Deposits in HP Turbine

[jsteverm]

Description of Unit.
The machine is a base loaded 250 MW 2400/1000/1000 unit with a GE turbine and a CE boiler. Water chemistry consist of Equilibrium Phosphate Treatment and conforms with EPRI recommended guidelines.


Problem:
Prior to the current outage the unit was suffering a 20 MW load reduction, due to inability to pass steam flow through the HP Turbine.

History:
First stage pressure (adjusted to constant steam flow) had increased at a steady rate, ~ 1% per month, for the past 18 months.
There was slight reductions in first stage pressure when forced outages occurred.

Internal Inspection.
Internal inspection showed severe magnetite deposits (I don't have a complete deposit analysis yet but the material is black and magnetic)on the first three stages, and lesser amount of deposit in the remainder of the HP turbine. The IP turbine was clean.

Questions
Has this condition, severe magnetite deposits, been observed elsewhere? What were the remedies?

 

[davefitz]

Usually the magnetite deposits are from exfoliation from the superheater tubes- they oxide scale will psall off during startups, and since you probably do not have a HP bypass system to dump startup steam to the condenser, thenthe oxide will deposit in the turbine. Usually this type of deposot will erode the blades and lead to a lower HP stage efficiency , and usually you can pass more steam for a given inlet pressure . But your problem statement indicates the opposite problem.

Two other sources of magetite are : (a) excess superheater spray water flow and the feed water to the spray has high levels of iron (magnetite from corrosion of the feedwater heaters) and (b) high water carryover leaving the drum ( should show up as high conducttivtiy) due to problems in the steam drum- usually high water level not correctly montitored by the level gauges.

 

[jsteverm]

Thanks for the information. We don't see much evidence of hard particle erosion (still have sharp trailing edges and no blunting of leading edge). Our monitoring of steam chemistry has shown less than 1 ppb iron but maybe that is still too high, or maybe the sample is not representative of the bulk chemistry.
We don't know why it is coating rotating and stationary parts, what mechanism could make the material stick?
We won't have the deposit analysis for a few more days.

 

[poetix99]

It is common for a turbine rotor to have some residual magnetism. For example, API Standard 612 (not applicable to your rotor, but cited as a convenient standard for comparison) requires that the residual magnetism of rotating elements is not to exceed 3 gauss (0.0003 tesla). There might be analogous EPRI guidelines...

Perhaps your HP rotor has particularly high residual magnetism.

Perhaps there are localized eddy currents being set-up while in operation...

These are guesses on my part, and not particularly instructive.

 

[jsteverm]

Thanks poetix,
We did think of that and just had a degauss firm in. They indicated that the level of residual magnetism was what they would expect after degaussing. I appreciate the thought and any other guesses that might be out there.

 

[kclim]

Where are you monitoring the iron level in the steam? I believe steam iron monitoring isn't one of the core parameters listed by EPRI. I have a source which suggests 20 ppb iron for steam is tolerable.

I'm inclined to go with davefitz option a) on this one, in the absence of any other information. I imagine iron may be introduced by the feedwater sprays - steam monitoring at the drum won't pick that up as it occurs after this.

The pressure conditions at the HP turbine stage may have been conducive to the precipitation of iron - if this has occurred, your analysis of the turbine deposits should not only have turned up magnetite but also significant amounts of elemental iron (not just from the turbine blades).

 

[EdStainless]

If you are comfortable with having the sources that have been mentioned in earlier posts under control (great list davefitz), you might look to see if there is any FAC in your system. If you have run wet for periods of time you may be thinning some pipe walls also.


= = = = = = = = = = = = = = = = = = = =
Corrosion never sleeps, but it can be managed.

http://www.trenttube.com/Trent/tech_form.htm

 

[jsteverm]

Thanks all. Here is a brief update of the situation.
The deposits were mostly copper but with bands of iron (condenser is AL-6X and FWHs are Cu-Ni and Monel). Deposits were mechanically removed. Condenser, FWHs, superheater and piping insected for unanticipated material loss (none identified). Also sampled waterwalls, superheater, both primary (horizontal) and secondary (pendant) for deposit analysis (some Cu found). Performed boiler chemical cleaning where about 300 lbs of Cu was removed. Similar Cu removal had been seen in past cleanings.

Other things done included: Steam sampling system upgraded. Steam chemistry analysers upgraded. Preparations and pipe modifacations were made to be able to perform an in-place HP turbine chemical cleaning in the future.

Thanks again for all your suggestions and help.

Jack

 

[rmw]

I did not get in on this one initially, but I am curious as to the condition of the blading/buckets in the hot reheat entry stage, wherever that is in your machine.

Do you have any accelerated erosion in that stage, and/or any deposits of any kind past that point.

BTW, I suspected copper initially but did not post since I did not know the metallurgies of the cond, and FWH's.

rmw

 

[jsteverm]

The reheat section looked good. Not much erosion and buckets clean.

Jack