Gas Turbine Alternator RTDs Meaning and Explanation

Hello everyone,

We are in operation of 19.6 MW GE make gas turbine (Mark Vie control system) with Brush make alternators. There are nine RTDs installed for alternator i.e. :

1. Stator coupling end : 3 nos, (DTGSF1, DTGSF2, DTGSF3)
2. Stator collector end : 3 nos : (DTGSA4, DTGSA5, DTGSA6)
3. Cold gas TE : 1 no ( DTGGC10)
4. Hot air TE & CE: 2 nos (DTGGH18. DTGGH19)

Can anyone explain what are meaning of the tags, its functions and its importance?

Thanks and regards

>1. Stator coupling end : 3 nos, (DTGSF1, DTGSF2, DTGSF3)<pre>
DTGSF1 Detector, Temperature-Generator Stator Forward #1
DTGSF1 Detector, Temperature-Generator Stator Forward #2
DTGSF1 Detector, Temperature-Generator Stator Forward #3</pre>
There are three RTDs, embedded in the generator stator at the "forward" end of the generator (also referred to as the turbine end, or the drive end, or the driven end), usually one RTD in each phase (of the 3-phase generator)

>2. Stator collector end : 3 nos : (DTGSA4, DTGSA5, DTGSA6)<pre>
Detector, Temperature-Generator Stator Aft #4
Detector, Temperature-Generator Stator Aft #5
Detector, Temperature-Generator Stator Aft #6</pre>
There are also usually three RTDs, embedded in the generator stator at the "aft" end of the generator (also referred to as the collector end, the non-drive end, the non-driven end), usually one RTD in each phase (of the 3-phase generator); I have always presumed DTGSA4 is measuring the temperature of the same phase as DTGSF1, and DTGSA5 is measuring the same temperature as DTGSF2, and so on

>3. Cold gas TE : 1 no ( DTGGC10)<pre>
DTGGC10 Detector, Temperature-Generator Gas Cold #10</pre>
This RTD measures the temperature of the cold "gas" (air in an air-cooled generator; hydrogen in a hydrogen-cooled generator--the cooling "gas")

>4. Hot air TE & CE: 2 nos (DTGGH18. DTGGH19)<pre>
DTGGH18 Detector, Temperature-Generator Gas Hot #18
DTGGH19 Detector, Temperature-Generator Gas Hot #19</pre>
There is usually an RTD at each "exit" of the air cooling the generator, one at the turbine end (the "forward" end) and one at the collector end (the "aft" end)

Should a high temperature develop in the generator stator (turbine end or collector end), the embedded RTDs could help in detecting the high temperature. Again, as was mentioned in another thread--operators should always be "trending" readings like these, because if they rise, and the rate of rise, can be an indication of a problem--possibly a serious problem if the rate of temperature rise is very quick.

By monitoring the temperature of the air entering and leaving the generator one can also gauge the effectiveness of the cooling of the generator. Again--trending these temperatures is key to spotting a problem BEFORE it becomes a serious problem.

Hope this helps!
Hello CSA ...

Yes this will help. recently the RTD DTGSA-4 was increasing gradually till 96 deg C (normal 70-74 deg C) and operator raised alarm. While troubleshooting b instrument engineer, the GT got somehow tripped. Later we came to know its trip setting is 120 deg C.

Now I have a query.... will it really require to set the particular temp at collector end to be set at 120 Deg C. I think it is not.

But anyway thank you CSA sir for the explanation and help


Would it surprise you to know that most generators supplied with a unit packaged by GE don't have trips configured for the generator RTDs? Alarms, yes; but trips, no.

GE's philosophy as a packager is that an alarm should alert a conscious operator to a problem--which might be or become a serious problem--and the operator can raise the issue with the operations supervisor or other plant personnel to decide what corrective action to take.

This is a good philosophy--because when the turbine control system is configured to alarm AND trip for just about every possible condition then the unit becomes unreliable. In the case you are describing if only one stator RTD is indicating a problem the plant personnel should be able to decide how to handle the situation. In your case, it seems the I&C technician was troubleshooting the issue without understanding how the Mark VIe handles RTDs and what can happen when an RTD is disconnected or the wiring is otherwise "disturbed"--and probably was also unaware that a single out-of-range RTD signal (high in this case) would also result in a turbine trip (if the scenario you described is actually what happened--a LOT of times trips are incorrectly attributed to something OTHER than what really tripped the turbine by people who should know better, but don't--the Trip History and the Alarm Display have the required information to be able to say precisely what tripped the turbine!). And, so, if what you say happened actually happened, this is another case of making the unit unreliable by trying to make it trip for any and all conditions and taking the ability of operators and their supervisors to make appropriate decisions.

