Different CPDs

K

Thread Starter

k00

sir,

we are having two gas turbines frame 6B. it is observed that both the machines are having different compressor discharge pressure profiles.
let us take following situation:<pre>
machine1 machine2
Power Generation MW 19.5. 25.5
First Spread (SP1) oC 67.1 30.6
Inlet Air Filter DP. mmwc 45.7 69.5
Inlet Guide Vane DGA 57.00 57.00
Exhaust Temp. oC 505.5 576.4
Max Vibration 9.52 6.79
Gen Winding Temp. oC 76.0 103.0
Comp. Discharge Pr bar 7.9 7.8
Comp. Discharge Tr degC 332.0 337.0
Inlet Air Temp. CTIM oC 31.0 32.5
Last Comp Water Wash 17.9.2015 2.04.2018</pre>
However, compressor/turbine rotor of machine 1 was replaced in recent MI in august 2017. Machine 1 is having high spread 1 problem as clear from data. For being on safe side L83SPMB impulse signal is kept forced to true to make spread limit constant to its higher value.

Is this condition of different CPD profiles are normal with machines?
is there any significant relation between CPD and spread?
 
khoriwal000,

Unit 1 is operating at 25 MW; Unit 2 is operating at 19 MW. Both Units have the same IGV angle, and both Units have roughly the same CPD though the inlet filter differential pressure is higher on Unit 2.

You DIDN'T say what the FSR was for each machine, OR what fuel was being burned. It is presumed both Units have conventional combustors (not DLN combustors). It is also presumed both Units only have one CPD transmitter, and not three redundant transmitters.

It surely seems something is amiss. When was the last time the CPD transmitters were calibrated?

As for the statement about forcing L83SPMB to prevent a trip, ..., well, that's NOT safe at all. With a TTXSP1 of more than 67 deg C there's definitely a problem with something in at least one of the combustors, and it could well be a problem with a cracked combustion liner or transition piece or a plugged fuel nozzle orifice--none of which is a good thing to be ignoring by bypassing the Combustion Monitor protection. Just because something can be forced does not mean it should ever be forced. In this case, the problem could (very likely will) cause damage to one or more combustors and possibly worse (turbine nozzles and/or buckets).

As for whether or not CPD has or might have an effect on exhaust temperature spreads, it would likely be the opposite, that a spread problem would result in a CPD issue, which would be bad for the machine.

If you want to pursue the CPD issue you must check the calibration of the CPD transmitters. But, if they have been checked recently and are both in specification then the SAFEST thing to do is shut the Unit down (the one with the high spread) and at least borescope the combustors.

And unforce L83SPMB, and never force it again.

Also, you should read past control.com threads about how the Combustion Monitor works, because it's NOT all about the magnitude of any spread (TTXSP1, -2 OR -3). It's about magnitude, and adjacency (except for one case). It had been written about and discussed many times before in control.com.

And I believe you can post your email address and glenmorangie will send you a document or documents he has about spreads and the Combustion Monitor is you ask him to.

But, there's a problem with Unit 1 (the unit with the high spread and the low power output) based on the information provided).

Good luck--it doesn't sound good.
 
>Sorry, don't understand the question. CPDs are 7.8 & 7.9,
>do you expect them to be exactly the same?

Reply-
at different loads as mentioned, there should be significant difference while the difference is only 0.1kg/cm2.
 
>You DIDN'T say what the FSR was for each machine,
Reply- FSR machine1 =39.90 % for machine 2=45.11 %

>fuel was being burned.
reply- natural Gas


>conventional combustors (not DLN combustors)
reply- DLN 1.0 type

>It is also presumed both Units only have one CPD transmitter, and not
>three redundant transmitters.

reply- machine1 is having only one while machine 2 is having three transmitters.


>It surely seems something is amiss. When was the last time
>the CPD transmitters were calibrated?
reply- machine 1 transmitter was last calibrated in august 2017 while machine 2 transmitter was calibrated in 2016.

-spread high logics are known well to me and machine load is maintained such that logics are not satisfied. all the check ups as you mentioned were already done. (more than that are done). machine will be stopped in coming 2-3 days.

-how does spread problem will lead to a CPD issue?
 
It's atypical for a machine with DLN-I combustors to have only a single CPD transmitter since CPD is SO critical to DLN control and protection. But, if Unit 2 has three CPD transmitters what are the three pre-voted values of CPD? It's entirely possible that one or two of the three CPD transmitters is/are out of calibration.

I kick myself for not asking (because I'm sure I know the answer), but what Diagnostic Alarms are present on EACH Unit (INCLUDING the ones you don't believe are relevant)?

Also, since we are on the subject of Alarms, what Process Alarms are present on EACH Unit?

If there was a hole in one of the combustion liners or transition pieces then it's possible that could lead to a lower than normal CPD for the same operating conditions. In other words, a high spread could lead to a serious problem with one or more liners or transition pieces that could increase the air flow into one or more combustors such that the CPD would be lower than normal.

It would probably be a very good idea to check the calibration of all the CPD transmitters of both Units at the earliest possible time.

But based on the information provided it doesn't seem the operating conditions listed in the original post are similar, except for the values of CPD and TTXSP1. And I don't believe you have told what the values of TTXM are or what the exhaust duct back pressures are, either. Both important pieces of the puzzle--along with the Process- and Diagnostic Alarms....

I would also like to know what the values of TTRX and TNR are for both machines, as well as if IBH is active on either machine, and what combustion mode each machine is operating in.

Finally, there is an array that lists the physical locations of each exhaust T/C when sorted from highest to lowest. It is JXD_0 through JXD_17. That would help to understand where the highest and lowest T/Cs are located with respect to each other.
 
Sorry, I'm kind of embarrassed by my first answer to you. I must have been having a bad day and only looked at the CPD without looking at the load.

CSA, as usual, has given you good ideas. I'll leave it to him for the moment.
 
Top