every hour we take readings of exhaust spreads from the mark V <I>. we take exhaust spread 1, exhaust spread 2 and allowable spread. examples are as follows:

exhaust spread 1 = 39.7
exhaust spread 2 = 34.3
allowable spread = 70

why do we have to take these readings and its significance? it is a frame 9E gas turbine.

 

Hi BBSALO,

The exhaust spreads parameters are required to protect the gas turbine from combustion problem that might originate from the combustion cans.

GE gas turbines employ can annular type combustor. Combustion happens in each of the combustion chambers distributed evenly around the shaft. As opposed to the type where combustion happens in a single chamber and then only distributed to the turbine, the can annular type will have a problem if the combustions do not produce almost equal energy output, meaning the combusted air temperature to the turbine stages is not balanced. This might be due to, most of the time, fuel nozzle clogging, and causing out-of-spec air/fuel ratio.

Now imagine that the combusted air will pass through the 3 turbine stages in almost a straight line (in real situation it will take a certain swirl angle) to reach the diffuser, so there will be 14 straight lines from 14 chambers. And imagine the turbine blades rotating tangential to the flow path of the hot air, cutting through the path 50 times per second (3000 RPM). If the all the 14 channels of hot air produce somewhat equal temperature, the turbine blades (bucket) will be safe. But what if one or two channels produce a significant cooler temperature (due to combustion problem)? The buckets will now experience rapid cooling and heating cycles at a rate of 50 times per second per channel. That will impose a detrimental effect to the bucket alloy.

So, the exhaust spread is a way to monitor this and warn the user should it detects symptom of combustion trouble and trip the turbine if it becomes excessive.

Hope it is clear to you.

P/S: I notice you are still using <I> HMI. If you are interested on a retrofit, our company has a 100% Mark V compatible replacement solution for it. Please visit our website at www.tmosscada.com for more information or you may email me at [email protected].

Cheers,
Kevin Goh
www.turbineservices.at

 

A high spread usually means a flame loss in one combustor and it is cause of trip.

There is a relationship between the bladepath that detects the spread and the combustor that is flameloss, taking into account vorticity of the flue gas.

 

Good answers, but the main question was, "Why do we have to take these readings?"

The reason readings are recorded is to develop a trend over time of some variable(s), and the biggest reason to trend readings is to try to anticipate problems before they become emergencies. Conscious operators will closely monitor readings, some even plot them, and when they begin to see a change--a "trend"--they alert supervisors and other operators to the condition.

The Speedtronic, while a very capable control system, does not monitor changes over time--which is just about the most important operating consideration. Yes; it will trip when the spread gets high enough, but unless that happens very quickly (and sometimes it does) there is usually a gradual increase, sometimes fast; sometimes slow. Usually, what happens after a trip occurs is that some operator or supervisor goes back and looks at the trended data and will usually find evidence the spread has been increasing for some time--but that's a little late to find out. As was said, conscious operators and supervisors will monitor the recorded data and spot changes and try to anticipate problems before they become trips or emergencies.

While these values listed by the original poster are good values to trend, they do not indicate where the spread problem is occurring. There are two arrays created by the Mark V for the purpose of detecting and alarming a high exhaust temperature spread: JXD1_n and JXD2_n. These two arrays list the exhaust thermocouples by reading, from highest to lowest and from the location of the hottest to coldest--and these two pieces of information, along with the actual spread trend, will help to find and track an increasing exhaust temperature spread and eventually lead the maintenance department to the exact combustor/fuel nozzle where the problem is occurring.

An exhaust temperature spread is significant because as the turbine buckets rotate around the turbine past the 14 combustors of a GE-design Frame 9E heavy duty gas turbine they will be heated and cooled as they pass the problem area. This happens, as was said, at the rate of 50 times per second--and if the temperature differential (the spread) is high enough it will lead to failure of the turbine buckets, which usually leads to very catastrophic and expensive problems. It's the continual thermal cycling of the buckets (the continual heating and cooling of the buckets as they pass the problem area) that causes the failure.

Trending of any variable while the turbine is operating is very important. Operators who just write values down without looking at the previous reading, the day's previous reading, the week's previous reading are usually going to find they could have anticipated an alarm or trip before it occurred.

Which is not when they should have realized there was a problem a-brewing (in progress).

 

Don't really understand why you do it on an hourly basis. We take it only once in a shift.

That being said, exhaust spread can assume significance in the following scenarios in liquid fuel:-

1] Incorrect calibration of flow divider to the 14 nozzles

2] Jamming of flow divider planetary gears due to insufficient lubrication [usually done by certain additives like nalco/hitech to the liquid fuel]

Even otherwise, there is the possibility of cracking/failure of cross-fire tubes or leakage through them. Similar is the case with your flow sleeves, combustion liners & fuel nozzles. However, you cannot determine what the exact problem is, just by seeing the exhaust spread. If you are having more than 1 high value of spread, first check whether the highest & second highest thermocouples and the lowest and second lowest thermocouples are close by. If they are, better to go for a planned shutdown as soon as possible. In your case, the values are a bit on the higher side. In case the spreads breech the 50 degree mark, better shutdown. Both a rising trend for exhaust spread as well as high exhaust spreads are detrimental to your gas turbine. In between, how much candela are your 4 flame scanners showing? While this is the case for reverse flow turbines with a higher flame stability, the situation is worse for straight flow gas turbines. In case you have a straight flow gas turbine, I think you should avail a shutdown soon.