Can't figure out why we have to continually vent generator casing to increase hydrogen purity. Plant manager wants to maintain 97% we bleed and refeed at 95%. Scavenging flow rates are 2500 on both CE and TE per GE. Any thoughts?

 

Doug Toner,

You didn't provide a LOT of information. Are the scavenging flow-rates the same at both ends of the generator?

Presuming its one of the ends, which end of the generator has the low(est) hydrogen purity?

Is the generator casing purity increasing or is it one (of both) of the ends of the generator?

Has vibration increased on either end of the generator recently?

How long since the seal rings have been examined or replaced?

How old is the machine?

What is the oil quality? How old is the oil?

And, of course, the most important question of all: <b>When did this problem start?</b>

If you have a GE-packaged turbine, then you will find some decent information about seal rings in the manuals. Also, I believe they have recently published a new GEx-nnnnnn publication about seals and seal oil (sorry; I don't know the designation). I know there is at least one TIL out about seal rings, also.

The first thing you need to know is that with hydrogen pressure inside the generator casing at approx. 2 barg (about 30 psig), air CANNOT leak into the generator from the atmosphere. Under normal circumstances, air comes from the seal oil.

The second thing you need to know is that seal rings and the springs that hold them together wear out over time, and sometimes the springs break.

When you are trying to make sense of a problem like this, it's important to have good data to analyze--and in this case it's Seal Oil Flow-rate over time. Unfortunately, the Seal Oil Flow-rate is not typically an input to the Speedtronic panel, and trying to get operators to take gauge readings is next to impossible. But, if you have data, and oil temperature is pretty constant, then you should see a rise in Seal Oil Flow-rate as the purity started decreasing--this is presuming the hydrogen you are putting into the generator is high purity. (I was at two sites that received low-quality hydrogen, and it was maddening to find--and no one could believe the supplier was not providing quality gas. We spent man-weeks of time trying to get to the bottom of the problem. And, it was--quite literally--the last thing we suspected, and looked at.)

I'm also presuming the seal oil differential pressure is constant.

999 times out of 1000, air comes from the seal oil (or the hydrogen gas cylinder--but that's pretty rare). Another source of air, that I've never encountered, has been leaking hydrogen coolers. Air can be liberated from the water that leaks into the generator casing. But, you should have generator casing liquid level alarms if that's happening.

One BIG problem with most GE-provided hydrogen-cooled generators provided with gas turbines is that there is only one Seal Oil Flow-rate "gauge" (flow-rator I think it's called) for BOTH ends of the generator. If it was my site, at the earliest possible time I would change that to be able to have individual flow-rators for each end of the generator. And, if possible, I would try to get the flow-rates into the Speedtronic somehow (with 4-20 mA inputs) to trend the flow-rates, because, again, even with two flow-rate indicators trying to get operators to take readings is like pulling teeth.

The last thing you need to know is that as the seal rings get older, it's very likely the scavenging flow-rates will have to be increased. They are not "set them and forget them" as is widely believed. High generator vibration can also increase seal oil flow-rates, and the scavenging will also have to increase as well.

The topic of maintaining hydrogen purity has been covered many times before on control.com. Cleverly hidden at the far right side of the control.com Menu bar is a 'Search' field. While not intuitively obvious, it is quick and powerful. (HINT: Use the 'Search' Help to get started. HINT: Try the search term "hydrogen purity" just as shown.) Going through search results can be trying, but most people who do so actually learn more than they were intending to, thereby expanding their knowledge more than they otherwise might have.)

 

Doug Toner,

My bad. You DID say the scavenging rate was the same for both ends. My apologies for mis-reading the original post and missing the scavening rate into for the TC and CE.

But, my confusion is that you are saying the casing purity is decreasing but you don't say what is happening to the TE and CE purities.

All of the F-class hydrogen-cooled generators I have worked on only had two hydrogen purity analyzers (I think three is an option which could have been purchased, or possibly installed during/after commissioning), so it's necessary to manually switch one of the TE or CE hydrogen purity analyzers from it's seal drain enlargement tank to the generator casing to sense casing purity. Is this the case at your site? Could it be a problem with not purging the tubing properly? Could there be liquid (oil or water) in a low point in the generator casing purity sensing tubing? Some analyzers are adversely affected by moisture in the gas passing through it.

In my estimation, if the TC & CE purities are relatively constant but the casing purity is decreasing, then I would say the problem is either a poor job of purging and charging was done the last time the casing was purged and charged (not likely, except if the generator is not operated and is not on cooldown for long periods of time), or the purity of the hydrogen being used to maintain casing pressure is not as high as believed.

If the casing purity problem started after a new batch of hydrogen bottles was received and started being used, because I believe the hydrogen from the bottles to replenish what's being lost to scavenging is admitted to the casing. Scavenging on the TC & CE is typically done on the hydrogen-side seal drain enlargement tanks, and hydrogen is not added to the generator casing through these enlargement tanks.

Again, atmospheric air can't leak into the generator because the generator is basically at two times atmospheric pressure. So, that leaves:

1) Air liberated from entrainment in seal oil (usually doesn't make it's way into the generator casing);

2) Low purity of the hydrogen being used to replenish that lost to scavenging;

3) Seal oil spilling over into the generator casing (it would be reasonable to assume that this would cause high generator casing liquid level alarm(s), unless the generator casing liquid level switch(es) is(aren't) functioning or are not properly valved in).

4) Air entrained in cooling water leaking from water-to-gas heat exchangers in the generator (there are generator casing drains which can be opened (with due caution) and any liquid flowing out of the drain captured and examined; and it would be reasonable to assume that water leaking into the generator casing would cause high casing liquid level alarms, unless the generator casing liquid level switch(es) is(aren't) functioning or are not properly valved in).

That's about all I can add. Again, my apologies for mis-reading the original post and missing the scavening rate into for the TC and CE.

Please write back to let us know what you find.

 

As H2 pressure > atmospheric inside the casing, air cannot get in but H2 leaks out. Typically, Casing purity is higher (by as much as 2 or 3%) than CE and TE purity.

Have you tested the Casing purity with a portable analyzer to confirm the readings of the H2 panel?

Could it be that the tubing is mislabeled or has been connected to the wrong port?

When casing purity starts to fall, CE and TE purity is usually below alarm set points.

 

What brand of H2 analyzers are you using? We use E-One and they are very accurate but need to be cal'd monthly or so. the units will increase scavenging if they detect low purity. tell your boss that 95% is ok. perhaps your seals are leaking, as someone else suggested. usually the case, from my experience.

 

I'll state, I have never worked on a scavenging design, however, from what I've read,

I though the seal cavity purity was to be maintained at ~85%. with scavenging out the seal cavities, this should maintain the purity in the casing near the H2 replenishing purity.