We have frame5 gas turbine machine with mark 5 control system. We are using windowsNT for HMI. now machine is at shutdown condition for combustion chamber inspection so that we removed both the spark plugs. now the question is we want to check both the spark plugs operation. for that what are the signals i have to force?

i checked in the rung display it is showing L2TVX- (ignition permissive) as Input and L2TVX1- (ignition transformer), L2TVX1A- (ignition transformer Aux relay K1), & L2TVX1A_REF (description not mentioned) are Output, ie, 3 outputs. Can anybody explain about this rung? Description? Outputs? and Logic signals required to force, to test the spark plug externally in a safe manner?

 

SHYAM,

The Mark V I/O Report will tell you precisely which software logic signal name drives the spark plugs. The file, found in F:\UNIT1, is named TC2KREPT.TXT. It is an ASCII text file and can be opened with MS-Notepad or MS-Wordpad, and you can search the file for "L2TVX" to find all occurrences to find the corresponding relay output which is assigned to switch the AC to the spark plug transformers/spark plugs.

A word of caution about TC2KREPT.TXT: It is not automatically generated from any other file(s) on the GE Mark V HMI; it is created at the time the original Mark V software for your site was created. After it is created, it must be MANUALLY edited by anyone making changes to the Mark V I/O for the turbine it was created for. QUITE FREQUENTLY if changes were made during commissioning or afterwards this file DOES NOT get edited to reflect the changes. So, while the file is good for MOST things, it <b>DOES NOT</b> reflect the actual way every input to- and output from the Mark V is actually wired to the Mark V at your site.

Having said the above, the spark plugs are intended to be wired to a specific discrete (relay) output of the Mark V panel, so for this particular output (spark plug transformers and spark plugs) it should be fine. (Unless it was improperly assigned to a different output during or after commissioning.)

The <b>ONLY</b> Mark V configuration file which defines exactly which inputs and outputs are assigned (hardware to software for outputs; software to hardware for inputs) is IO.ASG (also found in F:\UNIT1. This is also an ASCII text file, and can be opened with MS-Notepad or MS-Wordpad. <b>The caution with this file is DO NOT MAKE ANY CHANGES TO THIS FILE AND MAKE SURE WHEN YOU EXIT THE FILE YOU DO NOT CHANGE THE FILE NAME </b>(IO)<b> OR FILENAME EXTENSION </b>(.ASG)<b>.</b>

So when using any program to open and view the contents of IO.ASG, make sure that when you close the file and exit the program that you <b>DO NOT</b> save any changes if asked by the program. (A really good way to avoid making any changes to IO.ASG--or any ASCII text file associated with the Mark V--is to make a copy of the file and give the copy a unique name and then only open the copy when needing to refer to information in the file.)

I'm going to make an educated guess and say the signal you want to force to a logic "1" to energize the spark plug transformers, and hence the spark plugs, is L2TVX1. I believe this is the signal name you will find in both the I/O Report and IO.ASG files. I don't have access to any Mark V configuration files as I write this, but I believe the relay outputs the spark plug transformers/spark plugs are intended to be wired to are <QD1> CO16 or <QD1> CO17 (the "CO" stands for 'Contact Output). So, when you find the signal name (again, probably L2TVX1) in the I/O Report or IO.ASG you should find it is assigned to either Q_QD1_CO16 or Q_QD1_CO17. (If the unit has DLN-I combustors, you may find a second reference/output; this is for DLN primary combustion zone re-ignition during a normal fired shutdown. It is a second set of contacts from another relay output (contact output) in the Mark V wired in parallel with the "primary" relay output used during starting. I believe this is the L2TVX1A output, but, again, that's just a guess without being able to see the CSP at your site to be certain.)

The spark plug bodies must be firmly grounded to non-painted metal. (They are bolted to the combustion cans using metal bolts which serve as the ground during normal operation; usually just laying them on the galvanized metal grating outside the turbine compartment will suffice.) Personnel working in the area where the spark plugs are laid in the open should be notified and stay a couple of meters away from the spark plugs when they are going to be energized.

