C detectors:                                < 6                          <2             <60

(The ion chamber change is a relative reading - the BS gives the details.)

In BS 5446 detectors of either type must operate within 10sec of the external optical density being 0.5dB/m for the wood fires and 0.8dB/m for the petrol and PU foam.

A summary comparing these two standards for smoke detectors was published by the then Home Office FRDG in 1991 or 1992 - Cath Reynolds, now in charge of CLG's fire research, was the author.
Hope this helps.

John thanks for these details; but what do you mean by the unit 'dB'? I thought the smoke in either ionisation or optical is measured by percentage of obscuration isn't? which is then converted to change in current or voltage...

I was the author of the Fire Research Station report (we were contracted to do the work for the HO) on which the summary was based.

I guess you have then some more technical documents you may supply us, I am looking to know the reference to what the manufacturers have design their detectors, it must be some unique reference, otherwise we will be in a position to say i.e. one day Apollo are either more sensitive than Hochiki or either ways...

Am I making sense?

[John Webb]

The reference instruments, both for BS 5445 and BS 5446 are described in the BSs. I had to leave my copies behind at FRS when I retired, so I do not have the details to hand. The Ion Chamber is to a design which may have been from one at my colleagues at FRS or possibly from one of the European test/research labs. The optical density measuring system is, I recall, a 0.5m path-length with specified source, filter and receiver; I think there are detail differences between the two standards.

dB/m = 'deciBels/metre' obscuration and is defined as:

D=(10/(path length))*log((Incident light* no smoke)/(Incident light* with smoke))

Sorry for the clumsy way it's put but it needs a decent word processer to do it properly!

*Intensity of light as measured by the specified receiver.

Without the 10, we have the formula for what is often refered to as 'Optical Density' - just that the continentals prefer to work with the dB for reasons unknown to me.

[Gel]

BS5446 Pt 1(2000)  is withdrawn from this Summer when the BSEN 14604 Standard becomes the mandatory one for manufacturers.

[John Webb]

Gel, thanks for that - do you know how it differs from BS5446 Pt 1?

One of the criticisms we had of BS5446 was that the fires generated insufficient airflow past the detectors to push smoke into (particularly) the optical detectors, and we thought the fires should be made more representative of 'real-life' fires. I do hope the new standard has taken this into account.

[Benzerari]

The reference instruments, both for BS 5445 and BS 5446 are described in the BSs. I had to leave my copies behind at FRS when I retired, so I do not have the details to hand. The Ion Chamber is to a design which may have been from one at my colleagues at FRS or possibly from one of the European test/research labs. The optical density measuring system is, I recall, a 0.5m path-length with specified source, filter and receiver; I think there are detail differences between the two standards.

dB/m = 'deciBels/metre' obscuration and is defined as:

D=(10/(path length))*log((Incident light* no smoke)/(Incident light* with smoke))

Sorry for the clumsy way it's put but it needs a decent word processer to do it properly!

*Intensity of light as measured by the specified receiver.

Without the 10, we have the formula for what is often refered to as 'Optical Density' - just that the continentals prefer to work with the dB for reasons unknown to me.

Thanks John for your add, the only thing I could not pick up why mixing 'dB' which is the unit of sound with the % of obscuration which should be the unit of the quantity of smoke ?

Any way thank you for the try!

[John Webb]

Thanks John for your add, the only thing I could not pick up why mixing 'dB' which is the unit of sound with the % of obscuration which should be the unit of the quantity of smoke ?

Any way thank you for the try!

It is a continental thing and beyond my comprehension.

By the way, to convert from Optical Density (OD) to % obscuration S use the following formula for a given path length:

OD=2- log(base10)(100-S)

OD is more popular in the scientific literature as visibility in smoke can be calculated by

Visibility in metres=1/(Optical Density of smoke per metre)

ie for the often quoted 10m visibility as the point at which people become reluctant to move through the smoke the OD/m=0.1
And by the dB system = 1
It's just possible that the continentals like to use the dB version of Optical Density because this gives them '1' for a smoke density at the limit of movement - well, that's just a theory!

0.5dB/m= 0.05OD/m= 11% obscuration approx
1.0dB/m= 0.1 OD/m = 21% obscuration
2.0dB/m= 0.2 OD/m = 37%     ..

(Just to give you a comparison in the post above regarding the grading of the detectors.)

[Benzerari]

Thanks John for your add, the only thing I could not pick up why mixing 'dB' which is the unit of sound with the % of obscuration which should be the unit of the quantity of smoke ?

Any way thank you for the try!

It is a continental thing and beyond my comprehension.

