Megawatt Transducer Failure

S

Thread Starter

Sadiya

Dear Sir,

We have 3 Gas Turbine Units of 20 MW capacity each. GT - I ann GT - II have MARK - VI and MARK - VIe Control System.

There was some maintenance work going on in UPS (Uninterrupted Power Supply) and power to HMIs of both GT - I and GT - II; Power to MW transducer got switched off by mistake.

The power to MARK - VI and VIe control system was intact. The turbines was running normal. But Generator Breaker got tripped. As HMI was also powered off, we could not find out exact reason for Generator Breaker Trip.

As per the literature in GE manuals, the generator breaker WILL NOT trip on Reverse Power Trip Signal if the MW Transducer is in Faulty Condition.

Can anyone please explain, what must have caused Generator Breaker Trip?

Regards
Sadiya
 
Sadiya,

>As per the literature in GE manuals, the generator breaker
>WILL NOT trip on Reverse Power Trip Signal if the MW
>Transducer is in Faulty Condition.

Specifically, which GE manual did you read this in? Please tell us the GE publication number and section, and please type the passage you are citing.

What is the output of the MW transducer? Is it 4-20 mA for say 0-30 MW, or is it 12 +/-8 mA, for -30M to +30 MW?

If it's the latter then it's highly likely the Mark VIe saw the signal go to 0 mA, which would be -45 MW (for -30 MW to +30 MW), and open the breaker on Reverse Power. If it's the former, it's still likely the Mark VIe saw the signal go to 0 mA, which would be -7.5 MW (for 0-30 MW), and open the breaker on Reverse Power.

Unless the MW transducer was 0-1 mA output transducer (which many older MW transducers were), the signal would NOT go negative when the transducer lost power--but most of the 0-1 mA transducers were self-powered (meaning they took their power from the PT inputs and weren't reliant on a separate source of excitation, such a UPS or station power).

Before I go on a rant (which I am), I want to remind you of a feature of the Mark VIe (not the HMI--but the Mark VIe) that would allow you to download a list of parameters and alarms from the Mark VIe as soon as the HMI was powered-up and available which could be helpful in troubleshooting the problem. Trip History.

Even if the HMI isn't running, there is a section of the RAM (Random Access Memory) in the Mark VIe which is storing data and alarms for retrieval in the event of a turbine trip, but this data can also be retrieved prior to a trip, or after a normal fired shutdown, and reviewed to try to understand the sequence of events.

Also, if the power to the HMIs wasn't lost for more than an hour or so, most GE Mark VI HMIs automatically upload data and alarm lists every hour, storing it in the 'History' section of the HMI. You can use WorkstationST Alarm Viewer to look at this alarm history using the various filters, and you can even print it for even more helpful analysis and review.

Please write back to tell us what the configuration of the MW transducer is.

It's very difficult to inform you of this--but the <i><b>only</i></b> "document" which matters to the Speedtronic turbine control panel (the Mark VIe or the Mark VI or the Mark IV) is the application code and configuration running in the Mark VIe or Mark VI or the sequencing running in the Mark V.

They don't give a hoot what's written in any manual--be it a GE manual or a training manual from GE or a third-party--they control and protect the turbine based on the configuration and programming in the turbine control panel memory, not what's written in some manual or document.

This is a very difficult lesson for many people to learn--that they must learn to "read" the application code or sequencing running in the Speedtronic at their site in order to understand how their turbine is controlled and protected. It can't be done by reading a manual or document--even if it was produced by GE and provided by GE with the turbine or turbine control system. (Some divisions of GE are better about this than others, but, as a general rule the manuals do not always reflect how the Speedtronic turbine control system at a site is programmed.)

Now for the rant.... GE has an <b>extremely</b> bad practice of modifying the existing control and protection sequencing/logic when they provide a new turbine control system to upgrade or retrofit an existing system. Instead of just converting the existing sequencing and configuration from the current control system to the code and configuration required for the new control system, they literally start from scratch and begin with a whole different set of sequencing and configuration.

Why do they do this--even though in some cases it requires more effort to compare the existing sequencing and configuration to the new configuration line by line and make modifications (missing many in the process)? Well, there are three reasons.

First and foremost: the Legal Department. As the OEM of the turbine (if it's a GE-design heavy duty gas turbine) the lawyers have decreed that in order to prevent liability, lawsuits, and bad publicity (negative brand image) they believe GE must provide the latest and greatest sequencing (now called application code) and configuration similar to what would be provided on a new turbine and auxiliaries. And, in most large corporations the Legal Department, followed closely by the Safety Department, rules--and wins almost every argument.

