We are upgrading GE gas turbine control system from Mark IV to VIe
GE is proposing Redundant transmitters for FG and CD.
Could some one explain why we need Redundancy ?

 

Dear Gladymady,

typically the preferred arrangement is to have three transmitters so that each core (R,S,and T)can have its own transmitter to look at.

In the old days of MKIV there was limited I/O and the prices for transmitters were much higher, as I am sure you are aware. Back then GE would typically use one transmitter for CPD, and then fan that signal out to all three control cores. If this one transmitter was out of calibration, or sending a skewed signal that still looked good, turbine protection would be adversely effected, NOT GOOD!
Basically using 3 high quality transmitters for CPD and fuel gas adds to the reliability of a TMR(Triple Modular Redundant) controller setup, and also increases the turbine protection capabilities of control inputs using these signals.

 

Is it only for redundancy? If you've been operating successfully with single transmitters on a Mark IV TMR system, you can likely continue to do so with a Mark VIe TMR system, presuming you have conventional, diffusion flame combustors.

However, some of the older P2 and CPD transmitters used a differential power supply (+12V/-12V) instead of a +24 VDC power supply. While some Mark VIe terminal boards and power supplies are capable of providing the differential power required, some are not.

Those transmitters are also dated, as many of them are 0-5 VDC output, instead of the more common 4-20 mA. While you may have spares of the older transmitters now, they are getting increasingly expensive and more difficult to obtain.

Presuming you have the older transmitters, you could continue using them until your spares are depleted and then obtain the newer 4-20 mA transmitters. But, then you'd have to change Mark VIe configurations and some people aren't comfortable doing that and also don't want to pay GE to come and do it for them.

So, it's really up to you. The Mark VI was not capable of providing a differential power supply and so new transmitters were pretty much required (though GE found that out the hard way). If you want to continue to use the older transmitters, you just need to be aware that it might cost you a little more to make sure the Mark VIe can provide the required power, and that purchasing them may get even more expensive in the future.

One word of warning: It's not acceptable to use SMART transmitters for P2 and CPD pressure transmitters. They just aren't fast enough for the applications. So, you can't just use any old 4-20 mA transmitter in the future if you decide to continue to use the old transmitters now and want or need to change to different transmitters later. If you use a SMART transmitter for P2 or CPD sensing, it's likely you could introduce some instability in the turbine operation due to the lag inherent in the SMART transmitters.

There is something to be said for redundancy, though. If you have single P2 and CPD transmitters now, and either of them fails, the unit will trip. With redundant transmitters for either or both applications, a single failure won't result in a trip of the turbine (and lost production). If you have a TMR Mark IV panel with non-redundant transmitters, it's kind of defeating the purpose of a redundant control system. But, having said that, there are many sites, and yours may be one of them, that have operated for decades with non-redundant transmitters with no problems.

So, that about sums up your choices: Continue to use the older transmitters, and perhaps pay a little more for the Mark VIe, which can be modified to use newer transmitters later. Or, upgrade to the newer transmitters now, and pay the additional cost for the new transmitters (and spares) now, and not have to make any configuration- or transmitter changes later.

Redundancy is also a choice, and it depends on how much you value reliability versus upfront cost in light of your past experience with non-redundant transmitters.

Isn't this stuff fun? There's never a simple answer, is there?

 

It is possible that the Mark IV is simplex (maybe it was an upgrade to Mark I or Mark II or older), which would explain the single transmitters.

But if so, why wouldn't you go to Mark VIe simplex instead of TMR?

 

I guess I would question why the need for redundant sensors on a SIMPLEX panel.

There are many TMR Mark IV-equipped units with single P2 and/or CPD transducers. I think most of them were shipped during the early years of Mark IV production. Never seemed to make much sense on a TMR control panel, but they're out there, and many of them are still in service today.

As MIKEVI says, back in the day transmitters were pretty expensive.

MIKEVI also makes a very valid point about single transmitters being out of calibration and causing problems with power output or parts life. So, it's not just the potential failure of a single transmitter that should be considered, but also the potential for an out-of-calibration transmitter.

Isn't this stuff fun? (Really, isn't it fun?!!???!!)

 

Thank you so much...

 

Hello over there....

As many of the contributors explained, there are good reasons to use single robust CPD transmitter with any MKx family. There are also MKVI panels with single CPD transmitters with analog output, 4...20mA. The design philosophy for the new generation MKVI family is based on process control philosophy. Beside redundancy, the MKVI family algorithms uses the so called median select blocks which will take any time the median value of the CPD transmitters which will result in improved process control of the CPD (hopefully). Verify with the vendor how they have assigned the CPD measurement ,as median select or voted (unlikely). For power generation application, the MW transmitter should be back up for the CPD transmitter.
Same philosophy apply for the FG transmitters.
Good Luck....

A. Oztas