By the way, to convert from Optical Density (OD) to % obscuration S use the following formula for a given path length:

OD=2- log(base10)(100-S)

OD is more popular in the scientific literature as visibility in smoke can be calculated by

Visibility in metres=1/(Optical Density of smoke per metre)

ie for the often quoted 10m visibility as the point at which people become reluctant to move through the smoke the OD/m=0.1
And by the dB system = 1
It's just possible that the continentals like to use the dB version of Optical Density because this gives them '1' for a smoke density at the limit of movement - well, that's just a theory!

0.5dB/m= 0.05OD/m= 11% obscuration approx
1.0dB/m= 0.1 OD/m = 21% obscuration
2.0dB/m= 0.2 OD/m = 37%     ..

(Just to give you a comparison in the post above regarding the grading of the detectors.)

Thanks for that john, it seems your add makes more sense. Let's think in % of obscuration, could you find out through the document related to BS in your possession or the one you can pocess about the following intervals:

Fault condition or (No reading)------->from x% to y% ?
Quiescent condition-------------------->from x% to y% ?
Pre-alarm condition-------------------->from x% to y% ?
Fire alarm condition-------------------->from x% to y% ?

Thank you

[John Webb]

Thanks for that john, it seems your add makes more sense. Let's think in % of obscuration, could you find out through the document related to BS in your possession or the one you can pocess about the following intervals:

Fault condition or (No reading)------->from x% to y% ?
Quiescent condition-------------------->from x% to y% ?
Pre-alarm condition-------------------->from x% to y% ?
Fire alarm condition-------------------->from x% to y% ?

Thank you

Regret I have no such information available, and as far as I am aware (but I stand to be corrected) the BSs do not hold such information. All they are concerned about is that the detectors operate once the smoke reaches a certain level.
I think it is up to the individual makers to determine what the above levels should be - and of course with addressable detectors those levels are decided by the programing of the FA control, not at the individual detector heads. Hence 'spikes' (ie short duration readings above the usual ambient due to a manufacturing process etc) which might trigger the ordinary detector can be ignored by the program in the FA panel.

[Benzerari]

Thanks for that john, it seems your add makes more sense. Let's think in % of obscuration, could you find out through the document related to BS in your possession or the one you can pocess about the following intervals:

Fault condition or (No reading)------->from x% to y% ?
Quiescent condition-------------------->from x% to y% ?
Pre-alarm condition-------------------->from x% to y% ?
Fire alarm condition-------------------->from x% to y% ?

Thank you

Regret I have no such information available, and as far as I am aware (but I stand to be corrected) the BSs do not hold such information. All they are concerned about is that the detectors operate once the smoke reaches a certain level.
I think it is up to the individual makers to determine what the above levels should be - and of course with addressable detectors those levels are decided by the programing of the FA control, not at the individual detector heads. Hence 'spikes' (ie short duration readings above the usual ambient due to a manufacturing process etc) which might trigger the ordinary detector can be ignored by the program in the FA panel.

Thanks John for the reply and do appreciate your interpretations, do you agree from a commun sense point of view that at least it should be a commun reference say i.e. all makes of smoke detectors should report fire alarm condition when reaching i.e. 2% of obscuration / meter... But if it is left to manufacturers, then we will be in a position to say some makes could be more sensitive than others or prone to false alarms and sort of stuff... is that correct?

I am insisting at least about fire condition reponse level! never mind about the rest of levels

[John Webb]

The BSs do specify the smoke density at which any smoke detector, optical or ionisation, should alarm.

For BS 5445 on commercial use detectors, the dB/m are as the table in my post#4, which equate to 11%, 21% or 37% obscuration/m for class A, B or C detectors respectively (see conversions in my post #10).

For BS 5446 for domestic detectors it is 11% obscuration/m for the wood fires and 16%/m for the PU foam and petrol fires (0.8 dB/m which I hadn't converted in post #10).

2%/m is far too low for an alarm level and is the sort of thing you can easily get from local activities. I remember working in the Cardington airship hanger one day and we had to stop work on detectors in one of our rigs because smoke from the brickworks at Stewartby, south of Bedford, was being blown into the hanger and we were measuring precisely that sort of level of obscuration on our instruments.

Doing the calculations in reverse also shows the following: 2%/m = OD of 0.0088/m = visibility of 114m, so no problems in seeing your way out at that level. So the 11%/m above avoids false alarms most of the time but gives warning when visibility is around 20m - and few escape routes in the average home approach that sort of unbroken distance. Which may well be the reason it was chosen, at least for the domestic detectors.