Many times the safety procedures for an outage require that the AC source for the spark plugs be isolated (by opening an circuit breaker or lifting a wire somewhere in the low voltage circuit (120 VAC or 220 VAC, depending on locale/configuration). I've been to several sites where the maintenance technicians forgot that the AC power to the plugs had been isolated during preparations for the outage and were claiming the Mark V wasn't working correctly when they forced the logic signal to energize the spark plugs because they weren't sparking.... The Mark V just switches an AC source which is external to the Mark V to energize the spark plug transformers. The Mark V does NOT provide the AC sources for the spark plug transformers; it simply serves to switch the AC source on and off (by closing or opening relay contacts). The AC source must not be isolated during this test.

Hope this helps--wish I could be more specific about the exact logic signal which drives the relay output at your site, but it's always best to refer to the documentation for your site (including configuration files).

 

Shyam

Just check on your wiring list (or interconnecting diagram) where is your ignition transformer connected it will give you the name of the signal to force, but usually forcing L2TVX will do the job all other signals are some "spare" one (at the time there were two ignition transformers such as with Frame3 GT).

I am just a little worry about you last signal L2TVX1A_REF, is your card the VPRO card connecting the ignition transformer? and in that case it would be useful to also force the signal L4_XTP (also sometime named L4NEG) but usually ignition transformer are connected to TRLY card.

 

pascal,

Mark V's don't have TRLY cards; Mark VI's and Mark VIe's do.

The "standard" for discrete outputs for all digital Speedtronic turbine control panels (if adhered to by designers and field people) is to use a separate logic signal name to "drive" an output and NOT to use contacts associated with that logic signal anywhere in logic. That way, when an output is forced the force does only one thing: acts on the output. And does NOT affect anything else in logic/sequencing/control.

Sometimes, but not always, the separate logic signal is driven, in a separate rung, by the "main" logic signal, but frequently--especially in Mark V--the output logic signal coil is in configured to be in parallel with the logic that drives the "main" logic signal (L2TVX in this case). So, just forcing L2TVX will likely NOT force L2TVX1.

And, if the original poster looks in the CSP no contacts associated with L2TVX1 (and possible L2TVX1A) should be found anywhere else in the CSP--because they are only driving outputs.

I, too, was surprised to see the L2TVX1A_REF signal, but I wouldn't be surprised to find the signal in a Mark V that's recently been "upgraded" to a Mark Ve or whatever they're calling it these days, or as some "artifact" of some unused logic signal that somehow was mistakenly left in the CSP for that site. But, without being able to see the CSP and know more about the "Mark V", it's impossible to say for sure.

In the Mark V, trip solenoid outputs driven by the <P> core frequently had _REF signals associated with them, but the spark plug transformers were not usually driven by <P>....

 

Thanks for your valuable reply. for a couple of days i am not in the site so i could not check the TC2KREPT.TXT and IO.ASG files. but as per wiring drawings it is showing there is 2 relays named-L2TVX1 and L2TVX1A. the control signals (relay drive) for L2TVX1 connected to the,<QD1> DTBD - 61&62 and the control signals (relay drive) for L2TVX1A connected to the,<P> PTBA- 65&66.

230V AC taken from the external source and going to the ignition transformer through these relays. ie, phase is going through one relay and neutral is going through another relay. Still i am confusing about the last signal "L2TVXIA_REF". Expecting your feed back..

In my view i think i will have to force both L2TVX1 & L2TVXIA signals to make spark plug activation. can you give feedback?

 

shyam,

Okay, so part of the mystery is solved. L2TVX1A is assigned to an output of <P> core, which explains the L2TVX1A_REF (the Mark V uses the _REF signal as a check to see if the output has, in fact, been energized--or de-energized--when commanded).

I don't have access to a Mark V Application Manual, GEH-6195, at this writing to know exactly which output is being used from the <P> core. My concern stems from the fact that AC is necessary for the spark plug transformers, and many of the <P> core outputs are hard-wired/configured to pass DC (for trip solenoids). So, I'm a little confused about how the panel is configured.

I would suggest actually tracing wires in the panel to be certain one wire is actually physically connected to the <P> core; as was said before, it's extremely common to find documentation that was never updated during installation/commissioning to reflect wiring changes made to the Mark V Panel.

