I have a query to experts of this forum:

We have Fire Detection System for our Gas Turbines. The detectors are thermal detectors which "close" at high temperatures (indicating fire). They are otherwise open. There is +ve power (62.5VDC) on one side of the thermal sensor. The coil of CO2 release SOV is connected to the other side of the Thermal Detector. The second end of coil is at -ve part voltage (-62.5VDC).

In this configuration, during the normal condition the SOV is de-energised (as because the Thermal Detector is open) and CO2 is not released. In case of Fire the temp is expected to rise and at the set temp value the Thermal Detector will close and energise the SOV to release the CO2.

Is this concept of energising the SOV to release the CO2 safe? Usually most trip coils remain normally energised to keep the equipments running. And this seems to be a safe practice as well because in case of a coil failure or power supply wires coming out of TBs the Trip coil will de-energise and prevent the equipement from running. Should there be a similar approach for the CO2 release SOV coil as well? Should it not be kept normally energised so that any failure of coil or power supply cables would de-energize the SOV and result in a trip. This can be a loss in terms of inadvertant CO2 relase, ofcourse, but wouldnt it be better than a condition when fire really happens and its discovered that the SOV coil was open or the power supply cable was open? What are the indusry standards governing Fire Logics.

Thank you
Regards
Ritika

 

You have identified a difference between the majority of GE-design heavy duty gas turbine control functions and one of them. One philosophy doesn't fit every application, even on one unit. GE and the packagers who use the systems designed by GE all feel these systems as configured provide sufficient protection to protect them from liability; so that should also be considered when considering making any changes.

This is one of the frustrating things about understanding GE heavy duty gas turbine control philosophy; "rules" don't seem to always apply to every application or situation. There are lots of reasons for this, but every situation or application should be individually considered.

Unless you have some kind of agreement with GE or some other company for engineering and/or operational services for the units at your site which precludes making any changes, you are free to configure the turbine control system and any of its auxiliary systems as you see fit once the unit is out of warranty. That means taking responsibility for any changes made.

You should know that early Speedtronic turbine control systems had a *lot* of false trips and CO2 discharges and there have been many iterations of the fire detection system over the decades. This one seems to have provided the most reliability and fewest false trips and discharges over the years.

If it's felt that the fire detection system is inadequate at your site, make it adequate to meet your requirement. Many owners have opted for other vendor's solutions to replace the original fire detection systems. Some use IR sensors; some use UV sensors; some use a system which de-energizes to discharge; most use a system that energizes to discharge. Some owners have been forced to change fire detection systems by local regulatory agencies or insurance providers.

Any change you make should be reviewed with your insurance provider and any regulating agency(s) in your part of the world that have authority over such equipment and installations.

Industry standards don't mean much in some parts of the world; it's whatever the laws and regulations require in your part of the world. And, whatever your insurance provider will accept or require.

 

Just to add to the previous post a couple of things.

1) Traditionally (in my expeience) fire detection systems have been normally de-energised, energising to release extinguishant set of alarms and even in some cases to trip equipment. This is probably to due with a history of spurious operation in the past.

2) Balancing the risk between spurious operation of the CO2 (and possible consequent personnel dangers - suffocation) against dangers arising from not protecting plant is a matter on engineering judgement to a certain extent as well as local regulations. It is not necessarily wholly safe to have a de-energise to release CO2 policy. Sometimes people will make different decisions based upon their philosophy or measures of risk.

The system you describe seems to me to be somewhat on the simple side lacking some features I would expect on systems protecting large, expensive and potentially hazardous equipment. However that is not to say that the original designer got it wrong. If your heat tectors are the bi-metallic type they are pretty reliable from a spurious operation point of view.

Check your local/site/company regulations if you have any. Check the GT manufacturers recommendations if they have any. It might help you decide whether your site is making appropriate judgements on risk, or those that have made the decisions for you in the past.

Some useful general references are on fire systems NFPA 72 (US), EN 54 (Europe).

Hope this is helpful. Nothing is risk free.

DaveMH

 

GE fire systems have always been marginal in design and operation. In most Fire Systems you can install a coil monitoring module to interface with the system to know the coil is operable. Note, this only monitors the coil and does not let you know the valve will positively operate.

In many older systems they do use a NC activation that open and activates on "fault"/"sensor". This has/will cause many false discharges of CO2 and often costly damage and down time. Monitoring the VOC's is easy, safe, and the best way to go.

Ernie