OT: Gas vs Steam Turbine Warmup

I work at a 2 on 1 combined cycle power plant; we also have 2 simple cycle (gas only) units next door. I used to work at a steam only power plant. The gas turbines ramp up at startup very fast compared to the steamer and compared to the steamers at my former plant. I am hoping someone can explain to me why the gas turbines are able to ramp up so quickly vs steamers. And please don't say, "Because they are designed that way." I already know that. It's obvious. But surprisingly, this is the answer I get from seemingly otherwise intelligent people.

I know there are pretty tight tolerances between the rotor blades and stationary blades, or buckets in the steamers. The Siemens turbines I worked with before have roughly .030" clearance. So I fully understand the need to warm things up slowly to ensure uniform rates of expansion between the rotor (relatively small mass of metal) and the casing (pretty big mass of metal) and prevent contact, called "rubs", which damage blades. No doubt the gas turbines have similar tolerances? Someone I was discussing this with speculated that the gas turbine casing might have less mass than the steamer, so it maybe expands at a rate closer to that of the rotor. Is this accurate?

Anyone know about this stuff?

northeastconfederate said:

Anyone know about this stuff?

Click to expand...

The guys who spent time on the Navy ships (not me) certainly know this stuff.

I believe the major difference is what happens to the steam if it cools too quickly--water. You DO NOT want water condensing halfway through the turbine.

steam

northeastconfederate said:

I work at a 2 on 1 combined cycle power plant; we also have 2 simple cycle (gas only) units next door. I used to work at a steam only power plant. The gas turbines ramp up at startup very fast compared to the steamer and compared to the steamers at my former plant. I am hoping someone can explain to me why the gas turbines are able to ramp up so quickly vs steamers. And please don't say, "Because they are designed that way." I already know that. It's obvious. But surprisingly, this is the answer I get from seemingly otherwise intelligent people.

I know there are pretty tight tolerances between the rotor blades and stationary blades, or buckets in the steamers. The Siemens turbines I worked with before have roughly .030" clearance. So I fully understand the need to warm things up slowly to ensure uniform rates of expansion between the rotor (relatively small mass of metal) and the casing (pretty big mass of metal) and prevent contact, called "rubs", which damage blades. No doubt the gas turbines have similar tolerances? Someone I was discussing this with speculated that the gas turbine casing might have less mass than the steamer, so it maybe expands at a rate closer to that of the rotor. Is this accurate?

Anyone know about this stuff?

Click to expand...

.
i do not know for sure but i do know a steam radiator warming up with steam prematurely condensing makes that knocking , ping, bang sound. i know pockets of water have a habit of delayed boiling and they when start they explode (probably the bang sound).
......... i had a alcohol still not do anything, then bang blow the top off and go woosh all the way to the ceiling. maybe the warm up is to minimize the knocking, ping, bang thing ???
......... in a alcohol still i eventually used porcelain or ceramic beads. they were recommended to break up big bubbles into little ones. after the alcohol still exploded and left a big stain on the ceiling i did not under estimate any more warnings and recommendations. i only had to look at the ceiling stain which i never did get totally washed off to remind me.
.......... on a steam turbine how much is absolutely necessary ? i do not know but maybe they got experience like turbine damage that is minimized by a longer startup. the turbine manufacturer would know best usually

I work on flight engines which spool up relatively quickly compared to ground units. Look at the amount of rotating mass involved in the steam engine.

On the .030 of clearance. Those blades are going to grow as they warm up but also stretch with the centripetal force (that the case is not experiencing) so it's going to get even tighter. Some rub isn't the end of the world (they usually have specialized coatings on the tips for this purpose). The tighter the blades the more performance out of the emgine.

The power industry has a lot of Navy people. About half - or more - of the employees here are either retired or former Navy. Same thing at my last plant; I bet 70% were former Navy. They know how to operate, for sure, but not much about engineering, at least from a design standpoint. And they all have steam turbine experience, but I haven't heard of GT's in the Navy. Maybe they do have them; aren't GE locomotives gas turbine driven? A ship could do the same thing, I guess, right?