If, in fact, the spark plug transformers are connected to the <P> core the task of energizing them is going to be quite a bit more complicated. Presuming one leg of the spark plug transformer supply IS connected to the <P> core you will need to ensure that the emergency trip push-button circuit connected to the <P> core is complete (no P/Bs are latched in the trip position; many LOTO procedures require at least one E-Stop P/B to be latched and "locked" in the tripped position.

Next, you will need to force the following signals to the following logic states:<pre>
L2TVX1 = "1"
L2TVX1A = "1"
L4_XTP = "0"</pre>
If you don't force L2TVX1A_REF the output will work, BUT you will get a Diagnostic Alarm output feedback not agreeing with reference or something to that effect. You could also for L2TVX1A_REF to a logic "1" to prevent that Diagnostic Alarm, but it's an unnecessary step.

Without being able to look at a Signal Flow Diagram (in GEH-6195) I can't understand how an AC signal can be passed through a DC circuit (I'm assuming the <P> core has a TCTG card, which most GE-design heavy duty gas turbine Mark V control panels use). So, I'm a little leery of whether or not the documentation actually reflects the current wiring/configuration of the Mark V--but stranger things have happened!

I would suggest, again, that someone take the time to physically trace the spark plug transformer wiring in the Mark V panel, draw a schematic of the circuit, and use this in the future when wanting to perform this test again. It will likely save a lot of effort--and frustration. This could be part of a procedure that outlined what needed to be forced to what logic state--again, this could be very helpful in the future.

Please write back to let us know what you find and how the test worked (what you had to force; etc.).

 

We successfully tested both the spark plugs simultaneously by forcing L2TVX1, L2TVX1A and L4_XTP. Thanks for your support and feed back.

 

shyam,

You are most welcome!

Thanks for the feedback, also. 'Feedback is the most important contribution!'(c) here at control.com. It is what distinguishes control.com from a lot of other Web-based forums. When posters who receive assistance from responders on control.com provide feedback it lets other people who read these posts--now, and in the future--if the information provided was helpful, or not (as the case may be).

I want to be very clear about this particular post, though. It is NOT typical for either leg of the spark plug power source to be connected through the <P> core of a GE Speedtronic Mark V turbine control panel, so, it's not generally required to force L4_XTP in order to be able to energize the spark plug transformers/spark plugs.

It's unclear (to me, anyway) why any leg of the spark plug power source would be assigned to the <P> core, unless it was something demanded during commissioning or by some local technical code or regulation. Even so, it's not clear how an AC source can be routed through the <P> core trip solenoid outputs. (I do NOT have access to Mark V Signal Flow Diagrams at this writing, so I'm not clear what function PTBA-65 & -66 are connected to in the <P> core.)

Again, congratulations on the successful--and safe--test. And, thanks for the feedback!

 

It's new subject for us too, spark plug is connected in <P> core.

But questioner Mr.Shyam can make sure without L4_XTP=0 spark plug is not energized.

Take care
G.Rajesh

 

sorry i was not in the site at the time of testing. i just given instruction to my colleague for testing the spark plug, so i couldn't check your point. this both relays are mounted separately.

anyway i will try to confirm your point and come back to you.

 

Sorry for the late reply. We are using 2 relays to give the supply to ignition transformer, one relay drive signal is connected to <QDI> DTBD & another relay connected to PTBA. So without forcing this L4_XTP.

The relay which is connected to <p> will not energize and cannot perform spark plug test..ie, one relay (NO)contact for phase and another relay (NO) contact for neutral of 230v ac as the input of ignition transformer. Once again sorry for the late reply.

 

shyam, or SHYAM,

The configuration you are describing is not typical, so without being able to see the schematic drawings for the circuit it's very difficult to say for sure what might be the problem.

And I'm not clear if there is second, physical relay that's connected to <P>, or if it's a second relay output of <QD1> (or even <QD2> if there is a <QD2>).

I'm going to venture a guess and say that the second relay (the one connected to <P>) is actually being used to interrupt the AC neutral to the ignition transformers--but that's just a guess; it may, in fact, be in series with the relay output of <QD1>. Again, without being able to see the electrical schematic drawing of the circuit it's very difficult to say with any degree of certainty. You could try measuring the AC voltage across the contact(s) of the second relay (the one that's connected to <P>). You could also use the multimeter to see if one side of the second relay's contact(s) is connected to neutral (earth; ground).

