Getting a better understanding of gas turbine control

Question Regarding Inversion Masking

Dear CSA,

I've been following this thread (I read this whenever I get time) and I appreciate the time and energy you devote to the community. Please continue this thread. As your last two posts are not appearing when I open this thread, I'll be good if this discussion is continued in other thread (part II as Neo said).<pre>
L86CB L4Y l27qel L3CP l63fdl
---|/|-------||--------|/|-------||-----|/|--------( ) L3STCK2
| |
| L43FTG |

L3STCK2: Start check 2
L86CB : Compressor bleed valve or IGV position lockout
L4Y : Loss of master protective
l27qel : Emergency lube oil pump under-voltage
L3CP : Customer permissive to start
l63fdl : Distillate fuel supply pressure switch
L43FTG : Transfer to Gas

As already said I'm also on this journey with you and Neo. I've learned to read the Rungs and interpret how it'll work. I've managed to get some of the Application Code running in our Speedtronic Mark VI, and I've a question regarding inversion masking.

let me read the Rung:

when L86CB is logic "0" AND L4Y is logic "1" AND l27qel is logic "0" AND L3CP is logic "1" AND l63fdl is logic "0" OR L43FTG is logic "1" L3STCK2 will be logic "1".

now interpreting it:

When there is NO compressor bleed valve or IGV position lockout AND loss of master protective timer is complete AND Emergency lube oil pump is NOT undervoltage AND Customer permissive to start is selected AND Distillate fuel supply pressure is NOT low OR transfer to gas is selected, start check 2 is completed.

Coming to my question, I think the signal l27qel and l63fdl are inverted.

For l27qel: When EOP is not under-voltage, the NC contacts will remain closed and thus it will allow L3STCK2 to pickup. When EOP is under-voltage the NC contacts will open and will prevent L3STCK2 to pickup. Thus when the contacts are closed then only L3STCK2 coil is allowed to pick up. This ensure when the contacts are closed EOP is not under-voltage and the switch is also in healthy condition, thus doing the work of supervision too. Instead, had GE had chose the contacts to close when EOP is under-voltage, we wouldn't have known even if the circuit is intact or not.

For l63fdl, When the distillate supply pressure is not low the NC contacts will remain closed thus alowing L3STCK2 to pickup. When the distillate fuel supply pressure is low, the NC contacts will open, thus preventing L3STCK2 from picking up. Thus here also the contacts must be close, which ensures that the distillate supply pressure is normal and the switch circuit is intact.

Am I right? I've tried hard to understand the concept of inversion masking by reading (and re-reading) your explanations in this thread. Please correct me if I'm wrong.

As you asked about the machine at our site in It is a PG5361. It is having conventional combustors. The machine is duel fuel capable. It can be operated on Natural Gas or Naptha, but it is operated on NG only since last 10 years. out of two machines, one was installed in 1988 and the other in 1996. Both the machines can be operated in Droop and Isochronous mode. Our plant generally (not all the time though) remains separated from the grid. Both the machines exhausts in HRSG. This arrangement is having a By-Pass stack.
Thank you CSA. your explanation made me understand L4Y better, will start focusing on rest of the control signals.

Inversion masking. Okay. I have to make this short as I've just been given a last-minute opportunity for a holiday in the South Pacific Ocean that one dreams about, and I'm going to take advantage of it. So, I'm leaving tomorrow and have just a few minutes. I'll check this thread on my return in about a week-and-a-half.

HOPEFULLY the contacts of the 27QE relay are NORMALLY OPEN, and close when there is voltage to the starter and open when there is no voltage to the starter. (When there IS voltage to the starter, 27QE should energize, and that will close a NO contact connected to the l27qel input of the Speedtronic. Now, if the input were NOT inverted, that would mean that l27qel would be a logic "1", but that's in 'violation' of proper signal naming conventions because l27qel is supposed to be a logic "1" when the Emergency L.O. Pump voltage is <b>L</b>ow (the L (or l) at the end of the signal name means the signal name is a logic "1" when the voltage is <b>L</b>ow). So, the signal is inverted by software on the discrete input card BEFORE it drives the contacts in the associated with the input signal in the sequencing (or application code).

Let me try to draw it here by attempting to show the physical contact from the undervoltage relay of the Emer. L.O. Pump motor starter connected to the Speedtronic's input screw for the Emer. L.O. Pump Motor Undervoltage input, and show the software coil driven by the contact input, because that's what EVERY discrete (contact) input does: it drives a coil in software, and contacts associated with that software coil are used in the sequencing (application code). In the case of inverted inputs, the coil status is the opposite (the inverse) of what one would normally expect it to be when the contacts are closed or open.<pre>
Physical | Software
Contact |
27QE Input Screw l27qel
---| |----------o------------------(I)---</pre>
The "I" in the software coil for l27qel means the input signal is inverted, which means that when the normally open contact of 27QE that's connected to the Speedtronic input screw is <b>closed</b> l27qel <b>WILL NOT</b> be a logic "1"; instead, it <b>WILL be</b> a logic "0"--which is the inverse (opposite) of what one would think it should be when the contact was closed.