Condensation partway through the turbine is what makes the most sense to me so far; but we have turbine drains that stay open during warmup/startup for this reason, so I don't think it's that. Water induction is a very big deal with steam turbine operation, though, that's for sure. I am pretty sure it's the differential expansion issue that is the driving force here. I just don't get why one type can ramp up faster than the other given similar concerns. The turbine manufacturers publish very specific ramp rates, or warmup curves, at least on the steamers. I expect the same is true for the GT's, but I haven't seen them, at least not yet (I'm relatively new to this plant). Depending on the starting temperature of the turbine casing these vary greatly. They aren't linear, either.

I'm not so sure about a little rub not being a big deal in these turbines. I've had blades in my hand (we have several boxes of them in our warehouse) - there isn't any visible coating on them, and in fact, some of the old ones have the corners knocked off them from apparent contact. So I think any rubbing is a no-no in this application.

At this plant the combined cycle GT's have solid rotors while the simple cycle GT's have hollow rotor shafts (different manufacturers); so it seems like the expansion rates of these rotors would differ from each other and therefore the required ramp up curve would be different, but they both come up to full load in roughly the same amount of time (this may be a totally irrelevant comparison, since they're different manufacturers and different capacity - the CCGT's are ~80 MW each while the straight GT's are 180 each).

We had a GE engineer in here giving us some training on the starting devices for these things (the generator is used as a motor to roll the GT's), but he was an electrical engineer and spent his entire career with GE (40 years) working on the starting devices. He's one of the people who told me, "Because that's how they're designed."

I've been poking around to try to find a real answer to this, and it suddenly occurred to me that this forum might be a good resource. There's been some interesting discussion on boilers of late, for instance. Hopefully someone will come along and see this who knows.

I'm wondering if the gas turbine is exposed to such intense heat from combustion gases that they're built with larger clearances which then tighten up with heat. Maybe the clearances are so much bigger cold that the rotor can expand that much more quickly. But then, I understand that GT's use "cooling air" from the compressor section to keep from burning the turbine blades up, so this theory may be a dead end. In any event, I am purely speculating. I'm hoping to put an end to my speculation with some hard answers.

The navy does use gas turbines in some smaller warships - if memory serves they are duel drive ships with diesel's for long cruise and GT for high speed. Think destroyer or frigate, rather than aircraft carrier.
For example: Oliver Hazard Perry-class frigate - Wikipedia, the free encyclopedia
Also: Navy Gas Turbine Systems Technician Rating


How clean/dry is the gas going into a GT? And how hot is it? Versus how hot is super heated steam?

It might well be that at the temps and pressures gas is introduced into a GT, condensation inside the turbine is a non-issue, while with the steam condensation is high probability hazard.

[If the gas is "perfect" and the burner is "perfect" the burner exhaust driving the turbine would be CO, CO2, H2O, right? The gas plume driving the steamer is presumably H2O....]

Just a WAG, but...

I would think that the slow startup of the steam turbines is all about condensation and thus accumulation of water in the turbine. The same procedure had to be used on piston type steam engines.

If there's no coating on the blade tips check the case. Certainly on flight engines they often use an abradable coating on either the blade tips or the case. And the tolerances aren't significantly different. What's the operating temperature of the turbine blades?

For reference an air cooled inconel turbine blade in a gas engine is going to see around 1800 deg coming off the combustor (from memory, didn't confirm).

northeastconfederate said:

The power industry has a lot of Navy people. About half - or more - of the employees here are either retired or former Navy. Same thing at my last plant; I bet 70% were former Navy. They know how to operate, for sure, but not much about engineering, at least from a design standpoint. And they all have steam turbine experience, but I haven't heard of GT's in the Navy. Maybe they do have them; aren't GE locomotives gas turbine driven? A ship could do the same thing, I guess, right?

Click to expand...

A friend of mine was in the Navy in the mid 70s as a Machinist's Mate aborad the USS Forrestal (CV-59).