My other scientific wild-arsed guess is that there is a second relay output of the Mark V that is being used to apply power to the coil of the relay to energize the second relay's coil to close the contacts. This should be very evident in the CSP because it would have to be a logic "1" at the same time as the L2TVX1 relay (that's the typical signal name; look at the I/O Report to find the actual name used at your site). So, it should be possible to find that second relay's name, and then use the I/O Report and/or IO.ASG To find which relay output the coil of the second relay the contacts of which are connected to <P> is located. But, it's a pretty safe bet that a second relay output from the Mark V has to be used to drive the relay that's connected to <P>--and this is presuming that there is a second physical relay (an "interposing" relay) connected to <P>.

Have you traced the wires in the ignitor circuit from <P> to see where they go? That will tell you a lot about how the relay needs to be energized.

Please write back to let us know how you fare.

 

CSA,

I have a question on ignition excitation logic, but before asking that I will post Mark vie version of the control sequence. L2TVX is "ignition permissive" & l2tvx1 is "ignition permissive relay", which is energized when the output of the block L2TVX is "1". Apart from supplying the AC voltage by external means to ignition exciters 95SG-2A,3A which in turn make the spark plugs 95SP-2 & 95SP-3 "on", l2tvx1 is also used in a block, which is used for L30SG1_ALM i.e. "EXCITER CHANNEL FAILED WITH IGNITOR ON alarm" with a time delay of K30SG1_ALM (1sec). I guess L30SG1_ALM is like L2TVX1A_REF in Mark- V, which is a check to see if the output is energized or de-energized when commanded.
<pre> l30sg1 l2tvx1 L30SG1_ALM
|/|--------| |-----------( ) </pre>
Here comes the confusion. In the device summary, the description of 30SG-1 is
"DIAGNOSTIC IGNITION EXCITER SWITCH
NORM=NO; CLOSE ON EXCITER FAILURE;".

If we go by the logic block l30sg1 is an "NC" (NOT Gate used), but the device summary says it's an "NO" contact, and it closes on exciter failure. It seems to be contradictory & creating confusion.

Note: The info posted above is collected from the O&M manual & device summary, not from the ToolboxST.

 

Batman,

This is an all-too-common alarm and problem, and may involve ... <b><i>inversion masking.</i></b>

The high-energy ignitor power source, 95SG (there's usually only one of these ignitor power supplies, which drives two (2) ignitors ("spark plugs")) has a discrete (contact) output which changes state under some condition. I don't recall if it changes state when <b>both</b> ignitor outputs <b>ARE</b> working, or when one of the ignitor outputs <b>is NOT</b> working. Let's say it changes state when BOTH ignitor outputs <b>ARE</b> working, first.

l2tvx1 is the discrete output of the Mark VIe that, when a logic "1", will apply single-phase AC to 95SG, energizing 95SG. So, when l2tvx1 is a logic "1" the input to 95SG should be energized, and the outputs of 95SG should also be energized.

(There may be two 95SGs, and the discrete outputs may be connected in series or in parallel.... Things do change over the years.)

For now, lets' say there's only on 95SG with two outputs, and in this case, when the ignitor power source is energized and BOTH ignitor outputs are working if the NO contact changes state and closes that would be an indication that both ignitor outputs are working (it doesn't say both ignitors are firing--just that the outputs to both ignitors are energized). In this case, if the input was NOT inverted and the contact closed then l30sg1 would go to a logic "1" and the normally open contacts of l30sg1 in the rung you provided would OPEN and there would be NO alarm--which would be correct, right? Both ignitor power source outputs are energized when the input to 95SG is energized, and when the application code is telling the ignitors to be energized that's what is desired.

Now, let's continue to say that the discrete output of 95SG changes state when BOTH outputs are energized--<b>however,</b> one of them does not get energized when the input to 95SG is energized. In this case, the discrete output would NOT change state (the NO contacts would not close) and if the l30sg1 input were NOT inverted then l30sg1 would NOT go to a logic "1" which means the normally closed contacts of l30sg1 in the rung you provided would remain closed and one second after the input to 95SG was energized the alarm would be annunciated--which would be a true and correct condition.