Second, their turbine control system upgrade "process" requires the requisition engineer to begin with a configuration and application code generated by a computer based loosely on what would be provided with a new turbine with a few computer-generated modifications based on some queries answered by the requisition engineer during the process. These queries try to take into account older configurations, such as high-pressure trip solenoids or older starting means, etc. But, they can almost never take into account every configuration found on a site--nor are they designed to.

Third, the sales people selling these turbine control upgrades want (in some cases they demand--yes demand) to be able to tell the prospective purchaser that they are getting all of the latest and greatest improvements to sequencing and configuration for safety and efficiency. They can't tell the prospective customer exactly what these improvements are--but they repeatedly say these improvements can't be purchased from any other control system vendor. In their eyes, this is a distinguishing feature of the GE control system--again, even though no one, NO ONE, can document exactly what these improvements are. The requisition engineer can't. The sales person can't. The commissioning service person can't. No one can.

So, what happens in the field when these new upgraded turbine control systems are installed and commissioned is that changes have to be made to restore many of the site-specific operating sequences and configurations to allow normal operation of the equipment.

Why? Because GE didn't just "translate" the existing sequencing and configuration from the old control system to the new control system--which would have taken into account all of the site-specific configuration and sequencing--and, then make the changes the Legal Department and the Sales & Marketing department demand. Instead, they try to do it bass-ackwards--starting with completely new configuration and application code and modifying it to make it site-specific.

Sound familiar?

And, I suspect that you read this passage in some older, existing manual, and that new MW transducers were supplied and installed and the "new and improved" application code and configuration include breaker tripping on reverse power--which the older sequencing in the turbine control systems did not include.

Again--the only "written documentation" that matters to the Speedtronic turbine control system is that which is running in the microprocessors of the turbine control system.
 
Dear Sir,

The MWATT Transducer installed at our site is operating on 4-20 mA with range configuration as Neg 20 to Pos 40 MW. The literature which I was referring to is from "Control Specifications" Section 9.02.00 "Reverse Power Sequencing" ; Para No 3:

"If the MW signal indicates an excessive negative value, the megawatt transducer is assumed to have failed. The MKVIe reverse power sequencing will be faulted and prevented from opening the breaker. This condition is annunciated in an alarm to the operator."

I have checked the corresponding logic in MARK - VIe. In case the Megawatt Transducer has failed L32DWF (Generator Watts Transducer Failure) bit will be reset and it will prevent Reverse power Generator Trip sequence from execution. L32DWF bit is generated as an output of a comparator block which compares real time value of MW transducer with Neg 68 MW. That means the MW transducer is termed as faulty if the O/P generated by the transducer is less than Neg 68 MW.

So, this, in conjunction with the manual confirms that the breaker will not trip on reverse power generator trip if MW Transducer is in failed condition.

As you have rightly said sir, nothing except the logic implemented in MKVIe is the ultimate documentation for GE Gas Turbines. And it takes some amount of time to decipher those.

I have also checked in WorkStationST AlarmViewer. There are no alarms logged for that particular trip event.

So sir, after reading your reply, can I conclude that, the Generator Watt Transducer is not detected as failed if its power is lost or the O/P is 0 mA. So, this is indeed an erroneous condition leading to breaker trip. The generator breaker is getting tripped on a false reverse Power Condition.

Please confirm this sir.

Thanks a lot for the reply.

Regards
Sadiya
 
Sadiya,

The total range for -20 MW to +40 MW is 60 MW. One-quarter (4 mA) of that range is 15 MW. When the MW transducer is at 0 mA, the Mark VIe will indicate -35 MW (-20 MW - 15 MW = -35 MW). The value of the failed Constant is much too negative; it must be less than -20MW, but more than -35 MW, probably a value of -30 MW would be a better choice. This should have been caught by the Commissioning field service person.

Using WorkstationST Alarm Viewercan take some practice. If the Mark VIe opened the generator breaker--and it would have if the MW failure Constant was set to -68 MW, there should have been a Process Alarm. Sometimes it's necessary to open an Alarm History file using WorkstationST Alarm Viewer directly by double-clicking on the file from Windows Explorer (NOT Internet Explorer--Windows Explorer.)

Is it possible that some other protective relay function on the Generator Control/Protection Panel opened the generator breaker and no one noticed the flag? (Not likely, but not impossible, either.)

Hope this helps!
 