He said they would do random "unannounced" emergency drills simulating battle readiness. He said they always knew well in advance of the drill because they had to use all available steam on the ship. This meant 'turning on' an additional 10" steam pipe, which was shut off most of the time. Problem is you cannot just turn the valve and get power like a waterline. You had to SLOWLY open the valve and get all the piping up to temperature over a 2 HOUR period--which the crew could clearly hear as the piping hissed and banged as it warmed up. Only after the steam pipes were up to temperature and ready for service would they sound 'General Quarters' for the drill.

I would think the longer warmup for steam is not so much water in the turbine as it is water condensed in all of the steam piping. That water has to be evaporated off into stream and not shot as a liquid into the turbine. Liquid into a spinning turbine will unbalance it, cool it, and the casing, unevenly possibly causing rubbing or liquid locking. Think about why a jet engine has a hard time when a bird goes through it. A bird is roughly 65% water. At those pressures and tight clearances water acts almost like a solid.
Bill D.

The startup process has a warmup procedure for each part of the system. The piping is well heated before steam is ever admitted to the turbine. It's a permissive to admit steam to the turbine. I don't know what the numbers are at this plant, but the steam being admitted to the turbine is plenty superheated and there are traps along the way to get rid of any possible condensation before it gets to the turbine inlet. These are sophisticated drain pots with drains that are piped to a tank that is under vacuum, so believe me, the water gets to the tank, not the turbine. There is a permissive that the steam temperature and the turbine casing must be within so many degrees of each other (forget the number but it's not a big spread) on a warm start. In fact, on a warm start the danger is not that the turbine will cool the steam and condense it, but the opposite: that the steam won't be hot enough to match the turbine metal and cool it, causing differential shrinkage and subsequent rubbing.

The issue here is differential expansion, not condensation, that much I am sure of. The condensation issue is dealt with by various means, like I mentioned, including internal and external drains. As a matter of fact, the plant I just left has indication of the HP turbine exhaust temperature including how many degrees of superheat. So the "cold reheat" steam (HP turbine exhaust, which then went to a reheater section before being admitted to the IP/LP turbine) leaves the HP turbine as superheated steam. I think this is true even during the runup.

So the question is: why do steam turbines have such different differential expansion constraints than gas turbines?

As for GT temps, i'm not 100% sure of these figures, and won't be able to confirm them until Tuesday night when I go back to work, but I THINK the combustor temps are in the 2200 degrees Fahrenheit area, so that's going to be the approximate hot gas entry temp to the turbine; the exhaust temp I am sure of: It's roughly 11 to 1200 degrees. So the hot gas gives up 1000 degrees approximately as it goes through the turbine.

Fuel gas doesn't ever enter the turbine (unless it doesn't get burned in the combustor), so we're talking about combustion exhaust when we are talking about the hot gas path.

I never really thought about the amount of heat given up; it's more than in a steamer, but not b much. Our HP steam enters the turbine at 927 and finally enters the condenser at about 110. This control system doesn't have turbine exhaust temp indication so I don't know exactly what that number is, but let's say it's 150. Further proof that condensation isn't the issue is the admission of LP steam to the turbine. I don't know what stage (or even how many stages in this particular turbine) but 75 #, ~450 degree steam is admitted to the LP section of the turbine to help cool the HP steam as it passes through. This is probably to make sure it doesn't overheat the exhaust hood and condenser, or overcome the hood sprays, which spray water into the turbine exhaust to help it condense. So it's giving up a little less than 800 degrees.

The more I type on this, the more I realize I've gone off topic on my own off topic thread: I'm asking about warmup and differential expansion differences between GT's and steamers, not operating differences. Oh, well, I'm not going to delete it........

Typically the more heat extracted between turbine entry to the exhaust means more power extracted by the turbine.

The only limitation I ever observed on a gas turbine was the requirement for an indicated oil pressure rise passing 60% rpm on the start sequence. Once oil pressure was noted, only an extreme fuel flow rate would cause a shutdown. Regards, Clark

I expect it is about condensation primarily, but also about evenly heating everything for expansion.

That said, I know how to warm up a steam engine slowly. You have steam pipe valves, cylinder valves, and you can let in just a bit, letting it condense and warm the system as it blows through.