All good so far--if we presume the NO discrete output of 95SG changes state when BOTH ignitor outputs are energized AND when the discrete input driving l30sg1 is NOT inverted.

Let's take the other condition: The discrete output of 95SG changes state when ONE of the outputs is NOT energized. In this case, if both outputs of 95SG are energized when the input of 95SG is energized the discrete output will NOT change state, and if the discrete input driving l30sg1 is NOT inverted, the normally closed contacts of l30sg1 will remain closed and after the time delay the alarm will be annunciated. Which would be incorrect.

But, if one of the 95SG outputs did not energize when the input of 95SG was energized then the discrete output WOULD change state from NO to closed, and if the discrete input driving l30sg1 was NOT inverted then l30sg1 would go to a logic "1" and the NC contacts of l30sg1 would open and there would be NO alarm. Which would be incorrect.

If the inversion mask of the input driving l30sg1 were changed to INVERTED, then the alarm logic would work correctly if the NO discrete output of 95SG changed state when one 95SG output was NOT energized, and when both 95SG outputs were energized.

So, you need to know a couple of things at this point. One, what is the state of the discrete output when 95SG is NOT energized. Two, what is the state of the discrete output when the input of 95SG IS energized AND both outputs ARE energized? I don't know how to simulate the condition when only one output of a working 95SG is not energized, so I can't help with this.

You may have to have both ignitors out and "on the deck" to see if they are both sparking when the input of 95SG is energized. Of course, if one of the ignitors wasn't working even when it was energized that would be a bad test, so it would be necessary to switch the ignitors and perform the test a second time. (The outputs of 95SG are too high to measure with a normal multi-meter.)

You also need to know if the Mark VIe discrete input driving l30sg1 is inverted or not inverted.

I generally read the information in the Device Summary and the Manual, but I trust only what I personally observe. And, then I go from there. Ideally, if it were my call as the responsible design engineer for the system, I'd want a NO discrete output from 95SG that changed state when the input of 95SG was energized and BOTH outputs were working--as in our first example above. I would then connect that to a NON-INVERTED input of the Speedtronic to drive l30sg1, which would be logic "0" when the input of 95SG was not energized, and would change to a logic "1" when it was energized and BOTH outputs were working. If 95SG was energized and one output did not work, then the discrete output of 95SG would NOT change state, and l30sg1 would NOT change state and the NC contacts would remain closed and the alarm would be annunciated.

This would be the most correct indication of the state of 95SG's outputs, and would provide an alarm if a wire were to come loose in the circuit at some point. Using a NO contact to close when there is a fault will prevent an alarm from being annunciated when there is a fault.

I hope this is helpful; it's not very easy to describe in writing. And, when it's not properly commissioned it can be a nuisance alarm for years.

Again, my recollection is that the discrete output of 95SG changes state when BOTH output of 95SG are energized when the input to 95SG is energized. If that's true, then it should be connected to a NON-INVERTED input of the Mark VIe, that changes state when the NO contacts of 95SG change state. And, then the rung you have provided will work correctly because l30sg1 will be a logic "1" when BOTH outputs of 95SG are energized when the input to 95SG is energized.

 

CSA,

Thank you for the quick reply. I was under the impression that,

a) 30SG-1 & l30sg1 are same, but I was wrong. From your reply I understand that 30SG-1 is a discrete output of 95SG and l30sg1 is a discrete input to the Mark VIe through an I/O pack, which changes it's state when 30SG-1 is closed.

b) "NC" & inversion (using NOT Gate) are same, but now I realize that an NC relay (l30sg1 in this case) can be inverted again in the logic block.

I have info only about 30SG-1 that it's an "NO" relay, but I am not sure if l30sg1 is an "NC" relay.

Hope I am correct this time, will wait for your reply before asking more questions.

