Having difficulties to synchronize the Gas Turbine Generator (Frame 9, MK VI) with the grid. All permissives are met but the Auto synch permissive is not making -L83AS. Also, L27BN is 0 but other permissives like the HP speed signal, Generator Differential voltage are 1. Let me note here that there is actually voltage on the bus as recorded by the bus PT's.

 

Can you let us know why the genset was down? Was there any maintenance or other during period it was last stopped? What caused it to stop on its last stop?

 

Yes. There was maintenance carried out as a result of cable earth fault. The fault has been rectified but the 52G is unable to synch to the grid. The breaker closure and opening has been tested and its working fine to prove that the breaker interlock is not operational. The GT was on normal stop before the cable earth fault was seen on the ELIN relay panel - which is used for Generator protection and Excitation.

 

If the Buss PT's are showing OK output but the 27 permissive is 0, there must be some problem between the output of the relevant PT and the "zero" logic. Can you measure the voltage going into the relevant logic converter? Which device converts the voltage to logic? Have you checked the input and output from this device?

 

I'm going to take exception to the statement that all permissives are met. L83AS is just about <b>the</b> most important permissive. It's basically the permissive that's set when the operator selects AUTO synch.

Have you used Toolbox to determine why L83AS isn't picked up?

L27BN is <b>usually</b> driven by a discrete (contact) input from an electromagnetic relay. Have you investigated why L27BN isn't picked up? (It usually signifies the running (bus) PTs are sensing voltage.) There might be an analogue input for the bus PTs but if there is a discrete input that's not satisfied then, ..., well, ....

Lastly, when there is no bus voltage from L27BN there is usually a Process Alarm indicating that Auto Synch is not possible because there is no bus voltage. What Process Alarms are active when you are trying to Auto Synch?

 

Actually, its from the toolbox that i noticed L27Bn was 0 and not 1, hence blocking the Auto-Synch permissive.

The discrete (contact) input from an electromagnetic relay module in this case is a TM 1703 MIC terminal module (ELIN RELAY used for Generator and transformer protection) with 6 binary input, 2 relay output and some I/O modules etc.

When there is no Bus voltage, "Bus undervoltage - NO Auto Synch" alarm is annunciated.

On the relay module point for L27BN, a close signal is measured; showing that there is actually voltage on the BUS but the LED indication states otherwise.

> (It usually signifies the running (bus) PTs are sensing voltage.) There might be an analogue input for the bus PTs but if
> there is a discrete input that's not satisfied then, ..., well, ....

> Lastly, when there is no bus voltage from L27BN there is usually a Process Alarm indicating that Auto Synch is not
> possible because there is no bus voltage. What Process Alarms are active when you are trying to Auto Synch?

 

Are all of the other modules in the ELIN system working OK?

For the 27bn, is there only one input to the ELIN logic for this (Bus PT) or is there any other?

If all other modules are working OK AND if there is only one onput to the 27bn, then ......

If I am right in saying that the relay module shows closed but the LED shows otherwise and you are getting a 0 out, then it would appear that the relay module is not working correctly. Are there spare (unused) module points in the ELIN setup that could be used?

 

Ekemini...

1) Has the protective rely that operated to clear the cable earth-fault been reset?

2) Has the Electrical Fault-Lockout Relay been reset?

3) If provided has the close-relay circuit supply fuse been checked?

4) Upon meeting all perquisite conditions for synchronization does the 'close' signal reach the close-relay?

Regards, Phil Corso

 

L27BN is typically a non-inverted ("normal") discrete (contact) input to the Speedtronic. The device output contacts should be closed when the relay senses a proper level of bus (running) voltage available to synchronize to/with.

Presuming your Mark VI uses 125 VDC for discrete input wetting voltage--and there are no 125 VDC battery grounds on the system!--you should be able to measure nominal +65 VDC with respect to ground at <b>both</b> device terminals of the relay providing the signal to the Mark VI, if the device contacts are indeed closed. (You should probably carefully remove one or both of the wires from the relay output terminals and use an ohmmeter to determine if the contacts are closed or open. The Speedtronic discrete input circuits are current-limited, but you should try to prevent the wires from coming in contact with ground or from touching any other terminals when they are loosened from the device.)

