Control.com - Nerds in Control

On site we have a 324 GE generator hydrogen cooled. 8 months before GE made a MAGUC inspection when it removed the end shield and seal ring from collector end. During the last month we have experienced sudden drop of H2 purity while the unit starts up and mainly when it synchronizes. From 98% we go to 93% during the period from 2700 RPM to synchronization. It is also interesting that during the last two outages we have the lube oil pumps stopped along with turning gear and the aux seal oil worked to keep the dif pressure. Is there any suggestion for the sudden drop of H2 purity? From the time that the H2 pressure is higher than the oil we can't have entrapped air. We also don't have oil drainage to the seal oil enlargements drain pots, but bear in mind that due to a mistake last outage 8 months before a lot of oil was inserted in the generator. If we don't have air or oil contamination then it should be something from the generator itself (e.g. dust from fretting?) that creates the contamination.

After the unit synchronizes, what happens to the hydrogen purity? Does it begin to increase? Does it increase up to the 98% value and then remain there during operation and shutdown and turning gear operation?

The only way that air can make its way into a generator which is at 2 barg (twice atmospheric pressure, or, 30 psig) is by entrainment in the Seal Oil. The medium for Seal Oil is Lube Oil and Lube Oil is in contact with air in the L.O. Tank, the drain returns, etc. So, the hydrogen sides of the generator seals is how air gets released (liberated) from the Seal Oil (L.O.) into the generator. (Seal Oil is at approximately 5 psig above generator casing pressure.) As the Seal Oil is flowing through the seals, the air is released (liberated) from the oil into the area which should only contain hydrogen.

It's not clear seal oil drain enlargement flows could be blocked, but if that was done then it's likely that when the unit is started and the generator rotor fan starts circulating the gases (hydrogen and air liberated from the Seal Oil) inside the generator casing that the purity would decrease. What was done with the hydrogen scavenging lines from the hydrogen sides of the seals? The hydrogen purity sensors are purposely set (under normal valving set-up) to monitor the lowest possible hydrogen concentration, which should be the hydrogen side seal drain enlargements.

It's pretty common for oil to flood the generator when proper procedures are not followed during re-start after an outage (sad, but true). But, that's no reason to block normal flows and could lead to dangerous conditions inside the generator.

Hi CSA
After the gt synch, the purity drops steadily, not rapidly. This incident also started from the time that we changed the moisture indicators of the hydrogen purity analyzer, E-ONE DHCP.
When you say seal oil drain enlargements these are the CE and TE of the generator where the purity analyzer is measured.

You need to consult the hydrogen and CO2 piping drawings provided with the generator and then go out and locate all the devices and trace the piping and tubing and check the valve positions on the hydrogen control panel. The purity sensors can usually be set to monitor either casing purity or seal drain enlargement purity (Turbine End and Collector End). Do you which purity each sensor is set to monitor?

It's not clear how changing moisture indicators could cause a problem, even if they were not properly installed. Again, the pressure in the generator casing and hydrogen scavenging and purity measurement systems is twice atmospheric, which means that even if the sensors weren't properly installed hydrogen should be leaking out of the system instead of air leaking in.

If the purity decreases after the generator rotor is at rest for some period of time and then run up to rated speed, one would have to think that the scavenging system isn't properly set up. When the generator rotor starts spinning fans on the ends of the rotor circulate the gas(es) inside the generator. If air is being liberated from the Seal Oil and isn't being scavenged properly (you've alluded to some flows being shut off to some components; it's possible all the valves are not properly aligned as they should be) then circulating the gas(es) inside the generator will most likely cause the purity in the generator casing to decrease.

Hydrogen purity is nothing to be too casual about. Valve positions are critical. Locate all the drawings; review them until you are comfortable with them; this is *very* critical. Make sure you are looking at the drawing for normal operation (there are usually four to five drawings which look very familiar but are for different purging and charging conditions); locate all the valves and components on the generator and hydrogen control panel. Then make sure each valve is in the proper position. It's always a good idea to have two individuals on a site who are very familiar with the hydrogen generator, and can review the valve positions and configuration, and even check each other to make sure the valves are in the proper positions.