 

Batman,

30SG-1 "drives" l30sg1. You should think of it like this:<pre>
NORMAL (Non-Inverted) Discrete Input
to Digital Speedtronic Turbine
Control Panel

Hardware | Software
-------- | --------
|
Physical |
Contact |
30SG-1 Input Screw l30sg1
---| |----------o------------------( )---
|
|</pre>
It's important to note that we are talking about a NORMAL (Non-Inverted) discrete input to a digital Speedtronic turbine control panel. (We'll discuss INVERTED discrete inputs below.)

It's a rung, yes, but the vertical line in the drawing represents a "division" between hardware and software. What I'm trying to represent is that a physical device (a pair of normally open contacts in this case) "drives" or in GE-speak "writes to" a software coil, energizing or de-energizing the software coil if the hardware contacts close or open, respectively. Yes, l30sg1 is a discrete input to the Speedtronic, but it's really a software coil--with associated normally open and normally closed contacts. The "coil" of l30sg1 never appears in any ToolboxST representation, or Legacy Toolbox representation, or in any Mark V representation. However, the Mark IV Speedtronic Elementary--the finest digital Speeedtronic document ever produced by GE--<b>DID</b> show the representation above. It's how I visualize <b>EVERY single</b> discrete input to any digital Speedtronic turbine control panel. Because, it's what is happening--but it's never shown in modern digital GE Speedtronic representations.

<b>Moderator's Note:</b> the sentence below has been corrected at the author's request.

In the rung above, when the physical contact (30SG-1) is open, the software coil (l30sg1) will be a logic "0", false, dropped out. Just like with any coil (software or hardware) there are contacts associated with the coil: normally open contacts and normally closed contacts.<pre>
NORMAL (Non-Inverted) Discrete Input
to Digital Speedtronic Turbine
Control Panel

Hardware | Software
-------- | --------
|
Physical |
Contact |
30SG-1 Input Screw l30sg1
---| |----------o------------------( )---
| |
| |
| |
| ---| |---
| |
| ---|/|---
|
| <b>OR</b> l30sg1
| ---| |---
| l30sg1
| ---|/|---</pre>
In the drawing above, I'm trying to show normally open and normally closed software contacts "driven by" the software coil l30sg1. The vertical line down from the l30sg1 coil is meant to show that when L30sg1 changes state, the contacts change state. Below that, I'm just trying to show another normally open- and normally closed set of contacts associated with l30sg1--and they <b>also</b> change state when the software coil l30sg1 changes state. These software contacts can be used in any rung or function block in application code--there's no limit to the number of normally open or normally closed contacts, and the state of the contacts (open or closed) is a function of the state of l30sg1. Just like with a physical relay, or pressure switch or temperature switch. When the relay or temperature switch or pressure switch is NOT actuated, normally open contacts of the device will be OPEN, and normally closed contacts of the device will be CLOSED. When the relay or temperature switch or pressure IS actuated, normally open contact of the device will be CLOSED and normally closed contacts associated with the device will be OPEN. (I don't know about NO coils and NC coils; I only know about NO contacts and NC contacts.)

When the normally open contacts of 30SG-1 are open then l30sg1 will be false, a logic "0". This means that normally open contacts associated with l30sg1 will remain open, and normally closed contacts associated with l30sg1 will remain closed. And, when the normally open contacts of 30SG-1 close the software coil l30sg1 will be a logic "1" (true). When l30sg1 is a logic "1" (true), normally open contacts associated with l30sg1 will close, and normally closed contacts associated with l30sg1 will open. Here's a table to describe this:<pre>
NORMAL (Non-Inverted) Discrete Input
to Digital Speedtronic Turbine
Control Panel

30SG-1 NO l30sg1 Coil L30sg1 NO l30sg1 NC
Contact State Logic State Contact State Contact State
------------- ----------- ------------- -------------
OPEN "0" (False) OPEN CLOSED
CLOSED "1" (True) CLOSED OPEN</pre>
So, to your a) statement, I respond that you are correct--mostly--in that 30SG-1 is a discrete input through an I/O Pack which changes the state of l30sg1 based on the state of 30SG-1. But, I want to be clear: It's a software coil (l30sg1) that changes state when the discrete input (30SG-1) changes state. And, normally open- and normally closed software contacts associated with l30sg1 change state when the discrete input (30SG-1) to the I/O Pack changes state.