If the normally open contact of 27QE were connected to real physical electromechanical relay and it was closed (when there WAS voltage to the Emer. L.O. Pump motor starter) the electromechanical relay coil would be ENERGIZED, it would be "1". Without some additional electronic devices there can be no "inversion" of a hard-wired contact driving an electromechanical relay coil. But, using software a closed input will NOT cause the software coil it drives to be a logic "1"--it will cause it to be a logic "0" (the inverse of what one would expect). And when the contacts are open, an inverted software coil will NOT be a logic "0", but it will be a logic "1" (the inverse of what one would expect).

Now, after the inversion, in software normally open and normally closed work exactly as one would expect when the associated coil driving them is a logic "1" or a logic "0". When l27qel is a logic "0", normally open contacts associated with l27qel will be open, and when l27qel is a logic "1" then normally open contacts will be closed. Normally closed contacts of l27qel will be closed when l27qel is a logic "0", and normally closed contacts of l27qel will be open when l27qel is a logic "1".

So, when the contact of 27QE (in this case, it should be a normally open contact, that is open where there is an undervoltage, and closed when there is NO undervoltage) is CLOSED, then l27qel will be a logic "0" (NOT a logic "1" as would be expected) and normally closed contacts of l27qel would be CLOSED--because the Emer. L.O. Pump motor voltage is NOT low.

Just apply the same logic (thinking) to l63fdl. A normally open contact of the pressure switch will be open when the liquid fuel supply pressure is LOW, and closed with the liquid fuel supply pressure is NOT low (when the supply pressure is normal). l63fdl will be a logic "1" when the liquid fuel supply pressure is <b>L</b>ow, and because that input is inverted that means that the contact will be OPEN when the supply pressure is low--which will be the case when there is low pressure (or no pressure) and the input is wired to a normally open contact of the pressure switch. The normally closed contact in the rung will be closed when the liquid fuel supply pressure is NOT low (when l63fdl is a logic "0", which will be when the contact of 63FD is closed, and a normally open contact of 63FD will be closed when the pressure is above the setpoint of the switch).

Finally, think about it this way. If a normally closed contact of 63FD were wired to the Speedtronic, then when there WAS pressure to the switch the normally closed contact would be open. If that contact was wired to a non-inverted input called l63fdl then l63fdl would be a logic "0" when there was pressure (which is all well and good!).

However, if a wire falls off one of the terminals of that normally closed contact of 63FD <b>AND</b> at some point after that the liquid fuel supply pressure drops below the switch setpoint and the normally closed contact closes the Speedtronic will NEVER know that the liquid fuel supply pressure dropped. So, GE uses contacts that are to be CLOSED when the condition is NORMAL, and OPEN to alarm or trip (when the condition is NOT normal). Better to get a false alarm when a wire comes loose or a switch fails but the condition is normal, than not to get an alarm when there is a real problem.

WHEW!!! I gotta finish packing, kick the wife and kiss the dogs--er, ..., uh, ..., I mean kiss the wife and kick the dogs. Er, ..., uh, ..., I gotta go! Please read this, and re-read it and re-read it. It's a LOT to digest in even three readings, but at some point it should start to make sense.

First of all happy journey!
Hope you'll enjoy a lot and come back even with more energy!

Thanks for taking the time to respond. Your replies are valuable to me (and many more like me).

It has started to make sense as you said. I'm gonna read it many more times till you come back.

I am following this interesting discussion few days ago. I would like to join it and ask you about GE philosophy in arranging start check parameters into groups and go deeper asking if each parameter related to the others in each group or just random arrangement.

Good question, and welcome to the discussion. In the beginning, it was intended that the groups be of like conditions, but over the years that has ceased to be the case. And, in some cases, some conditions would not fit into a group with any other condition. Which would mean a lot more "auxiliary" start-check rungs. Really, a start-check is a start-check is a start-check, and the only reason they are broken up in digital control systems is, again, because of the limitation of the sequencing editors to no more than eight elements (seven contact in a horizontal series string, plus the rung coil--for a total of eight elements). In the days of hard-wired electromechanical relays there was no limit to the number of contacts in a series string (which made troubleshooting rather difficult at times).

So, while the GE-design heavy duty gas turbine control philosophy was originally intended to group start-check conditions, that has stopped being the case in the last 20 years or so, particularly as more and more start-check permissives have been added to the unit as they become more and more sophisticated.