From your second post: "After the gt synch, the purity drops steadily, not rapidly...."

From your first post: "...From 98% we go to 93% during the period from 2700 RPM to synchronization...."

Let's try to be clear. You say the purity drops from 98% to 93% at 2700 RPM to rated speed (it's presumed that's 3000 RPM). Then you say that the purity decreases steadily after synchronization (while the unit is at rated speed, again presumed to be 3000 RPM). But, what does it decrease to? 93% or lower?

When does the purity increase up to 98%, during operation while at rated speed, or during shutdown while on Cooldown (turning gear)?

When posting for assistance, please include all relevant information, such as purity while at rated speed, purity when on cooldown, the minimum purity reached, when purity starts to increase, if any manual intervention is done to try to increase purity, what is rated speed (that's why GE uses percent of rated speed instead of RPM for the normal speed display; if you said the purity decreasesd from 98% to 93% as speed increased from 90% to 100%, it would be easier to understand what's happening as we are just guessing that rated (100%) speed is 3000 RPM).

Usually, when hydrogen purity drops below a certain level, the increased scavenging flow-rate solenoids are energized. Is this happening on your site? Could that be what's helping to restore the hydrogen purity when it drops? It's just not clear how the hydrogen purity gets back to 98% from whatever minimum it decreases to (which we're also not sure about).

We also need to know which area each hydrogen purity sensor is
monitoring: TE, CE, or casing? Are you telling us the casing purity is
decreasing, or one of the Seal Drain Enlargements, and if it's one of the Seal Drain Enlargements, which one? Specifically, what is the purity of each sensor indicating?

As you have noted, air by itself, can't enter a space that is at twice atmospheric pressure. The only way that air can get into the generator is when it's liberated from the Seal Oil (which you've noted is at 5 psig above casing pressure) on the hydrogen side of the seals. That oil, from the TE (Turbine End) and CE (Collector End), is supposed to be directed to the TC and CE Seal Drain Enlargements, which, typically are connected to the TC and CE hydrogen purity sensors. By monitoring the areas which should be at the lowest hydrogen purity, it's felt that problems with generator casing purity can be avoided. Usually, the Seal Drain Enlargements are vented to atmosphere through the scavenging lines, with scavenging flow-rates controlled by the needle valves on the Hydrogen Control Panel (through the scavenging flow-rate monitors).

It's common for Seal Oil flow-rates to increase over time, and increased flow-rates result in an increase in air being liberated from the Seal Oil. The scavenging rates must be increased to return the hydrogen purity(s) to the desired levels, though it doesn't usually take a lot of increased scavenging under normal conditions. Many people mistakenly believe that once the scavenging flow-rates are set, they are never to be adjusted again; this is false. They must be adjusted over time to maintain desired hydrogen purity, but any adjustment should be recorded in an operational log.

One of the *most* important variables which should be closely monitored on any hydrogen-cooled generator is the Seal Oil flow-rate. If I ever worked at a plant with a GE-provided hydrogen-cooled generator I would immediately put individual Seal Oil flow-rate indicators at the TE and CE, replacing the single, total Seal Oil flow-rate indicator. With a single, total flow-rate indicator one can see total flow-rate increase but one has no idea which seal, or if both seals, are consuming more Seal Oil. This information, total flow-rate, but more importantly individual flow-rates, is very critical when trying to troubleshoot hydrogen purity problems. Since air liberated from the Seal Oil is almost the only method for air to get into the generator, monitoring Seal Oil flow-rates is critical to anticipating problems and troubleshooting problems. (I did see one Southeast Asian installation that was getting impure hydrogen from their supplier; difficult to prove, but the supplier did finally acknowedge a quality problem.)

Can you tell us what the Seal Oil flow-rate was before the MAGIC inspection? What is it now?