Now, if the unit was not configured/programmed to trip on a single high stator temperature and the tech caused an out-of-range signal which would have resulted in a trip had it been configured/programmed to trip on a high-high value then all that would have happened is that the signal would have been out of range and the unit would have continued to run (which is probably what the tech thought would happen because it appears he didn't check to see if the unit would trip on a high value--which should have been done).

So, trying to use the automated turbine control system to protect the unit in just about any or every case can actually have undesired results. And, guess what gets the blame? The Mark* turbine control system! Because it tripped the turbine! And, someone thinks--now--the turbine should not trip on a single high-high generator stator temperature (real or caused by an instrument tech). BUT, when the unit was built and ordered someone in the organization that was purchasing the unit decided the unit SHOULD trip on high-high generator stator temperatures, and so it was configured/programmed that way. And that person is probably long gone and no longer associated with the unit, and now someone (you, included) doesn't believe the unit should trip on a single high-high generator stator temperature reading.

Can you see what the issue here is? It's trying to substitute machine intelligence (that of the Mark* turbine control) for operator and operator supervisor intelligence. And, more and more sites are thinking this way, and this is leading to more and more turbine trips and blame being incorrectly attributed to the Mark*. Because owners and managers don't want to pay operators and technicians to be properly trained and want to let the hero turbine control system make all the decisions about whether or not to trip the turbine to protect it from inexperienced operators and technicians.

Yes; it happens everywhere around the world. Not so much in Europe, but pretty much just about everywhere else. Cut training budgets; pay operators and technicians less; and let the hero turbine control system make all of these decisions instead of trained, experienced operators, technicians and their supervisors. (Most operations supervisors get promoted to that position from a job as an operator. And, if they didn't get much, if any, training as an operator how does that make them a good supervisor? Learning by mistake is not the best way to learn how to run a turbine-generator or a power plant.)

As for the trip value of a reading--120 deg C is about 250 deg F. And, that's a fairly high temperature for a normal, air-cooled generator being operated properly (with a good supply of clean, dry cooling air, an unobstructed air way, a clean generator stator and clean generator cooling fan blades).

Here's the deal: It's not uncommon for embedded generator stator RTDs to fail over time. And it's made worse when generators are operated at high stator temperatures for long periods of time. Now, sometimes, it's just the wiring that gets damaged, but sometimes it's the stator insulation. And, it's very difficult to replace most embedded generator stator RTDs. So, they just get left "open" circuited in most cases. I'm NOT saying that has happened or was happening at your site (you didn't tell us what the results of the tech's checks were, which should have been completed after the turbine tripped and before it was re-started). I'm just saying this is not an uncommon occurrence for many generators around the world.

Finally, RTD wiring is very confusing to most people. VERY confusing. It's actually very simple, but it can appear very complicated. And, the Mark* turbine controls will react differently when one wire is removed as opposed to another (most Mark* RTD inputs require three wires, not just two). (And, that's not documented in the manual, either.)

Anyway, you are learning. And that's good. You should be learning a LOT from your queries and the information provided in response to them here on sometimes things are NOT just technically black and white. Like hero, automated turbine control systems. They aren't what they are made out to be by salespeople and what owners and managers want them to be. Or, what they want them to be when they realize what they actually are.

Someone wrote to tell me I should have explained the "high-high" references.

Many control systems or control system providers/packagers refer to an alarm setting value as a "high" condition (when it exceeds some value in the increasing direction), or a "low" condition when the condition drops below an alarm setting value (drops below the alarm value in a decreasing direction).

And, if the condition results in a further alarm, or a trip, then the next alarm setting is referred to as a "high-high" alarm, or a "low-low" alarm.

For example, when the bearing L.O. temperature for a GE-design heavy duty gas turbine exceeds approximately 165 deg F (approximately 74 deg C) an alarm "L.O. Temperature High" is annunciated. If the bearing L.O. temperature is allowed to continue increasing and exceeds 175 deg F (approximately 79 deg C) the unit is tripped and another alarm, "L.O. Temperature High - Trip", is annunciated. The first alarm setpoint is called the "high" value, and the second alarm (trip in this example) is called the "high-high" value.

Hope this helps!