We just carried out a CI on the unit (Frame 6B - MS6581) and have re-used all CI hardware as they were all in good condition. This included the Fuel Nozzles. While the unit was being started and running up to FSNL, we get very hight spreads with the T/C profile looking like a bathtub curve with the low or cold area over the T/C numbers 6,7,8,9 & 10. The spreads gets as high as 274Degc (Sp1 274/ Sp2 269/ Sp3 265) but eventually settled back down to about 66 DegC (Sp1 66 / Sp2 57 /Sp3 54)when the machine gets to FSNL and settles over a period of about a minute. The Allowable spread also follows a downward trend untill it gets to around the mid 50's. The machine then trips of\n combustion trouble.The spreads on this unit were fine before we shut down for the overhaul. However we only have the base load spread info which was Sp1 30, Sp2 27 and Sp3 24. We do not have the FSNL data prior to shutdown to compare.

What is interesting is that the low or cold area stays fixed over the same thermocouples (6,7,8,9 & 10) as the machine runs up to FSNL.The curve or depression gets smaller as the machine settles at FSNL but the cold area is still stays over the same T/C's. I would have thought that if it was a fuel nozzle or can hardware problem, the cold spot would move along as the engine runs up to FSNL, just as you would expect if the laod was changed. We are unable to synchronsie the engine and load it up to see what happens due to a transformer issue that we are working on.

Any comments would be welcome.

 

James Murray,

There's not a lot of air flowing through the compressor at part speed (not nearly as much as when at rated speed).

High spreads during acceleration are <b>generally</b> a function of too little fuel during acceleration. This is especially true of liquid fuel start-ups (you DIDN'T say what fuel is being burned), but it also happens to gas fuel-fired machines during start-up if the fuel flow is just a tad shy of what it should be.

Usually, if that's the only problem (too little fuel flow during acceleration) the spreads will diminish at about 90-100% speed. The machine cross-fires again reigniting all of the combustors as the air flow increases as the unit nears rated speed.

However, if there are problems with fuel nozzles or check valves (liquid fuel check valves or purge air check valves if the unit is dual fuel and being started on liquid fuel--we don't know what fuel is being burned) then even if all the combustors are lit the spreads will be high at FSNL.

The allowable spread limit (TTXSPL) is enabled when the unit reaches FSNL and will ramp down to the calculated value, and if the actual spreads (because it's not the "absolute" values of the spreads, it's also adjacency) are outside the limits the unit will trip.

If the unit is being fired on liquid fuel, you can use the manual selector valve at the Liquid Fuel Flow Divider to check the pressures to the combustors. Any pressure which is roughly 10% higher or lower than the average of the 10 combustor pressures is suspect. The Liquid Fuel Check Valves may not be working correctly; the fuel nozzles may not have been reassembled correctly. Just because the same hardware was re-used does not mean it was reassembled properly even if it was put back in exactly the same locations. Some fuel nozzles have internal components which have to be torqued to proper values and have crush gaskets which must be replaced. Some internally threaded fittings require staking to prevent loosening. Again, just because it was "fine" and re-used, does NOT mean it was reassembled correctly.

If the unit is dual-fuel and is being fired on liquid fuel, then look at the Tell-Tale Leakoff to make sure none of the Liquid Fuel Purge Check Valves is leaking (there should be NO flow coming out of the Leakoff; if there is flow, then one or more of the Purge Check Valves is leaking, and the low pressure readings at the Flow Divider will point you to the failed Purge Check Valve(s)).

If I recall correctly, there are seal strips between the transition pieces where they fit into the nozzle segments. These seal strips must be assembled correctly and have tolerances which should be checked and corrected during assembly/re-assembly and if they're not correct then excess CPD can pass through the seal strips and cause excessive spreads.

But, exhaust temperature spreads are almost NEVER caused by the control system. The spreads you are reporting were borderline "high" at Base Load before the CI. If you doubt the control system, inspect the exhaust thermocouple radiation shields to be sure the exhaust thermocouple tips are properly centered in the opening and are not bent or touching metal. (Millwrights and pipe fighters just LOVE to stick their thick fingers into the openings and bend the T/C tips; I've never understood this fetish behaviour.) You can also use a thermocouple simulator to test the exhaust thermocouple inputs to the turbine control system. But, I'd bet pretty good money that the exhaust thermocouples and wiring are good (unless they were disturbed) during the CI, and usually they are not.

Another thing to check, the allowable exhaust temperature spread is calculated using Compressor Discharge Temperature T/C readings (have a look at the logic/sequencing/application code running in the Speedtronic!). If the CTD thermocouples were disturbed during the CI and not reconnected properly after the outage, then that could cause a problem with the calculated allowable exhaust temperature spread. It usually doesn't have a large effect, but it can have an effect if the spreads are borderline high to begin with.

 

Dear James Murray

You never said what kind of fuel you’re the machine works in start time?

However you can Inspection the machine's component's before restart:
Inspect Flow divider press. and Fuel check valve
Inspect Combustion liner (hole's fracture & c rack's)
Transation piece seals for proper installation and leaks
Check leakage from cooling flow on the nozzles (step 1&2 & exhaust frame)
Inspect Fuel nozzles
Inspect Cross fire tubes
Check Exhaust thermocouples

 

We had sort of the same problem once. We had lost the diffuser off our compressor bleed valve discharge. It would blow cold air on the thermocouples and cause a temperature spread trip.