System healthy? and Panel healthy? should they be the same? and why?

[Benzerari]

Last week a serious event happened in one of our customers sites, it was a genuine fire and the system did not trigger at all, it is an advanced 4000 (I am not blaming advanced system), fortunately there was no substantial damage in live/property, the system was displaying healthy, but when checked through View/Edit, No device was found in the loops. What ever the cause was human error ... of the last engineer who worked on or not..., why should the system not displaying 'No Device Log' instead of system healthy, because the panel was healthy with itself without the rest of device. I mean not the whole fire alarm system was healthy...

Should not be better, to be considered by BS, that first you buy an analogue addressable panel and power it up, instead of showing system healthy, it should show i.e.  'No Device Log' yet, just to make the difference between ‘Panel healthy’ without Device Log in it, and ‘System Healthy’ with the Device Log in it…
I hope I am making sense.

Thank you

M C Benzerari

hey Benz

i was working on a panel today that jogged my memory about your question.

If you wipe the memory all the devices on the loop will all show up as disablements until autolearned.

Aritech panel.

Grame;

What do you mean by wipe the memory? is that flashing it? If yes where the disablement message is stored to be shown on the display then? that's new to me! it must be based on a different architecture, and it must be two seperate memories in the main PCB then, one of them is used as back-up or recovery memory, this just my guessing!

Even Aritech is new to me too, I would like to see its guides if possible, this has made me currios...?

Thanks Grame

[Graeme]

if you erased the memory or auto-learned a loop with nothing connected(and then reconnected) the panel knows that all the devices are there but logs them as all disabled until the loop is auto-learned again.

so if you had 70 devices on the loop,the panel would show 70 conditions. A disablement on this panel is a condition.

[Benzerari]

if you erased the memory or auto-learned a loop with nothing connected(and then reconnected) the panel knows that all the devices are there but logs them as all disabled until the loop is auto-learned again.

so if you had 70 devices on the loop,the panel would show 70 conditions. A disablement on this panel is a condition.

What makes the panel then to keep remembering them in its memory if they are back connected if its memory is  totally erased?

It must be a second recovery memory in which when the panel's software scan the loop then compare it  against the previously recorded in the recovery memory... therefore the panels report them as disabled or isolated till logged back on...

If you have got any guides of this system that would be appreciated

Thanks Grame

[Graeme]

it's not memory.The panel knows if you install another device on the loop but do not autolearn.

so say you installed an additional optical,the panel see the detector but this detector will not do anything until you tell the panel to enable it or autolearn the loop.

[Benzerari]

it's not memory.The panel knows if you install another device on the loop but do not autolearn.

so say you installed an additional optical,the panel see the detector but this detector will not do anything until you tell the panel to enable it or autolearn the loop.

This feature does exist with Notifire or Kentek if my memory still intact, not like Morley or advanced they do not see devices which are connected in the loop but not learned yet...

[Wiz]

it's not memory.The panel knows if you install another device on the loop but do not autolearn.

so say you installed an additional optical,the panel see the detector but this detector will not do anything until you tell the panel to enable it or autolearn the loop.

This feature does exist with Notifire or Kentek if my memory still intact, not like Morley or advanced they do not see devices which are connected in the loop but not learned yet...

If you programme a new detector with an address that is not already used on the system, then it depends if the particular panel operating system looks for all addresses or only those addresses that it has been told should be connected. Looking for addresses that shouldn't be there wastes vital time if carried out on every loop scan, so many panels normally interogate only those addresses it is expecting (programmed) to see on most loop scans and then every so often interogates every possible address and if it then sees one it wasn't expecting (programmed) to see, it will fault as something like 'additional device detected'

[Benzerari]

it's not memory.The panel knows if you install another device on the loop but do not autolearn.

so say you installed an additional optical,the panel see the detector but this detector will not do anything until you tell the panel to enable it or autolearn the loop.

This feature does exist with Notifire or Kentek if my memory still intact, not like Morley or advanced they do not see devices which are connected in the loop but not learned yet...

If you programme a new detector with an address that is not already used on the system, then it depends if the particular panel operating system looks for all addresses or only those addresses that it has been told should be connected. Looking for addresses that shouldn't be there wastes vital time if carried out on every loop scan, so many panels normally interogate only those addresses it is expecting (programmed) to see on most loop scans and then every so often interogates every possible address and if it then sees one it wasn't expecting (programmed) to see, it will fault as something like 'additional device detected'

How long is that waste of vital time? if it's hours that would make sense but if it's just few mili seconds that would still beneficial I think!

The other issue is it because of that waste of that vital times the BS5839 did not sets this issue as an obligation for panel manufacturers to look even for none logged on devices?

I need to know the position of BS too

Thank you

[Wiz]

This feature does exist with Notifire or Kentek if my memory still intact, not like Morley or advanced they do not see devices which are connected in the loop but not learned yet...

If you programme a new detector with an address that is not already used on the system, then it depends if the particular panel operating system looks for all addresses or only those addresses that it has been told should be connected. Looking for addresses that shouldn't be there wastes vital time if carried out on every loop scan, so many panels normally interogate only those addresses it is expecting (programmed) to see on most loop scans and then every so often interogates every possible address and if it then sees one it wasn't expecting (programmed) to see, it will fault as something like 'additional device detected'

How long is that waste of vital time? if it's hours that would make sense but if it's just few mili seconds that would still beneficial I think!

The other issue is it because of that waste of that vital times the BS5839 did not sets this issue as an obligation for panel manufacturers to look even for none logged on devices?

I need to know the position of BS too

Thank you

If scanning for 'unused addresses' was an obligation in BS/EN then all control panel manufacturers would do so as a matter of course. Otherwise they couldn't say that their panels complied with the standards. It is obviously not required in the standards.

Even wasted milliseconds are critical in addressable systems.

The scanning process invariably includes re-checking some information received back a number of times to ensure it is not being corrupted by outside interference. This increases the loop 'scan' time.

If you can't understand why milliseconds must be critical, have you never noticed how it can sometimes take some seconds for the 'operated' LED on a MCP using XP95 protocol to illuminate after the glass is broken? Because the control panel is trying to deal with other 'time-critical' processes, it doesn't prioritise this as an urgent function. However, BS asks for the alarm warning devices to operate within a specific time limit after the actual operation of a MCP, and actioning this process is more important than turning on the LED.

I'm sure that improvements in microprocessor speeds will eventually lead to the ability to check all addresses on all loops, re-check information received a number of times, and operate all other functions, in a blink of an eye. It may also lead to protocols allowing more than approx. 126 devices on one circuit to become more common.

Speak to panel manufacturers about this and you will realise how critical 'wasted' time in scanning 'unused' addresses can be for them.

[Benzerari]

I will do;

High speed microcontrolers or microprocessors may be beneficial in this issue, or dual core microprocessors would be ideal, to deal with more advanced features too!

How ever I still see that it is more beneficial for the fire alarm software to detect none logged on devices if they are in the loop...