We are facing a problem in one of the Mark V system for our Frame V GTG. The <C> computer is showing A8 error and not booting. We are not able to access 186 monitor from the LCC display. However 196 monitor display is accessible from the LCC display. As I/O status is available from 186 monitor display further trouble shooting is not possible.

We have replaced the SDCC, SLCC and TCCA Cards with new ones, but unable to proceed further (the error message is still showing as A8). Due to the above error, Mark V is not communicating with the HMI thru ARC net.

Remaining computers, R, S & T are ok (A7 status).

Request to provide trouble shooting help, if any.

 

What does the SLCC display show as the Voter ID for <C>? If it says B instead of C, then the SDCC card hasn't told it's in <C>. See the following thread:

http://control.com/thread/1296512618

If the Voter ID wasn't set, it's a safe bet that the the StageLink ID wasn't set, either. You'll need to get the StageLinkID of <C> from F:\CONFIG.DAT, and then use the SLCC display keypad, following the instructions in GEH-5980 in the section immediately prior to the one in the above thread (if I remember correctly) to set the StageLinkID of <C> so that the HMI can communicate with it.

I'm sorry to ask, but are you certain all of the hardware ("Berg") jumpers were set correctly on the SDCC, SLCC, and TCCA cards? Are you certain all of the cables have been firmly seated in their card connectors, and with the proper orientation (remember, the trace conductor of the cable goes toward Pin 1 of the connector). The 3PL cable that connects the SDCC, SLCC and TCCA (and TCCB if present, if I recall correctly) is notorious for failing intermittently. It's the only cable which doesn't have pull tabs and it's very easy to disturb the crimp connectors of the 3PL cable if it's not pulled gently and easily out of the connectors. So make sure all of the connectors of 3PL are very firmly pressed together and are firmly seated in their card connectors.

Do all of this with the power off of <C>.

If all of that has been done, then the only other two cards which might be in or associated with <C> are a TCCB card (which is an optional card, which if present would be in Loc. 3 of <C>), and the TCCA card in Loc. 1 of <CD>. You could try swapping either of those cards because they are the only other two cards which would be present in the 186 Monitor of <C>.

Please write back to let us know the result.

Lastly, you have a GE-design Frame <b>5</b> heavy duty gas turbine, with a Mark V control system. The turbines are designated with English numerals; the Mark* control systems are designated with Roman numerals. (Just to make things nice and consistent, as GE is wont to do.)

 

This is a bit of a nasty one..!! A8 is an I/O card initialization failure, basically one of the I/O cards doesn't match what is in the I/O config.for <C> Have you changed anything in <CD> core? Can you try the <CD> core cards from another panel in this panel?

 

Thank you CSA for the reply.

1. We do not have TCCB card. I have already swapped the TCCA card with a new one as well as with a working one from <R> and the status is same in <C>.

2. I have checked the berg jumpers in SDCC, SLCC and TCCA, they are correct, as per the original configuration. (the machine was running, we only powered off Mark V during the combustion inspection and powered on after completion of CI).

3. Regarding the 3PL cable and Voter ID, shall come back after checking.

 

As Bob Johnston has said, A8 is a pretty serious failure/indication. I was just trying to make sure that all the jumpers are in the right place, the cables are properly oriented and connected, and that the StageLink- and Voter IDs are all set correctly. This is basically the easy stuff, but it's worth checking.

I've also seen this problem occur when one or more of the fuses of the TCPS power supply card in location 5 of <C> have opened/failed. Without proper voltages, cards cannot properly initialize. So, while it's not easy to check the fuses, it's probably a good idea to do so. There should be some sheets of paper on the back of the <C> core door, and one of those sheets should show the locations of the fuses on the TCPS card. One of them is particularly difficult to see, and so a good thing to do is to remove the TCPS card, inspect all the fuses, then replace the card, making sure to firmly re-attach the earth lead!

Lastly, I've seen bent or shorted pins of the PROMs which must be moved from the card being replaced to the new card cause problems. Also, sometimes the chip sockets used are larger than the chips, so the placement of the chips on the socket is important. If I remember correctly, if the socket is smaller than the chip, the chip "notch" is not placed directly over the socket notch; the chip must be inserted with the non-notched end flush/in line with the non-notched socket end.