Any trip must have set your alarm, either generically in HMI, but must remain in the protective relay, up to the relay 86 is not reset.

For if it can be of use, we had a trip as synchronizing appeared alarm " ALM FSR LIMITED DUE TO FAILURE DWAT Xducer " and " ALM MEGAWATT TRANSDUCER SIGNAL TROUBLE " for failure Dwatt gross energy transducer. FSR control stays in ACC when it should happen to SPEED, to perform Master Reset, transuctor failure is normalized and control passes 30% (ACC ) to 50% (SPEED), causing a sudden opening of valve gas control and trip by EXHAUST OVER TEMPERATURE.
 
elomoon,

According to the original post, the <i>turbine</i> didn't trip, the generator breaker was opened "(tripped"). That's an important distinction. Loss of the MW transducer based on the information provided will only result in opening (tripping) the generator breaker, not in tripping the turbine (UNLESS the fuel gets cut back so quickly that flame is lost and the unit is tripped on loss of flame--which the original poster did not mention).

I believe your alarms are from an F-class turbine, and this is a Frame 5 (more like a B-class) turbine. Not all alarms are the same in all units--F-class have many alarms which are more particular to DLN (Dry Low NOx) combustion systems, which have their own peculiarities and idiosyncracies.
 
CSA Sir,

Thanks for clarifying... The feedback from Electrical engineers on site is that the Generator breaker was not tripped from generator breaker panel. Atleast there was no alarm.

That means transducer failure alarm configuration at our site is faulty. I will take care of that. This discussion has helped us understand configurations better. Thank you the support provided sir.

Regards
Sadiya
 
Sadiya,

Thank you for the feedback! We like to say here at the control.com forum, "Feedback is the most important contribution!"(c) because it lets others who may read this thread next week, next month, next year--or three years from now--if you found the information useful (or not, as the case may be).

And, it's good to let the responders who provided the information know if the information was useful (or not, as the case may be). It's really a win-win-win situation all around when people provide feedback--so thank you for the feedback!
 
We are facing problem of load limiting due to " FSR limited due to DWATT transducer failure" in Frame V gas turbine. The load doesn't increase in turbine beyond 30% FSR, and on increasing the load, it trips on underfrequency. Also, we have to manually " Master reset" to increase loading on turbine and avoid underfrequency.

Can we do any changes to avoid underfrequency tripping?

Regards
Awaneesh
 
Awaneesh,

I don't have any software to review at this time to address the specific alarm you mentioned, but when a gas turbine is operating below rated speed the air flow through the axial compressor is also less than rsted. This results in higher than normal exhaust temperatures for the same fuel flow which might be limiting the fuel flow-rate because the exhaust temperature is at the exhaust temperature limit. If that's true the unit load cannot be increased to help the grid frequency.

It's one of the "dirty little secrets" of gas turbines that when the grid frequency is much less than rated (as it sounds like it is at your site) the power output of a gas turbine is limited because of the reduced air flow through the axial compressor. And while this is not desirable it is unavoidable. Your turbine's speed is a function of grid frequency when the unit is <b>synchronized</b> to the grid. Your turbine can't run any faster or slower than the speed dictated by the grid frequency. And when the grid frequency is low the air flow through the axial compressor will be reduced which may cause the unit output to be reduced because of a higher than normal exhaust temperature preventing the turbine from putting out more power and in some cases even reducing the power output to protect against exhaust overtemperature.

The underfrequency setting is set to protect BOTH the turbine & axial compressor as well as the generator. It should not be changed without the OEM's consultation and input.

When I can get a chance to investigate the specific alarm you mentioned I will and get back with a description of the condition it is warning the operator about. But I suspect the root cause of inability to load, or at least the underfrequency tripping, is the low grid frequency which you can't do much to help.
 
Hello:

In our unit, the FSR was limited to 30%. When the alarm ALM FSR LIMITED DUE TO DWAT XDUCER FAILURE / ALM MEGAWATT TRANSDUCER SIGNAL TROUBLE appeared, the cause was the failure in one of the DWATT gross energy transducers.

When synchronizing the unit, it remains in the correct ACC command, until it reaches 30%, limited by the DWATT transducer failure. From this point the control of the FSR should have passed to SPEED control and continue its normal loading process.

When carrying out Master Reset, the DWATT transducer failure is normalized, but the FSR control goes from 30% (ACC) to 50% (SPEED), the required command. This action causes a non-stepped opening of the SRV, acting as protection for the TG, Trip of the unit, in our case due to high temperature in the exhaust.
 
Top