Issue: There were three cards (TCD1, TCD2 and TCE1) in A2 state in R core of one of our GT, i.e., GT1. After 3 days TCE1 came in A7 state after pushing IO net cable connections in R core, P core, QD1 and QD2 core.

We have again checked IO states in R core and found that TCE1 again came back in A2 state and TCD1 came in A2 state while TCD2 were remain in A2 state .

After 3 days by itself + Master reset TCE1and TCD1 again came back in A7 state and is remaining in A7 since 1 week.

Now we have TCD2 card in A2 state. Two queries.

1. Is it advisable to replace TCD2 card on line by switching off the power supply of TCD2 in QD2. Any negative impact?

2. The problem is repeating. Can someone share the experience?

We are a process plant and steam limitation from HRSG can affect process.

 

Sunil,

<b>When did this problem start?</b> After one of the cards you listed was replaced? After some other cards were replaced in <R>?

It certainly seems there is some loose or intermittent connection of the IONET cabling for <R>, which you have discovered runs from <R> (the TCQC if I recall correctly; I don't have access to a Mark V Application or Maintenance Manual at this writing) to Loc. 1 (TCEA) in <P>, then to Loc. 1 in <QD1> (TCD1), then to Loc. 1 in <QD2> (TCD2). The IONET termination resistors on the TCDA in Loc. 1 of <QD1> should be in the OUT position, and they should be in the IN position of the TCDA in Loc. 1 of <QD2>. (Consult the Mark V Application Manual for details.)

It is also common for corrosion to develop on the pins and cable connectors of Mark V cards. Have you tried (with the unit NOT running) unplugging and plugging the IONET cables at each point in the string several times to try to improve the electrical contact/connectivity? GE also recommends the use of conductive grease on all the connectors, and that grease should be sparingly applied (too much conductive grease is just as bad as none), and regularly applied (approximately every year or two) to ALL the connectors of the Mark V--and there are a LOT of connectors!

The power for the TCDA in Loc. 1 (TCD2) in <QD2> comes from the TCPS in <R>, so the only way to remove the power to that card is to power <R> down. I do not recommend ever replacing Mark V cards with power being applied to the card.

So, check the IONET termination resistors on the TCDA cards in Loc. 1 of <QD1> and <QD2> to make sure they are in the correct positions. I think there are also some "address" jumpers on the TCDA cards which must be in the proper locations, and which might also be causing problems if they are not set correctly. (Again, I do not have access to Mark V manuals at this writing.) Essentially, all of the TCDA cards in <QD1> must have all of the hardware ("Berg") jumpers in the same positions. Also, all of the hardware jumpers on the TCDA cards in <QD1> must essentially all be in the same position. <b>HOWEVER,</b> the jumper positions DIFFER slightly between <QD1> and <QD2>, so be careful! I do NOT recommend consulting any documentation when replacing cards and setting hardware jumper positions on the new cards. I ALWAYS recommend placing the hardware jumpers on the new card in the SAME positions as they were found on the old card--which was working fine before it needed to be replaced. Commissioning personnel rarely edit or correct any hardware jumper documentation supplied with a Mark V, and the hardware jumper documentation supplied by GE is almost NEVER specific to the site. So, any hardware jumper documentation (including any displays on the operator interface/HMI) is almost never to be trusted. It can be consulted when replacing cards, but if discrepancies are noted the as-found settings on the card being replaced wins!

Also, if you have not applied any conductive grease to the connectors in the Mark V panel, or if it's been more than a couple of years since conductive grease was applied, it is recommended to shut the unit down at the earliest possible convenience and properly apply conductive grease to all the cable connectors.

Please write back to let us know how you fare with your troubleshooting!

 

Dear Mr CSA,

Sorry for the delay. We had a process plant trip in lieu of Gas turbine trip and production line re establishment was critical

The actions we had taken on GT A are: We had replaced TCQA card, TCD2 cards of "R". Later TCQA card, was replaced on line at low load.

The trip logger alarm indicated Lube oil thermocouple #1fault, Compressor inlet thermo couple disagree. This generated a doubt of communication failure. Also time stamping sequence is FAGR voted value has shown variation, then P2 low alarm, load reduction, flame failure ,and consequent TRIP.

Unplugging and plugging the IONET cables at each point conductive grease application on all the connectors were also carried out.

Still the problem is repeating. Now TCD1,TCD2, and TCEA card is indicating in A2 status, where as R computer displays status A7. Alarms first appeared is DCC IO card objID #4 resetL63 QAL, L63HQ1L, L27MC2Nat same time (08.04.32.000). The same got reset. At 09.34.33 hrs L28FDE (Alarm ID 1828) & TNH_OS ( Alarm ID 3447) voter mismatch alarm appeared and DCC No IO card for socket exists (Alarm ID0017) appeared. Later at 11.32.11 hrs DCC I/O card went to A3 state alarm appeared, TCE1 TMR trouble ETR1& ETR2 -(Alarm ID 1696 & 1697) appeared. This is status right now.

The problem still prevails

 

Sunil,

The Mark V Maintenance Manual, GEH-5980, shows how the IONETs of <R>, <S> and <T> are connected. The IONET runs from the DCC/SDCC, to the TCQC, to the associated TCEA, then to the associated TCDA, and if there's a <QD2> to the associated TCDA in <QD2>.

You are taking a huge risk changing cards while the processor is powered-up. It's very easy to misconnect ribbon cables and short or ground something out.

If I understand your description all the cards that are not at A7 are on the IONET, from the TCQC in Loc. 4 of <R>, to the TCEA in Loc. 1 of <P>, to the TCDA in Loc. 1 of <QD1>, to the TCDA in Loc. 1 of <QD2>. So, there's either something amiss with the IONET termination resistors along the IONET, or there is a problem with the IONET cable(s), most likely between the DCC/SDCC on <R> and the TCQC in <R>, or between the TCQC in <R> and the TCEA in Loc. 1 of <P>.

I recommend you check and confirm the IONET termination resistor jumper settings all along the IONET of <R>. I also suggest you check the IONET address jumpers of the two TCDA cards in <QD1> and <QD2>. One of the Appendices of the Mark V Application Manual, GEH-6195, lists and describes all of the jumper settings.

If the jumper settings are all in the proper places, I would then suggest using an ohmmeter to check each conductor of each section of IONET cable for continuity, and to make sure each conductors is isolate from each other and from ground.

If that checks out, I would suggest replacing the TCQC card in <R> first. You will need to re-calibrate all associated LVDTs if you do that.

If that doesn't work, I would suggest replacing the DCC/SDCC card. There's a LOT involved in replacing <R>s DCC/SDCC card.... You have to set the Stage Link ID, and you have to set the Voter ID. It might be easier to move the EEPROM (U9, I believe) from the existing card to the new card, because I believe all those settings are stored in EEPROM.

But, it's pretty certain there's something amiss with the IONET--termination resistor jumper positions, the cable(s), or the cards.

<b>Based on the information provided,</b> that's all I can suggest.

Please write back to let us know how you resolve the problem.

 

Dear sir,

Sorry for delay. Finally we had taken shutdown and replaced the TCQC card. Also we had observed that humidity in the compartment was a bit on higher side. As a preventive action we are planning to provide air lock with door in front of TCC door.

 

Thank you for the feedback!