You should also be able to measure +65 VDC with respect to ground at both TBCI (presuming a TBCI is being used for discrete inputs on your Mark VI) terminals of the L27BN input--presuming there is no Diagnostic Alarm associated with the L27BN input or the VCRC (presumed) being used for the L27BN input.

If you can't measure +65 VDC with respect to ground at all four terminals of the L27BN circuit, and there are no Diag. Alarms associated with the input circuit/card, and you determine the contacts are, in fact, closed, then there is some wiring problem to be sorted and resolved.

L83AS is usually a logic output from a Command State block. When the associated PERMn input logic is TRUE (logic "1") AND the operator selects AUTO SYNCH from the HMI, the BBL will set L83AS to logic "1". (NOTE: No other "prior" PRSETn value can be a logic "1", or it will prevent the block from setting L83AS to logic "1". There should be 'Help' in Toolbox for the Command State block.)

If you would report all of the logic values of the PRSETn <b>and</b> inputs to that Command State Block when at FSNL, in order, we can probably help you determine if that's the problem. Provide the logic signal names associated with each of the inputs (PERMn and PRSETn) and their values if L83AS can't be set to logic "1" when the operator clicks on AUTO SYNCH on the HMI.

L3, Complete Sequence, is also usually required for synchronization (Auto or Manual). To achieve Complete Sequence it is necessary for the Aux. Pumps (L.O. and Hydraulic) to be off, and for the unit to be at FSNL, and for the unit to be on Droop Speed Control (L30F_ND, if I recall correctly, should be a logic "1"), and for the Emer. L.O. Pump to be off. Is L3 a logic "1" when you are trying to Auto Synch?

Can you provide a comprehensive (meaning all) list of alarms--Process and Diagnostic--alarm which are active when you are trying to Auto Synch?

Please write back to let us know how the problem is resolved, or if you need further assistance.

 

All other modules are working correctly and with the correct indication too. The Excitation Engineer is yet to look into the GCP for now.

 

The protective relay has been reset but we can't check for the "close" signal because the permissives has not been met for a breaker closure.

 

The Vdc split is okay.

The L3-complete sequence is also okay as the GT always attains FSNL before synchronization can be carried out. thus the auxiliary, hydraulic and emergency lube oil pumps are offline.

When attempting to close the breaker, the alarm "VTUR- Breaker #1 slower than adjustment limit allows" is annunciated.

The following signal/permissives must be 1 before a synch can be carried out: L52Z- Auto sync Enable, L52SX- Auto Synchronizing permissive, L3synch- Auto sync permissive, L14HS- HP operating speed signal, L27BN- AC Bus nornal, L3DV_Err - Generator diff voltage synch permissive, L1FX- Start up shut down Aux. to fast load start signal, L83s_Auto - Gen breaker synch mode-Auto, L3SFDIFF - system frequency diff (slip) synch permissive.

All the above stated signal were 1 except for L27BN, hence L83AS was 0.

The ELIN GCP protective panel has microprocessor board, pulse generation and signal processing board.

 

Ekemini,

Thanks for the feedback.

It would certainly seem the problem is with the L27BN input to the Mark VI. Again, that input is usually not inverted, and should be closed when the bus voltage is "Normal", and open when the bus voltage is abnormal. (Most Speedtronic logic signal names indicates when it will be a logic "1"; in this case, L27BN will be a logic "1" when the <b>B</b>us voltage is <b>N</b>ormal.)

Have you tried temporarily jumpering the Elin relay output terminals while observing the L27BN logic state using Toolbox? That would confirm if the wiring between the Elin output and the Mark VI input are good. You could do this anytime, but I wouldn't recommend jumpering this permanently to allow synchronization. This is just a wiring check, and basically to prove that the problem is the Elin relay output or something to do with the bus voltage sensing driving the output.