Also, some of the major PROMsets are comprised of two chips which must be in the proper sockets.

I'm just trying to make sure all the major and minor things have been covered here. I would have to say that if the problem occurred during power-up after an extended power down during an outage, that blown power supply fuses would be a good thing to check.

Also, have you checked the DC power at the PTB on the <PD> core? Specifically, have you checked to make sure the 125 VDC power is above 90 VDC and there are no grounds?

There's one last thing worth mentioning, though I've only seem it happen one time. On the DCC/SDCC card is a very important, socketed EEPROM chip that *DOES NOT* have a label on it. This is *THE* EEPROM chip, the one the EEPROM Downloader does it's thing with. I saw one spare SDCC card come without this chip, I believe it's identified as the U9 chip on the cards. If that chip is not present then no card can get any configuration information. I don't recall if the absence of the chip resulted in an A8 indication, but it's worth checking at this point. We just removed the chip from the card being replaced and placed in on the new card and everything worked just fine.

If the problem is just that the StageLink- and/or Voter ID wasn't set properly, you will likely need to download ALL to the new EEPROM chip on the SDCC card, then re-boot the processor (a hard re-boot, removing power for about 30 seconds, then re-applying power) in order to get the EEPROM formatted and new information downloaded to the EEPROM on the new card.

If all the above fails, then I would start by replacing the TCDA card in Loc. 1 of <CD>, then retrying. If that doesn't work, then I'm out of ideas at that point.

Again, please write back to let us know how the troubleshooting progresses and the resolution!

 

Identified the issue with SDCC PROMs, replaced the same with a new set and now the <C> communicator is ok.

Thank you for the support and suggestions.

 

Thanks very much for the feedback! Can you tell us precisely which PROMs were bad? Not the PROM numbers, but the DS200 numbers on the labels.

 

We are facing a problem in one of our Mark V systems (Frame V GT). The <C> computer is showing A8 error and not booting. Also we are not able to access 186 monitor from the LCC display.

We have replaced all the PROMs on SDCC card, ie., U11, U12, U22 & U23 (on a new SDCC card). We have not checked by replacing one by one, as we were afraid that probably the faulty PROM may damage the new card.

 

Interesting and learned.

Thank you CSA for your valuable command for help them to solve their issue.

Is there anyway to check the faulty PROM's?

Thank you
G.Rajesh

 

To check the PROMs, one would have to have a "PROM burner" and have a made a copy of the original PROM to compare against the suspect version. The data is stored in a binary format and without knowing what it was originally, it would be very difficult to tell if all the bits were in the right "places". One could check for the presence of bits or for an obvious read error, but that wouldn't say anything about whether the bits there matched what they were or should be without something to compare it against.

I have heard about a disturbing trend of people paying companies to come in an make ("burn") copies of running PROMs (of course the machine has to be shut down to remove the PROMs in order to make copies), and then the PROMs have to be re-installed and the unit re-started.

More than a couple of sites have had problems re-starting because something happened to the original PROMs in the process of removing them, copying them, and re-installing them. It's very difficult to say exactly what happened, but it was also found that the copies that were made were also corrupt. So, it was necessary to go to GE and purchase new PROMs (which was what was trying to be avoided in the first place) and that can take weeks (unfortunately).

The case where a PROM (EPROM to be exact) fails is rare in my experience, and almost always happens because of 125 VDC power supply faults (spikes, dips, lightning strikes, battery charger issues, etc.) or serious 125 VDC battery grounds which go unresolved for long periods of time. Or when cards and PROMs are mis-handled curing card replacement exercises during troubleshooting exercises.

While it may be expensive, buying new PROMs from GE is probably more cost-effective in the long run if they are purchased in anticipation of a possible need, rather than purchased under emergency conditions when one has failed for whatever reason. The delays that can occur when trying to purchase these from GE can run into tens, maybe even hundreds, of thousands of dollars whilst trying to work through the back-and-forth and finding the right GE entity to supply the PROMs, and then getting them burnt and shipped to site.

Much better to order them as "spares" and work through those issues when it's not an emergency than to wait until they are needed critically.

An ounce of prevention is worth a pound of cure.

 

Thank you CSA for your advice / information.

keep it up
G.Rajesh