advertisement
from the Forum department...
Meaning of DC (designated controller) in RST mark 6e
Power generation equipment control. topic
Posted by Controman on 25 February, 2013 - 5:48 pm
Hi,

We have Mark6e control system for 9E gas turbine, and in the RST controllers we have same LED alight. but for the R controller and differently to two others, we have the DC LED alight and I don't know what does this mean? and what specificity is added to R controller by this?

If someone has the answer I will be grateful

Best regards for all


2 out of 2 members thought this post was helpful...
Posted by CSA on 26 February, 2013 - 12:24 pm
Controman,

I think the Mark VIe System Guide, GEH-6721, has some (limited) information on the concept of 'designated controller.' But here goes, anyway....

In a TMR control panel there are three independent control processors (duh, I know!) but there is only one operator interface (HMI), and I'm referring primarily to CIMPLICITY or PROFICY or whatever it's being called these days, but also to Toolbox. The UDH connects the HMI to all three processors, each of which may have different values for some parameters and those differences can't be displayed on the HMI for every value on the HMI. So, one of the tasks of the designated controller is to report a single value to the HMI for display.

In ToolboxST when you 'Connect' to the Mark VIe one of the questions it usually asks is is do you want to connect to the designated controller, or to <R>, or <S>, or <T>? Most people just click past the window (because that's what they see the GE TAs do) without questioning or even reading the prompt and so they are connecting to the designated controller. Now, once connected, it's possible to see the Pre-Voted values of many parameters, but not all. So, to check the values of some parameters in a different processor (than the designated controller) that are not voted one would have to connect to that processor, not the designated controller.

In previous versions of Speedtronic, the "choice" of designated controller was done by different means. But, in the Mark VIe, as long as <R> is healthy <R> will be the designated controller in a TMR panel. If <R> is not healthy or has to be shut down, or shuts down for some reason, then the task of being the designated controller goes automatically to <S>. In this case where one of the three processors is not participating in the voting and control while the turbine is running, the Speedtronic no longer votes logic ("1" and "0") signals. In this case, the logic signal value in <S> is deemed to be the correct value and that's what's used in the application code. (Two processors can't vote the choice of two different values, so voting is suspended in this case and the designated controller values are used in the application code.) I'm including discrete (contact) input values in the list of logic signals in this discussion.

So, to start a turbine with a TMR Mark VIe control system <R> must be the designated controller, because to start a turbine all three controllers (processors) must be healthy and communicating--that's one of the rules of GE control philosophy of TMR control systems. But, once the turbine is running, the loss of a single controller will not (should not, in a properly configured system, with proper servo coil output polarities to all servos) trip or shut down the turbine and it will continue to run--using the designated controller's logic values to execute the application code.

It should be clear that <T> will never be the designated controller for a running turbine, because that would mean that both <R> and <S> were not healthy or were shut down, and in that case the turbine would be tripped.

Lastly, the reason the DC LED appears on all three controllers is that the controller can be used for <R> or <S> or <T>, making them cheaper to build and requiring less spare parts (if you had to have a separate controller for <R> and <S> and <T>, one that was not interchangeable or could not be used for <R> or <S> or <T>, then you would have carry more spare controllers).

The designated controller also has other duties, including keeping the three controllers in sync with each other and keeping the time synchronized for the three controllers, etc., but that's kind of "background" stuff.

Hope this helps!


Posted by Controman on 28 February, 2013 - 6:34 pm
Thank you CSA for your explanation

There was no information in GE documentation (in our site) about this DC.

I presume that there's a vote between the three different values and the accepted one is sent to HMI by the (DC)
And if i connect to <R> I will read what <R> reads, and if i connect to DC I will read what <R> takes from the voting operation.

> ...Most people just click past the window (because that's what they see the GE TAs do).

That's really what we do, but no more because connecting to only one will help for diagnostic of problems.

> The loss of a single controller will not (should not, in a properly configured system, with proper servo coil output polarities to all servos) trip or shut down the turbine

I didn't understand well what's the relation between coils polarities and loss of controllers signal (I presume that if there is a problem in coil even if controllers signals are good, we will have a problem or trip)

If you can add some information about the other duties I really will be grateful.

Dear CSA be sure that these information for this thread (and also for others) helps me and make me aware of many things.

Best regards


2 out of 2 members thought this post was helpful...
Posted by CSA on 1 March, 2013 - 1:24 pm
Controman,

Glad the information is of help. As for "other" Designated Controller tasks, that's about all I know of. Since these controllers (microprocessors) run on a Real-Time OS (RTOS), GE writes the program that makes the controllers do what they do for turbine control and protection. They don't document these tasks very well, and really, all we as technicians need to know about is what we've already discussed.

I think one of the things that most people get wrong about GE TMR turbine control panels, and wrongly attribute to the Designated Controller, is that voting, in particular SIFT voting (Software-Implemented Fault Tolerance voting) takes place in EACH controller. Each controller has an IONET port for the other controllers; in other words, all three controllers have IONET ports for <R>, <S> and <T>. So, each controller "sees" what the other two controllers have for input values (not all inputs, but most of them...) and votes those values with its own value to make the determination about what value to use in executing the application code.

This reduces the workload of the Designated Controller (if the Designated Controller did the voting) and spreads it between all the controllers.