As far as the 'close time out of limits alarm', this is a very "common" nuisance alarm for most Mark VI and Mark VIe installations--especially if the Mark VI/VIe was provided as a retrofit of an older control system. The Mark VI/VIe must have a direct-acting generator breaker contact input (to the TTUR card if I recall correctly) in order to properly sense breaker closing. Many installations do not have such a contact available. The contact MUST come directly from the generator breaker; it can't come from an auxiliary relay. <b>AND</b>, the breaker MUST be a stored-energy type breaker. Many start-up TAs don't understand these requirements, and don't disable the input. And, some versions of firmware for BOTH the Mark VI and Mark VIe can't be totally disabled. So, a work-around using an interposing relay can be done as part of the disabling of the feature.

[The feature being talked about above is called "breaker self-adapt correction." It is <b>***NOT required***</b> for every generator drive application, and can only be applied when the two criteria above are <b>both</b> met. Unfortunately, GE and its packagers and TAs don't all understand that, and MANY Customers mistakenly believe it <b>IS</b> required and so resist attempts to disable it and insist it be made to work, when many times it can't be made to work because the criteria can't be met.]

 

That's great info.

I still think the problem is in the ELIN system or upstream of it. What are the inputs to the processor enabling "normal" bus voltage out?

 

Quote:
"On the relay module point for L27BN, a close signal is measured; showing that there is actually voltage on the BUS but the LED indication states otherwise."

If you could measure the contact output at the relay that the contact is CLOSED when GT at FSNL, we can be assured that the relay works fine (provided that the contact is measured to be OPEN when GT is OFFLINE), please measure the contact that goes to the Mark VI panel. The LED indication might just be fused LED.

If the contact is also CLOSED, there can be few possibilities:

(i) Bad contact input point on the IO card. Replace if necessary.

(ii)Bad termination. Try to remove the termination and terminate again. (I personally have experienced similar problem for my turbine synchronization problem previously. Simple remove and re-termination procedure solved the 2 days unable to synchronize problem )

If the contact is measured to be OPEN, there can be few possibilities:
(i) Bad cabling (open circuit)
(ii) Bad termination problem.

Happy troubleshooting and update us the progress. Good Luck

 

To add to this, and to build on my earlier recommendation of jumpering the Elin relay output terminals while watching the L27BN logic in Toolbox, there could be a problem with the discrete (contact) input point or module, also. Though there would likely be a Diagnostic Alarm to that effect, there is the odd occasion when there isn't a Diagnostic Alarm when there should be one.

It will be very nice to hear the resolution on this problem.

 

> Happy troubleshooting and update us the progress. Good Luck

Thanks for your directives. The PT's input that is terminated at the ELIN processor is okay, but the output gives nothing. I think the issue lies within the ELIN processor - kind of re-configuration of the module and troubleshooting the processor is needed. If a jumper is placed at the module output terminal, the MK VI will have a 1 signal (with the false LED indication).

 

Yes. I think the issue is with the ELIN processor because the Wiring from the module to MK VI is very fine, the input from the Bus PT's are correct too. The processor is not giving out the right signal to MK VI - needs an Excitation Engineer to sort it out.

 

Thank you for the update! 'Feedback is the most important contribution!'(c) here at control.com; it's what makes these threads useful for many years after the original problem was resolved.

I'm presuming the "false LED" indication is at the ELIN module output.... I have seen some Mark VI discrete input VME cards with LEDs to indicate status, but, honestly I've never tried to use them or to equate them with inversion status (meaning if the LED will be lit when the contact is open if the input is inverted in Toolbox; or if the LED will be lit if the contact is closed and the input is not inverted ('Normal') in Toolbox).

As was said before, the important thing is that if L27BN is 'Normal' (not inverted) and you jumper the input to the Mark VI that the signal change from logic "0" to logic "1" when the jumper is applied to the terminals, and changes back to logic "0" when the jumper is removed. Again, the usual configuration for L27BN is to be not inverted, and to be a logic "1" to indicate the presence of bus voltage so that synchronization can be enabled.

Thanks again for the update. Please write back to let us know when you are successfully able to synchronize the unit to the grid, and what the final resolution was.