I don't believe that a poor purge of air from the generator before charging with hydrogen would allow the air/CO2 to settle in the bottom of the generator casing if the unit were off Cooldown (turning gear) for some substantial period of time. It might be possible since both air and CO2 (used to purge the air from the casing prior to charging it with hydrogen) are both heavier than hydrogen, and then when the generator rotor starts spinning (fast enough) the air/CO2 is mixed with the hydrogen, decreasing the purity. But, it would seem unlikely--not impossible, but unlikely (in my opinion). Again, air at 0 psig can't flow into an area which is at 30 psig, so the only way that air can get into the generator is through the Seal Oil which is at a slightly higher pressure (5 psig greater) than casing pressure. Increased Seal Oil flow-rates are the usual reasons for decreasing hydrogen purity, and being able to detect which seal is using more oil is critical to troubleshooting. Also, monitoring (recording and reviewing) Seal Oil flow-rates is critical to troubleshooting.

Does the unit have a Hydrogen Dryer? When is the Hydrogen Dryer used? The piping to and from the Hydrogen Dryer must be properly purged before and after use of the Hydrogen Dryer. Is there a procedure for using the Hydrogen Dryer? Is it being used? This might be another possiblity, however unlikely.

Finally, are you having any indications of Hydrogen Cooler tube leaks? There is probably some air entrained in the Cooling Water, which might be another (again, unlikely) source of air. But one would think that it take a fairly substantial leak (or leaks) to allow enough air to accumulate in the casing.

But, it's not clear what area(s) the purity sensors are monitoring or what area each sensor is monitoring. There's also the problem of understanding what flows were stopped and how they could be stopped (easily). That's why it's important for you and another individual at your site to familiarize yourselves with the drawings and the locations of all the components and the valving, and ensure that all valves are in their proper locations. Also, it's important to review any Seal Oil flow-rate information (no, this is not usually monitored by the turbine control system; people have to read an indicator and record the values).

We can try to answer any questions you might have, and point you in the direction of possible causes; but you need to understand how the system is to be operated, and if it's being operated properly. I've been to sites where one particular "outside" operator, while making rounds during his shift, was continually reducing the scavenging flow-rates because he thought the unit was using too much hydrogen. He was not recording the adjustment he made or reporting that he was making any adjustment. He wasn't very familiar with the generator, how it maintained purity, how it was to be operated, or even what the expected hydrogen usage should be under typical conditions. He just arbitrarily decided that one bottle of hydrogen every three days was excessive, didn't consult with anyone about it, and took it upon himself to do something about the situation without informing anyone else. It wasn't until the turbine control panel annunciated a low purity condition and the increased scavenging solenoids opened and the hydrogen usage increased dramatically, did a supervisor go back through the logs and see that hydrogen purity had been decreasing for about three weeks and no one had done anything about it, not even notify a supervisor. It took several days to get to the bottom of the situation, and then only because the outside operator volunteered that he had trying to reduce hydrogen consumption did the whole story finally get out and understood.

Does your site log scavenging flow-rates? Can you say if the scavenging flow-rates have increased or decreased recently? I don't know many sites that do anything other than put a mark on the hydrogen control panel where the scavenging flow-rate indication was when the unit was new. Of course, over time if that flow-rate isn't adjusted to maintain purity, the purity will steadily decrease.

Operating a hydrogen-cooled generator and maintaining hydrogen purity and understanding how purity can increase or decrease is a very delicate balancing act, and requires accurate logs of flow-rates and accurate recording and reporting of any changes. Unfortunately, since most of these flow-rate readings are not monitored by the turbine control system and the values must be recorded by operators and those logs must be read by supervisors, nothing happens until the turbine control system annunciates a low hydrogen purity condition. Most operators, even if they are required to take readings, do so in a robot-like fashion and grudgingly. As long as there isn't any alarm, they're fine. When the alarm is finally annunciated, if anyone bothers to go back through the logs and discover that the Seal Oil flow-rates have steadily been increasing for three months then they will begin to understand they should have been increasing the hydrogen scavenging flow-rate to maintain purity along with the increasing Seal Oil flow-rates. That's even if there are logs to refer to.