Hope this helps!
Dear CSA,

It seems that further discussion on L4T rung is not continued in this thread. I would like to know if the discussion was continued on any other thread. Also, I am very new to gas turbine control. I am hoping to learn a lot from you.

Thanks in advance!!!

If you click on a Member's name it will show all of the posts which the Member has made. I clicked on 'Neo' and I found all of the posts which he has made. And, there are several which you might benefit from.

Can you tell us a little about your position, the equipment you work with, and your work experience?
Dear CSA,

Presently, I am working with 3 nos. GE frame 5 gas turbine, co-gen cycle, Naptha fuel while running & HSD for start up. I am an electrical engineer with 6 month of experience in this field. I am working as a Power Plant operation engineer. Right now, I have gone through the GT support system manual and have read Part-I of this topic & there is still so much to grasp. Let me complete all the parts of this topic and I will let you know if I will come across any problem in understanding the same.

Thank You!!
Hello CSA, would like to know some details:

We have Mark-VIe control system for running frame 3 gas turbines. Recently a unit tripped on "Excessive combustion trouble trip" alarm. Prior to this alarm, "Exhaust thermocouples trouble" alarm and "combustion trouble alarms" also appeared and persisted. We have 18 thermocouples whose readings just prior to tripping were as follows:
547/542/551/567/563/558/553/557/564/504/505/501/491/493/516/528/527/526 degC respectively in ascending order sequence i.e. TE-128.1~18. Average exhaust temperature was 534 degC. Combustion monitor actual spread was more than allowable i.e values were 73/66.7 degC respectively. My exact questions are:

1) What exact logic combination has caused tripping of unit, considering combustion monitor actual spread was already more than allowable, i.e. 73/67 degC respectively.

2) What is difference between turbine exhaust primary temp control reference, turbine exhaust temp control reference and speed biased temp control reference?

3) Value of allowable spread increased from 67 to 149 degC after unit outage; how is this explained. What calculation is involved in allowable spread?
Hello Neo,

Can you help me understand this rung. Unit is GE MS6000b with Mark IV speedtronics. What does it mean when the point is Highlighted and has a negative sign underneath?
Thanks for you time,

Hello Neo,

Can you help me understand this rung. Unit is GE MS6000b with Mark IV speedtronics. What does it mean when the point is Highlighted and has a negative sign underneath?
View attachment 1024
Thanks for you time,

Hello Neo,

Can you help me understand this rung. Unit is GE MS6000b with Mark IV speedtronics. What does it mean when the point is Highlighted and has a negative sign underneath?
View attachment 1024
Thanks for you time,

Hello All

It is called a CONTACT variation :
From Mark IV&V user Manual it is saying that One logic signal can be used to multiple contacts.

Have a search on the Right Technical manual From OEM it is always a good thing to do !

Thanks for sharing such Post

We are here to share and learn!

Always glad to support you!


I don't have access to the Mark IV Maintenance Manual at this writing, and I don't immediately recall ever seeing the ”negative” sign under a contact--a highlighted or un-highlighted contact.

When a contact is highlighted, it means that ”power” can pass through the contact. The highlighted contact in your post is normally closed. For power to be able to pass through normally open contacts the logic signal associated with the contact (L3COMF in this particular rung) has to be a logic ”0”.

This is proved by the other occurrence of L3COMF in the rung--it is a normally open contact and it is NOT highlighted. A normally open contact that is NOT highlighted means the logic signal associated with the contact is a logic ”0”.

The question, though, was help with reading the rung not with a single rung. Do you require further help?

Finally, the ”negative sign” underneath the contact is most likely the result of the location of a cursor on the display, or some similar artifact. If I can get access to a Mark IV Maintenance Manual soon, I'll try to look up the ”negative sign” beneath a contact on the Rung Display. I cut my teeth on Mark IV, but I don't recall that symbol. But it has been a very, Very, VERY long time since I've worked on Mark IV....

Hope this helps. If you require further assistance with the rung (and not just a single contact), write back. (But, it's a mystery why you didn't ask about the ”negative sign” beneath the un-highlighted contact in the rung....)
The answers are clearly given in the Rung display modes of the Markiv CRT in the OEM manual!

It is about Fn functions & Cursor selections used on CRT & Opeartor interface module to navigate and see which current signal status or dispaly the rung sequentially as per your rung !

I am sure that you can find what you are looking for on the manual just read it carefully!

Can the original poster try to respond to the last thread ...
It would be much appreciated ! as we are not here to discuss in the wind !

I mean we make effort to reply so please try to do the same...gentlemen.

Thank you for your attention& comprehension