I believe that since the Designated Controller also votes values before executing the application code when the HMI requests a value it sends its value--since it is the voted value used by all three controllers. Again, GE doesn't document these things very well (intellectual property and proprietary information and all that stuff).

The electro-hydraulic servo-valves used by GE and their packager for heavy-duty gas turbines are bipolar--meaning they operate on +10 mA to -10 mA, and anything in between. In fact, "zero" current is very important. Positive servo current increases the flow of fuel or air or steam; negative current decreases the flow of fuel or air or steam. And "zero" current stops the actuator and holds the present value of position or pressure or flow constant, until the current goes positive (to increase the flow) or negative (to decrease the flow) and when the actual flow (or position or pressure) equals the reference the current goes back to "zero" and the actuator stops moving and flow (or pressure or position) is held constant until a change is required.

The servo-valves have three coils--one connected to <R>, one to <S>, and one to <T>. If the wires to one of the coils are reversed, the turbine will still operate (usually with some Diagnostic Alarms--which nobody pays attention to anyway because the turbine is still running.!.!.!). Let's say the wires from <S> to a servo-valve are reversed and the turbine is running with all three controllers (albeit with unresolved Diagnostic Alarms associated with servo problems). Suddenly, <T> controller fails which means the servo current being applied by <T> to its servo coils is shorted--meaning there is NO current being applied by <T> to any of its servo coils. <R> is the Designated Controller in this case, so <S> will use <R>'s values in executing application code, however because the polarity of the current being applied by <S> is opposite of what it should be <S>--which has been trying to reduce the flow of fuel or air or steam all along--will now have enough power, along with the force from the null bias spring, to cause the flow of fuel through that device to be shut off which usually results in a turbine upset at a minimum, but more often than not a trip.

The servo-valve output currents are "voted" at the servo-valve, since the currents are "cumulative". When current is being applied to the servo-valve coils a minimum of two currents must be correct to overcome the force of the null bias spring which is also trying to shut off the flow of fuel or air or steam. If, at any time only one of two servo-valve currents is opposite what it should be then the device will move to shut off the flow of fuel or air or steam. And the only alarm that will be annunciated is "Loss of Flame Trip." (There were likely Diagnostic Alarms warning a conscious operator or technician of problems, but, as is usually the case since the turbine started and ran the Diagnostic Alarm(s) were ignored.)

(When properly performing servo-valve polarity checks, only one controller is applying current to the servo-valve at a time, and that is sufficient to mostly overcome the null bias spring for testing purposes. But, when only two controllers are applying current to servo-valve coils if one of them is incorrect then the electrical force of that current will add to the force of the null bias spring and move the device to shut off the flow of fuel or air or steam.)

So, to prevent unwanted problems (trips) in the (unlikely) event that one of the three processors fails it's VERY important to ensure whenever a new servo-valve is installed that the polarity of the current being applied by each controller is correct, so that when any of the controllers fails (or is powered down) when the unit is running it will continue to run.

Hope this helps!


Posted by Controman on 2 March, 2013 - 1:34 pm
Of course all these informations help and make us aware of problems to avoid.

So thank you again.

Best regards


Posted by sardar9 on 2 March, 2013 - 9:19 pm
Excellent and very thorough description, thank you SIR


Posted by CSA on 13 March, 2013 - 7:50 pm
!!!OOOOPPPSSS!!!

> Positive servo current increases the flow of fuel or air or steam; negative current decreases the flow of fuel or air or
> steam. And "zero" current stops the actuator and holds the present value of position or pressure or flow constant,
> until the current goes positive (to increase the flow) or negative (to decrease the flow) and when the actual
> flow (or position or pressure) equals the reference the current goes back to "zero" and the actuator stops moving and
> flow (or pressure or position) is held constant until a change is required.

That's wrong.

NEGATIVE servo current INCREASES the flow of fuel or air or steam.

POSITIVE servo current DECREASES the flow of fuel or air or steam.

Sorry for any confusion!!! (It's a common mistake, but it is a mistake, nonetheless.)


Posted by Controman on 14 March, 2013 - 6:25 pm
Thank you GETech for the added information.

And thanks to CSA for the correction.

Best regards


0 out of 1 members thought this post was helpful...
Posted by GETech on 12 March, 2013 - 12:04 pm
> It should be clear that <T> will never be the designated controller for a running turbine, because that would mean
> that both <R> and <S> were not healthy or were shut down, and in that case the turbine would be tripped.

This is not correct. By default <R>, <S> Then <T> is in control. But, GE uses an algorithm to look at which Processor is performing best, most healthy. It will select that one. If all are performing equally well then it will follow the default. Currently I am running on <S> but <R> and <T> are both up and running but <S> is performing the best overall.

Aside from that I do have a problem along this line. If I select to go to designated controller Controls ST says "it can not determine which is the designated controller". If I select any one of them they show correctly which is the designated controller. I show no other sign of this problem except when I try to connect.

I assume I just need to cycle power at the next opportunity to fix this, but I'm not sure what other ramifications this might have.

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2014 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.


Fortune
The best way to make a fire with two sticks is to make sure one of them
is a match.
-- Will Rogers
Advertise here
advertisements
Time to incorporate data handling, web HMI and motion in one system!
164-page eBook free download - EtherCAT Applications Guide
Control.com is the largest Automation community on the web. Learn how to advertise here now...
Servo, stepping motor control, analog & web HMI in one system!
View free setup and multi-vendor EtherCAT demo videos online
our advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive