8 replies to this topic

[smuk]

Hi,
I want to know about control scheme for auto starting firefighting water pump, and starting of subsequent pumps in case of drop in pressure in firewater header due to usage.

Thanks

[gvdlans]

I don't know what you mean with a control scheme, but you just have a pressure switch on the header that starts the first pump. If the pressure is still low after say 10 to 20 seconds the 2nd pump is started etc...

[Andrei]

smuk,

Usually there are two firewater pumps in parallel, one main pump to provide the full flow required started only during high demand; and one jockey pump, about 10-20% capacity of the main pump, maintaining the network pressure between high-low preset values.
In all the cases I know the main pump was 100% spared with an alternate driver system (diesel engine usually) that comes on line during power outages.
Depending on the required capacity there may be multiple main pumps.

[smuk]

Hi,

Thank you very much for your help. I want to state the problem as follows:
There are two firewater pumps, say P-100A and P-100B. during normal time (no fire) the firewater header pressure is maintained by jockey pumps. In case of fire, the jockey pump is not able to maintain header pressure due to large usage of water. The header pressure drops and, say P-100A comes on line due activation of a low pressure switch on the firewater header. If the usage increases, the header pressure falls further, and P-100B comes on line activated by another pressure switch set 0.5 kg/cm2 lower. The instrumentation should also take care if P-100B starts first and P-100A subsequently. I am looking for a typical P&ID showing the above instrumentaion. To the best of my knowledge, the control scheme works as I have described above. If you can give a link to a site that gives such P&ID or reference of a magazine, it will be a great help.

Thanks.

[gvdlans]

With your "control system", when the deluge valve(s) open(s), pressure will drop to such a level that both pumps will be started at once. A jockey pump is normally only sized to maintain pressure in case of leaks. That's why I recommend to start the 2nd pump based on a time delay, not based on a lower pressure setting. Note that it may take typically up to 20 seconds until your main firewater pump will be fully running.

This is in line with NFPA 20: "The controller for each unit of multiple pump units shall incorporate a sequential timing device to prevent any one driver from starting simultaneously with any other driver"

Your P&ID will show the pressure switch(es), a dotted line to the logic solver and a dotted line from the logic solver to the firewater pump drivers (indicating the pump start signal). Pump stop can only be done locally.

For integrity reasons I would use two pressure switches at different sections of the header, in a 1-out-of-2 configuration.

[fallah]

That's why I recommend to start the 2nd pump based on a time delay, not based on a lower pressure setting.

Provided that the fire water ring main pressure being still below its set point.

[gvdlans]

Provided that the fire water ring main pressure being still below its set point.

Correct, see my first post (post #2 in this thread).

[astro]

Hi,

Thank you very much for your help. I want to state the problem as follows:
There are two firewater pumps, say P-100A and P-100B. during normal time (no fire) the firewater header pressure is maintained by jockey pumps. In case of fire, the jockey pump is not able to maintain header pressure due to large usage of water. The header pressure drops and, say P-100A comes on line due activation of a low pressure switch on the firewater header. If the usage increases, the header pressure falls further, and P-100B comes on line activated by another pressure switch set 0.5 kg/cm2 lower. The instrumentation should also take care if P-100B starts first and P-100A subsequently. I am looking for a typical P&ID showing the above instrumentaion. To the best of my knowledge, the control scheme works as I have described above. If you can give a link to a site that gives such P&ID or reference of a magazine, it will be a great help.

Thanks.

smuk, in your process description and with the various comments from the forum you've got the basis of your operating philosophy. If a P&ID is not forthcoming, why don't you sketch up your own given this info and then post it for review / comment?

[Art Montemayor]

Astro is absolutely correct in his deductions.

If this issue is important enough to take up two days, then it certainly is worth one hour - or perhaps two - to sketch up a P&ID and submit it to the valued Forum members who already have invested their time and effort into this thread so that they can probably add further value and credibility to the logic employed.

I think that I have already demonstrated many times in the past that drawing a P&ID in Excel is nothing nuclear or out of this world - and it certainly can be quickly transmitted and marked up.

What about it Smuk? Will you submit the P&ID? It will probably take far longer to find a P&ID on the Internet.