CSA. I will try after a long time of absence to be more specific.

Usually when the unit is stopped we have a purity of 97%. When the machine accelerates to FSNL and then synchronizes we have a steady drop that finally drives purity to 95%. While the machine is running for 1 month continuously we see a drop until 92@. If we increase the low flow scavenging then we will drive the purity back to 97%.

we dont wait to let the automatic scavenging start, we interfere before with the low flow scavenging.

Sensors of purity analyzer measure for 55 minutes the purity pon the two ends and for 5 minutes they measure the casing. The numbers i give is from the lowest purity part the Generator TE. Casing is higher. BUT the drop is more or less the same in all measurements.

In the worst case, when we start eh manual low flow scavenging we have TE 92%, CE94% Casing 96%.
The main seal oil flow is staedy through the 2,5 years of plant commercial opeartion.
The hydrogen is 99,99999 purity.
Unfortunately we dont have individual seal oil flow rates.
The seal oil flow rate is the same before and after the MAGIC inspection.
The unit does not have hydrogen dyer and the seal oil is nto vacuum trated.
We had an issue with the coolers , a suspicion of leakage.

That is all i can say for the moment.
Something last , the problem started form the time that we changed the moisture indicatiors of the purity analyzer. I dont know and i dont think that this has anything to do with the purity drop but in any case i have to mention it.

From the previous reply:

"It's common for Seal Oil flow-rates to increase over time, and increased flow-rates result in an increase in air being liberated from the Seal Oil. The scavenging rates must be increased to return the hydrogen purity(s) to the desired levels, though it doesn't usually take a lot of increased scavenging under normal conditions. Many people mistakenly believe that once the scavenging flow-rates are set, they are never to be adjusted again; this is false. They must be adjusted over time to maintain desired hydrogen purity, but any adjustment should be recorded in an operational log.

One of the *most* important variables which should be closely monitored on any hydrogen-cooled generator is the Seal Oil flow-rate. If I ever worked at a plant with a GE-provided hydrogen-cooled generator I would immediately put individual Seal Oil flow-rate indicators at the TE and CE, replacing the single, total Seal Oil flow-rate indicator. With a single, total flow-rate indicator one can see total flow-rate increase but one has no idea which seal, or if both seals, are consuming more Seal Oil. This information, total flow-rate, but more importantly individual flow-rates, is very critical when trying to troubleshoot hydrogen purity problems. Since air liberated from the Seal Oil is almost the only method for air to get into the generator, monitoring Seal Oil flow-rates is critical to anticipating problems and troubleshooting problems....

Can you tell us what the Seal Oil flow-rate was before the MAGIC inspection? What is it now?"

It's simply impossible for Seal Oil flow-rates to remain constant over time. It's necessary to make adjustments to the scavenging flow-rates to maintain hydrogen purity as Seal Oil flow-rates increase, and they usually only ever decrease after seal rings and springs are replaced with new ones during some outage. It seems that you are increasing the scavenging flow-rates for some period of time, then returning them to their previous values. Have you just tried increasing the flow-rate slightly and monitoring the hydrogen purity, and continuing to increase the scavenging flow-rate by very small increments until the hydrogen purity stabilizes? This is normal operating and maintenance practice to adjust scavenging flow-rates to maintain hydrogen purity. Pure and simple; think or ass-u-me otherwise would be illogical.

Again, monitoring Seal Oil flow-rates and maintaining records of Seal Oil flow-rates and hydrogen scavenging flow-rates is the best method to monitor hydrogen consumption and hydrogen purity. Logs are critical, and, having Seal Oil flow-rate meters at each end of the generator is the best method of determining increased Seal Oil flow-rate at one end of the generator or the other.

It also seems that the data you are now reporting is slightly different from the original post....