But I have no clue how to slowly warm up a gas turbine. Once you light it off, the EGT is going to come up pretty fast, I'd suppose, fire is fire, and fire isn't a lot cooler at idle. Might have more cooling air mixed in, and it probably won't be as hot as at WOT, but it still isn't going to be coming up *that* slowly. It ought to be at a good percentage of max temp pretty soon.

That being the case, the folks suggesting "because they are designed that way" probably have a point.... it would HAVE TO BE designed that way or fail.

Turbines are really damn cool. Ive had the chance to spend a little time in somw gas power stations and those things are cool!

NE I work on all of them as a millwright. Happy to share what little I know.
The gas turbines are sold as a rapid start up machine. They do have less mass in the shell, especially since most steam machines have an inner and an out shell. They have more clearance but all of this is not enough. I have worked at a combined cycle plant near here that was originally intended to be a peaker. The maintenance manager told me that it was a maintenance nightmare until he convinced them to run it all the time, even at reduced load. The problems went away then. I recently worked on a Westinghouse 501 that had a little grenade problem. Like it became one. It was and is used as a peaker in a combined cycle plat running about 14 hrs per day. The rotor in these is a stack of segments bound by tie rods 21/2 inch in diameter. One of the rods broke at the first thread on the compressor end . the stump and nut went throught compressor. Most of the compressor was shovel size bits. My belief is that the repetitive expansion and contraction was the cause. Some veterans in the industry agree. It is hard for the OEs to admit that their selling point is wrong, especially in view of the liability.
The latest is the "aero-derivitive" turbines. You will hear some idiots claim that it is an aircraft engine. NOT!! I know of no GE aircraft engine with a 10 inch shaft coming out the exhaust. They are built like an aircraft engine, with a turbine shell maybe 3/4 inch thick instead of 21/2 inch or so. They are billed as capable of coming to full power in 10 minutes. Obviously this due to less metal to warm up to operating size.
So I think the difference is in the mass to be warmed up or "grown". The steam turbines have an inner and outer shells with a total thickness of 4-10 inches depending on the machine in the HP section. The gas engines are about 2 1/2 -3 inches. The "aerodirivatives" are much less and much faster warm up or " growth".
Joe Webster would know more than I , I suspect.

The aero derivatives that I've worked on had nearly identical cores to the flight engines. Instead of a fan they drove a shaft.

Hi

I was a project manager for a power station with 2x GE LMS100 100MW gas turbine aero-derivative. In the core is a Jumbo jet engine with a turbo charger and intercooler. These double the power of the standard engine.
The engine has a 10 minute start cycle from cold steel to full power. Much of that time is taken pushing fresh air through the engine to purge any remnants of gas/fuel mixture. The actual starting cycle from idle to full power is about 2 minutes.
GE were looking and reducing the total start cycle to 5 minutes.

Generally the limiting factor for steam turbine start cycles is the metal fatigue and creep in the boilers. Fast starts stress the boiler components.
If any liquid water reaches the turbines, it acts like shooting bullets on the blades and wrecks them.

I have only been around model liquid fuel turbines. Theres really not much slow about start up, you have to get the compressor bringing in enough atmosphere to get the burners to do there thing, some of the ones i was around had to be doing over 1K rpm to get enough to go from a stall to start generating enough energy to accelerate the turbine speed.

With a steam turbine you have the option of bringing things up slow, you don't have to have it spinning to bring in air for the combustion process. Really to a degree a steam turbines just a fan blade on start up and you can just softly blow on it to get it to start rotating.

One of the other temp - spool up requirements is probably related to the pressures. I don't know what a steam plant runs at pressure wise, but sayeven a couple of hundred psi is really going to bend those large steam rotors over side ways till they have some centripetal force to help resist it. Gas turbine has minimal pressure till its at a fair speed in comparison.

I'll ask the Siemens guys this week. I thought it was to warm up the iron as like said above water drops can wreck blades. Vibration from thermal differences may be part of it, as our turbines are one high pressure stage and two low pressure stages. It seems like it takes us 2 days from cold iron to get rolling. (1800Mwe). All the combustion turbines I have worked on could cold start in a couple minutes and hot start in 30 seconds. Those have all been 50Mwe and smaller though.