I don't know how to respond to your b) statement; I don't know what your experience or frame of reference is for inverted signals. Yes, a normally closed contact is an "inverted" normally open contact. But, that's not how I think of normally open versus normally closed. I prefer only to think of coils as being 'normal' or 'inverted'--I'll explain below.

Let's talk about inverted inputs, and inversion masking--as done by GE in their digital turbine control panels. Let's presume that discrete input l30sg1 was inverted in the I/O Pack configuration; it's pictorial representation would look like the drawing below, and it would also have normally open and normally closed software contacts associated with it:<pre>
INVERTED Discrete Input
to Digital Speedtronic Turbine
Control Panel

Hardware | Software
-------- | --------
|
Physical |
Contact |
30SG-1 Input Screw l30sg1
---| |----------o------------------(I)---
| |
| |
| |
| ---| |---
| |
| ---|/|---
|
| <b>OR</b> l30sg1
| ---| |---
| l30sg1
| ---|/|---</pre>
Note the "I" in the coil of l30sg1; this represents an inverted <b>coil</b>, or, an inverted discrete input. Note also that the inverted discrete input has normally open and normally closed software contacts associated with it.

The table below shows what the states of the various elements of the discrete input will be based on the state of the normally open contacts of 30SG-1:<pre>
INVERTED Discrete Input
to Digital Speedtronic Turbine
Control Panel

30SG-1 NO l30sg1 Coil L30sg1 NO l30sg1 NC
Contact State Logic State Contact State Contact State
------------- ----------- ------------- -------------
OPEN "1" (True) CLOSED OPEN
CLOSED "0" (False) OPEN CLOSED</pre>
For an inverted input, the l30sg1 coil will be a logic "1" when 30SG-1 contacts are open (the opposite of what one would expect!), and normally open l30sg1 contacts would be CLOSED and normally closed l30sg1 contacts would be OPEN. I prefer to think of inversions as only being applied to software coils, not to contacts.

This is where, I think, a lot of people get confused. Normally open contacts can be open or closed, and normally closed contacts can be open or closed. They're not always open or closed, respectively; they change state based on the state of the device (for real, physical contacts) or software coil (for software contacts) driving them. <b>ALWAYS,</b> normally open contacts associated with a device or software coil are OPEN when the device <b>is NOT</b> actuated or when the software coil is a logic "0" (False). And, <b>ALWAYS</b> normally closed contacts associated with a device or software coil are CLOSED when the device <b>is NOT</b> actuated or when the software coil is a logic "0" (False). When the device or software coil <b>IS</b> actuated or when it's a logic "1" (True), normally open contacts associated with the device or the software coil are <b>ALWAYS</b> CLOSED, and normally closed contacts associated with the device or the software coil are <b>ALWAYS</b> OPEN (when the device <b>IS</b> actuated or when the software coil is a logic "1" (True)).

I hope this helps. Again, I don't know what your experience has been with inverted signals; quite often people refer to "inverting" a contact when they are changing the contact from normally open to normally closed, or vice versa. To my way of thinking, I don't want to get that confused with inverted discrete inputs versus non-inverted discrete inputs. Both can have normally open and normally closed contacts that <b>ALWAYS</b> behave in the same way with respect to the device or software coil status. It's only the software coil status that changes based on it's inversion mask setting (NORMAL or INVERTED).

If it's not clear, let me know. I'll try a different tack (to use a sailing term). I have read and re-read and re-re-read the above in an effort to try to ensure there are no errors--but, I'm not the best person to proof-read my own writing. If there are errors, I will try to persuade the Kind, Friendly Moderator to correct them for me.

 

CSA,

I am really amazed by your sheer persistence to offer help to the needy. And after spending lot of time in typing & re-reading, you were even ready to offer more support if it's not clear. OMG!! I would not do that to a stranger if I were you (hate to say this, though). Really we are very fortunate to have you here.

