Does anyone have experience conducting an in-leakage survey (condensor system) at a petrochemical plant?

Where are typical sources of leakage?

What are the performance benefits and monetary effects?

Terry O

I suppose that condensers in power generation plants are no different in principle, having a service pressure below atmospheric, but they handle steam/water rather than hydrocarbons.

The first indication of excessive air inleakage is when the turbine exhaust pressure is above that expected for the operating conditions. This may not be that easy to tell, as weather and laod both affect it!

If air leakage is massive during unit start-up, but reduces to normal when normal load is reached, then the likely leak source is somewhere on the system at the point when the stage pressure reaches atmospheric. (Stage pressures are basically linear with load).

We used to spray Freon around all the likely leak points, and test for it at the air pump outlets. Less nasty tracer gases are now used- helium or SF6 -and more sophisticated testers than refrig detectors!

Ultrasonic listeners may help, but I had no luck with them on a unit in service. In one case, we used heaps of shaving cream, also without success. A feather from a feather duster worked though!

The cost of air inleakage? Each skerrick (a new technical term!) of higher condenser pressure means less output and worse efficiency. I set up a test on a 350MW unit to find the allowable level of air inleakage. I had some valves on the casing that I opened in turn. The exhaust pressure rose as soon as the first one was opened, so wwe were already at a critical level!

Sources vary with design. Leaks in joints of pipes going from packing to low pressure feedheaters, rubber ring decay (unusual design), and rubber flexible joint between turbine and condenser, are some I have seen.

Happy hunting! Ray Beebe

I have worked in a methanol plant (more than 15 years old) with 3 seawater condensers for steam turbines which air, oxygen and syngas compressors. Not aware of any excessive air ingress into the condensers that cause a concern. Is it really a problem?

This message has been edited. Last edited by: Josh,

IR can be an effective tool in monitoring condensers for air inleakage, if there is sufficient temperature difference between the skin temperature of the condenser wall, and the ambient air temperature. Also, there must be a line-of-sight to the surface that is leaking. Under these conditions, the air that is being drawn into the condenser at the leak point will cool or warm the condenser wall around the leak, and create and observable "halo". Power plants use this effectively, and I've attached a couple of images that show air inleakage at manway locations. It can also be used effectively at valve packing, rupture disks, and piping flanges. I will try to post an article by Mark Lanius of Exelon Power on such applications at a nuclear power plant, that may be relevant to petrochem condensers as well.


The advantage over mass-spec gas detection methods is that you can cover a large area quickly, and pointpoint the sources of concern for further evaluation. Ultrasound isn't always the best (as you might know, Terry), because the turbulence in a vacuum leak is created on the other side of the wall. Sometimes a vacuum leak can be confirmed with ultrasound by spraying a soap solution or leak amplifier over a suspected leak area, and hearing the "sucking sound" with ultrasound.

Rich Wurzbach
ASNT PdM Level III IR and Thermal Testing
Maintenance Reliability Group, LLC

rwurzbach@mrgcorp.com

Many years ago I assisted in an energy loss assessment of a nuclear plant steam turbine. I recall that calculations of heat rate had a threshold accuracy of about 1 to 2%, so the analysis was not definitive that condenser leakage or other issues were present. There were about 40 small-bore isolation valves (valves that are normally closed during operation) that I tested for leakage with ultrasound. I found several "suspect" valves that could be leaking. The energy savings (heat rate improvement) was well worth the testing and valve repair. I am continually amazed by many plants that don't have an energy savings program, and have not assigned costs to condenser leaks, compressed air leaks, steam traps, etc..

Walt

"Is air inleakage really a problem?", Josh asks.

In a power plant, yes indeedy. Each increase of 1 kPA in turbine exhaust pressure above design is worth about 0.7% drop in output, and also worsens heat rate (thermal efficiency inverted).

For the auxiliary drives that you mention, this cost will be in more steam consumption, and may be minor.

Ray

At our Power Plant, we are facing extensive foaming in the turbine lube oil since startup of the Unit afetr major overhaul. The anti-foaming additive has been effective for few days only. Does any one have experience about such occurances?

Secondly, and more importantly, what technique can be employed for confirming whether there is air ingress into the system or whether the oil is deteriorated. Will the methods like ultrasonic or IR work for detecting the location of air in-leakages?

Asad,

There are two distinct types of air issues with lubricants. One is Air Entrainment, the other is Foaming. Air Entrainment is when an excessive amount of air is introduced into the flow or in the reservoir of an oil system, typically either through excessive agitation, or through leaks in the system that introduce air into the flow. In a recirculating system like a turbine reservoir, there is always a little air entrainment as the oil returns to the reservoir. It is typically mitigated by the use of ramps or diffusers, or just in the layout and design of the reservoir. The important this is that the air can be released from the oil before it enters the pumps for return to the system. Compression of the air bubbles in a pump can cause "dieseling" which degrades the oil. The other source of entrained air is from leaks in the piping. These leaks are hardly ever "one-way" leaks, and typically can be seen on the outside of the piping as oil leaks from fittings. One of the first things to check is to thoroughly clean the lube oil piping, and then go back and check for any leaks. The use of ultraviolet dyes in lubricant, and searching for the leaks with a UV light source is an option, but I never had much luck--always seemed to find the dye everywhere, and couldn't pin down the source.

As oil degrades or is contaminated, it can lose some of its ability to release air bubbles. This property of oil can be tested by ASTM D-3427, and compared to the performance of new oil. However, since you are putting new oil in and it "loses" its "anti-foam characteristics" over a few days, I don't think you are seeing a true change in additive or lubricant performance with regard to air release.

Foaming is different, and occurs when entrained air rises to the surface of the oil. Instead of popping and releasing the air, the oil bubbles persist and accumulate. This can lead to leakage, oil level issues, and also air being drawn back into pumps. Turbine oils are provided with anti-foam additives, which can be diminished over a long period of time. More commonly, though, their effectivness is reduced by the presence of contaminants. Moisture and particulate contaminants can greatly increase foaming for this reason.

My suggestion is to start by identifying whether your problem is foaming (surface phenomenon) or air entrainment. I would also have the oil analyzed for contaminants such as moisture and particulate, and for degradation by either voltammetry or RPVOT (I wouldn't rely on TAN-can miss early signs of degradation). If you believe that there is an air entrainment issue, and there are no lubricant problems, take the time to clean the piping and check for leaks as I explained. If it is a foaming issue, rule out contaminants, and then look at the design of your reservoir to see if a diffuser or ramp or other design change is needed to reduce agitation of the oil. If none of these are successful, you may need to review the air-release properties of your lubricant, and see if another product will perform better--but that would be my last resort.

Below are images of foaming (first) and then air entrainment (second). This images are courtesy of Noria Corporation. Their training programs are outstanding, and I recommend them for a better understanding of lubrication issues such as air entrainment and foaming.





Rich Wurzbach
Certified Lubrication Specialist (CLS)
Maintenance Reliability Group, LLC

This message has been edited. Last edited by: Rich Wurzbach,

Here is the link to the article by Mark Lanius on this topic:

IR Monitoring of Condensers-Mark Lanius


Rich Wurzbach
Maintenance Reliability Group, LLC

quote:

Originally posted by Ray Beebe:
"Is air inleakage really a problem?", Josh asks.

In a power plant, yes indeedy. Each increase of 1 kPA in turbine exhaust pressure above design is worth about 0.7% drop in output, and also worsens heat rate (thermal efficiency inverted).

For the auxiliary drives that you mention, this cost will be in more steam consumption, and may be minor.

Ray