26-QT High lube oil temp switchs have malfuctioned a couple of time and our TG got tripped through PIM module. 26-QT is directly terminated on TBD of PIM along with other causes also e.g low hydraulic oil pressure etc. Six inputs AND logic (implemented by wires in series on TBD PIM) then give a single contact to <P>. Whenever TG trips we are completely blind that which switch among these 6 actuated.

Therefore initially we decided to install a RTD sensor in place of 26-Qt (which has max. frequency of failure) which will be terminated in <C> and through soft logic a alarm and a contact will be generated through <CD> to PIM TBD. All Other cabeling will remain as it is. Then afterwards we will do the same with other causes of PIM.

Plz let me know wether it is safe to configure a analog trip input on <C> core and getting trip contact through CD core. I have implemented the complete logic, alarms and simulated. Can i configure an RTD in <R> which seem not likely. Is there some other method also.

Best Regards

 

You haven't told us what version of Speedtronic turbine control panel you have, nor whether it's a SIMPLEX or TMR panel.

The fact the turbine control panel contains a TPIM (Turbine Protective Interface Module) suggests the current version of the Speedtronic was an upgrade from a previous version. You also seem to be indicating there is only one single 26QT temp switch, which would also indicate the turbine and auxiliaries are older and/or were originally packaged by someone other than GE, because it was typically GE's practice when packaging turbines to use redundant devices to generate a trip to prevent a single point of failure from causing a nuisance turbine trip such as you seem to be experiencing.

If the turbine control panel is a SIMPLEX panel, then connecting a trip input to <C>, while not recommended, doesn't really adversely affect turbine protection as for most Mark IVs, Mark Vs, and Mark IVs if <C> fails the unit will trip.

If the turbine control panel is a TMR panel, it's not advisable to connect an input that can trip the unit to <C>. There are exceptions to every rule, and if you only have an RTD to connect to the Speedtronic, and the turbine control panel is a Mark V (which it seems to be from your post), then you have no choice but to connect it to <C>.

If the unit has a 26QA switch (high L.O. temp alarm) switch, can you use the 26QT in parallel with 26QT in the trip circuit to ensure that the L.O. temp must be above the alarm and above the trip levels before the unit is tripped? (Of course you want to have a contact from the alarm switch connected to the turbine control panel to warn of an impending problem. Perhaps replacing both switches with ones that have two contact outputs each would be best. Use one contact from each as individual inputs to <Q> to generate alarms on high- and high-high temperature (to indicate activation of the switch(es)). Use the other contact of each in the trip string (in parallel).

I believe there are even devices which have 4-20 mA outputs and contact outputs which can be used to provide running indication of L.O. header temperature and a contact which could be used to alarm or trip the unit. These probably aren't inexpensive, but neither are nuisance turbine trips.

Finally, when posting questions it's best to provide as much information as you can about the configuration of the control system at your site in order to receive the quickest and most concise response.

 

Dear CSA,

We have SIMPLEX MARKV system which was upgraded from MARKII system on MS5001 Turbine. The dual contact needs an external supply in the field. Thats why we are not going for these switch. On the trainer, which is copy of our running system, when i turned off <C> it generated following ALARMS ON <BOI>

1) 248 l3combt_alm <C> TO <R> COMMUNICATION FAILURE TRIP

2)62 L3COM_B_ALM COMMON IO COMMUNICATION LOSS

 

A Mark V SIMPLEX is *not designed* to keep a unit running when <C> is switched off.

A Mark V SIMPLEX Speedtronic control system *should* generate these alarms on the <BOI> when <C> is switched off. The two alarms you listed are true and proper indications that communications between <C> and <R> have been interrupted and the I/O status from <C> has been lost because <C> was powered down.

Connecting inputs to the <C> processor of a SIMPLEX Mark V turbine control panel that can trip the turbine is sometimes unavoidable while not desirable. Since there is no redundancy on a SIMPLEX to speak of, it's not unheard of to connect trip inputs to the <C> processor of a SIMPLEX Mark V turbine control panel.

A properly configured Mark V SIMPLEX heavy duty gas turbine control system should *not* be able to keep a turbine running without the <C> processor. So, I'm confused about the confusion.

 

With regards to your statement, "Whenever TG trips we are completely blind that which switch among these 6 actuated", there is a simple solution to this problem. There is an input module that is a standalone DIN rail mountable unit that is designed to accept the inputs from any field condition sensor via existing wiring system that monitors wiring between the module and sensor through a unique diagnostic process and is able to distinguish between an actual condition (normal or active) and a fault condition - e.g. open, short or ground. When the switch functions normally, a normally open contact in the module closes, passing the signal to the host equipment. In the event of a fault, a normally closed contact opens, buffering the fault from false alarming and pinpoints the type of problem indicated on the color coded LED display. This would solve you problem of determining which switch actuated or is causing faults. If you need additional information feel free to contact me offline at tim at elusivesolutions. com.