I hope It's clear for me now. Inversion is applied to the software coils, not to the contacts associated with it. So the inverted discrete input to the TCP, would result in a block like this, as you said in your reply.
<pre>30SG-1 NO l30sg1 Coil l30sg1 NC
Contact State Logic State Contact State
-------------- ----------- -------------
OPEN "1" (True) OPEN
CLOSED "0" (False) CLOSED</pre>
I'll put it this way. 30SG-1 closes on ignition exciter fault (sensing one faulty output or both exciters fault is not known at this moment), and since inversion is applied to the l30sg1 COIL, it's output to the NC contact is "zero" & which in turn gives output "1" and generates the alarm. But how to identify that an inversion is applied to a software coil? Is there any other means to know, than to find it out accidentally & hoping that it must have been inverted? Has it got something to do with "inversion masking", like it's inverted but "masked"? Guess I am asking too many questions, but I am not in a hurry to learn, so reply me when you have free time.

And thanks a ton again for your generous reply to my previous post.

 

Batman,

I don't think we know when 30SG-1 changes state, actually. We don't know if it changes state when the input to 95SG is energized and both outputs are energized, or if it changes state when the input to 95SG is energized and at least one of the outputs is NOT energized. My suggestion to you is to use Trend Recorder to monitor l2tvx1 and l30sg1 during the next START. We'll presume that both outputs are energized when the input to 95SG is energized (that's a fair assumption for now, especially if the unit and ignitor power supply are relatively new). If l30sg1 changes state very shortly after l2tvx1 changes state then my presumption would be that 30SG-1 changes state whenever the input to 95SG is energized and BOTH outputs are energized. (This seems to be what happened on units I've worked on in the past. This is a "positive" indication that the ignitor outputs are both energized when the input to 95SG is energized. If one or both outputs were not energized when the input to 95SG was energized, then 30SG-1 would NOT change state--indicating a failure of one or more exciter "channels" (I've always hated that description, but no one ever asked me what the alarm text message should have read). This would also go along with GE's philosophy of using device contacts that open to alarm or trip--because the normally open 30SG-1 contacts would NOT close if one or both of the outputs were not energized when the input to 95SG-1 is energized.

As for determining if a discrete input is inverted or not, one needs to use ToolboxST and open the configuration of the I/O Pack(s) through which the signal in question is connected, scroll to the signal name, and then check the configuration of the signal in the main window of ToolboxST. It will read either INVERTED or NORMAL (Not Inverted), or, if the input is unused I think I will read NOT USED. I forget the name of this parameter for discrete inputs, but it should be easy to locate.

I'm pretty certain that if you were able to disassemble 95SG you would find a small electro-mechanical relay, and a normally open contact of that relay would be connected to the two terminals which are 30SG-1. (It might be a solid-state relay, too; who knows. The point is: A normally open contact of a relay of some sort in 95SG is connected to the two terminals being described as 30SG-1.) So, there is some internal circuitry that energizes and de-energizes the relay I'm going to call 30SG-1, and a normally open contact is wired to the two terminals that are ultimately wired to the two discrete input terminals that drive "imaginary" software coil l30sg1. Imaginary software coil l30sg1 may--or may not--be inverted; that's done in the I/O Pack configuration for that input using ToolboxST. Software contacts associated with "imaginary" software coil l30sg1 are used in the rung you provided to generate an alarm when there is a problem with one or both of the high-energy ignitor exciter power supply outputs.

I have a question for you, Batman. Why did you choose this particular alarm and discrete input and alarm? Does this alarm get annunciated every time the unit is started? Or, has this alarm just recently start to be annunciated every time the unit is started?

I have been to MANY sites that have this alarm every single time the unit is started--and no one ever takes any action. Why? Because the commissioning person from the turbine packager did nothing, and just said, "That's normal; it will happen every time the turbine starts." Which is hooey. So, please tell us why you chose this input and alarm to inquire about. Thanks!

 

CSA,

I have not faced any problem with this alarm, but when I was going through the start up sequence, this particular block confused me. Now I have a better idea, thanks to you & control.com. I worked in the commissioning, involved in few start ups in which the turbine was loaded to spinning reserve. Now I am on to a different assignment, trying to understand Mark Vie control philosophy during free time.
From one of your posts in a different forum I understand that the exact purpose of inversion masking is to make the logic signal associated with a discrete input a logic "1" when the contact or the circuit is open. I am not sure it it's used here. Please shed some light on this.

 

Batman,

Can you provide a link to the comment so I can build on the information for the explanation?