Dear CSA,

My company using GE GTG Frame 9 with Speedtronic Mark V system control..
Could you explain to me about Exhaust Temperature Feedback and Protection (TTXMV4) BBL?

i have read the Mark V manual from GE and do not quite understand so i think you can help me with that because i have seen lot of post from many expert in this forum and i admire with CSA answer.

thank you

 

Have you looked at this thread:

http://control.com/thread/1260993438

Do you still have questions?

 

thank you for your link,

if our unit has 24 exhaust thermocouple,how many T/C signal for each processor (<R>, <S>, <T>)?

and what is the signal TTX_TOT_TC and TTX_NUM_TC means?

And can you explain to me about over temperature and exhaust over temperature trip algorithm? I think it is on the bottom of the TTXMV4 BBL

thank you

 

pklman,

The Mark V operator interface has a "list" of all of the inputs and outputs of the control system, and where they are connected and what their signal names are. It's in the F:\UNIT1 directory and it's called TC2KREPT.TXT. It's an ASCII text file which can be viewed with any text editor (MS-Windows Notepad or -Wordpad; MS-DOS EDIT; even MS-Word will open ASCII text files).

The TBQA card on <R> core has 45 T/C inputs total on it. Those 45 T/C inputs are split evenly between <R>, <S>, and <T> in a TMR Mark V turbine control panel. The inputs on screws 1-30 are connected to <R> processor; the inputs on screws 31-60 are connected to <S> processor; and the inputs on screws 61-90 are connected to <T> processor. (Yes; the terminal board is mounted on <R> processor, but there are ribbon cables (those lovely ribbon cables!) from the terminal board to cards in <R> core, <S> core, and <T> core to get the signals to each of the three redundant control processors.)

The idea is to evenly divide the exhaust T/Cs between the three redundant control processors, so eight are connected to <R>, eight to <S>, and eight to <T>. Starting at the 12 o'clock position and continuing in the direction of rotation of the turbine shaft, the first exhaust T/C is connected to <R>, the second to <S>, the third to <T>, the fourth is connected to <R>, the fifth to <S>, the sixth to <T>, the seventh to <R>, the eighth to <S>, the ninth to <T>, and so on. So, in your case the exhaust is effectively divided into eight sections, with each section being monitored by exhaust T/Cs which are connected to <R>, <S> and <T>.

I believe the values you are asking about are Control Constants, and you can look them up in the Control Constant display or by looking at the block in Dynamic Rung Display. In your case, TTX_TOT_TC (the total number of exhaust T/Cs) should be 24. And I believe TTX_NUM_TC is the number of exhaust T/Cs connected to each control processor (I don't have access to a running Speedtronic panel any longer to verify this, but you do!).

In general, exhaust over temperature protection is done by looking at the "median" (average, actually) exhaust temperature, TTXM, as calculated from the exhaust T/C inputs to the block, and comparing it to a couple of setpoints. The setpoints are a function of the exhaust temperature reference/limit which is being calculated by the Speedtronic (TTRX) plus a bias. If TTXM exceeds (TTRX + 25 deg F) then an exhaust over temperature alarm will be annunciated, and if exceeds (TTRX + 50 deg F) then the unit will be tripped.

There is some "massaging" of the calculated "median" exhaust temperature that goes on in the block, to remove a limited number of exhaust T/C values which are deemed excessively high or excessively low, so as not to skew the "median". For example, if one of the exhaust T/Cs is only reading 178 deg C, and the others are reading approximately 500 deg C then that one signal could falsely skew the reading, so it is removed from the calculation. Years of empirical data have proven this to be acceptable to protect the turbine from sensor/wiring problems.

The exact method of massaging the median calculated exhaust temperature has changed over the years, but the concept still remains the same: to remove <b>some</b> values which might falsely skew the readings in order to improve the reliability and availability of the turbine to produce power, but not all of them. (Some people want to eliminate four or five or even eight exhaust T/Cs from the calculation because they don't want to shut the turbine down to replace faulty T/Cs or check the wiring/terminations. That would be defeating the combustion monitoring function (exhaust temperature spreads) so only a very limited number of readings which might falsely skew the calculated "median" exhaust temperature are removed by the algorithm/block. And it's not user configurable or modifiable (thankfully!).)

 

dear CSA,
thank you very much for your help..

regards
pklman

 

Dear CSA,
It's such a long time since we met here, right? i have been switched over to some other field and now i have come back to resume my control philosophy of GT. Here i have got a question:

Why do you wanna compare TTXM with TTRX or TTRXB to find Exhaust OT trip? Why don't we compare our controlled exhaust temperaure TTXM with higher constants value? Because our TTRX and TTXM almost going to be same (1-3 degC Difference),right? Is there any valuable reason behind this one? Please elaborate.

 

To detect an exhaust overtemperature alarm condition what's done is to add 25 deg F to the exhaust temperature reference (TTRX) and if the actual exhaust temperature (TTXM) exceeds (TTRX + 25 deg F) then an exhaust overtemperature alarm is annunciated.

To detect an exhaust overtemperature trip condition what's done is to add 40 deg F to the exhaust temperature reference (TTRX) and it the actual exhaust temperature (TTXM) exceeds (TTRX + 40 deg F) then an exhaust overtemperature trip is initiated.

So, the exhaust overtemperature alarm and -trip are actually "sliding" conditions, changing with TTRX.

If there was just a single exhaust overtemperature alarm setpoint and a single exhaust overtemperature trip setpoint then when the unit was operating at Base Load on cooler days with higher CPD the firing temperature would increase too much for the hot gas path parts before the alarm or trip were detected. So, the alarm and trip setpoints vary as a function of the exhaust temperature reference so as not to overheat the hot gas path parts.

There's a third condition where if the actual exhaust temperature exceeds an upper limit (usually 40 deg F greater than the maximum isothermal limit) then an exhaust overtemperature trip will be initiated.

The three Control Constants which define these setpoints are TTKOT1, TTKOT2, and TTKOT3. You can look them up in your CSP to see which one is which. And also look at their usage in the exhaust temperature monitor function.

When the turbine is on exhaust temperature control, it's putting as much fuel as it can into the unit while trying to keep TTXM as close to TTRX as possible. Variations in exhaust flow and T/C outputs cause the slight differences that you note.

 

Have you looked at this thread:

http://control.com/thread/1260993438

Do you still have questions?

HI CSA. The link you shared is no longer existing. Could you please send me another link if the thread still exist. Thanks

 

CSA doesn't seem to be posting to Control.com much these days; it seems he feels he's fulfilled his self-imposed obligation of trying to help others understand and troubleshoot GE turbine controls.

This is an interesting message; I've never seen this message before on Control.com.

The Moderators will have to look into this to see why the thread is not available/accessible.

Hopefully the Moderators can look into the situation and see what can be done. The ownership and day-to-day operation of Control.com has changed since this thread was started 13 years ago. It may have something to do with links or archiving technique, but we should have an answer soon--we hope!