Feasibility of tripping a Gas Turbine on LUBE OIL TANK level LOW LOW


Thread Starter

Info Seeker

Am new to the forum and am trying to share some thoughts. We have nine gas turbines, eight frame 7E and one 7EA with MKV simplex/TMR control system. Per design, our GT units are configured for lube oil tank level ALARM as under:

- 7E frame GTGs are configured for LUBE OIL TANK LEVEL - HIGH / LOW ALARM (L71Q), i.e. one common alarm is generated in case of high or low lube oil tank level.

- 7EA GTG is configured for LUBE OIL TANK LEVEL - LOW ALARM (L71QL), i.e. a dedicated low level alarm is generated in case of low lube oil tank level.

The GT design permits unit trip through low lube oil pressure only, 63QT, and not through low or low low lube oil tank level. I want to understand why the OEM considers lube oil pressure only and does not take into account unit tripping through a low or low low lube oil tank level as additional protection, or a combination of both the low lube pressure and low tank level.

If anyone has any opinion in this regard, kindly share please. Thanks.
Info seeker,

Information is a good thing. GE's basic philosophy with L.O. Tank level is that it should be monitored by operators (newer units have, or can be equipped with, level sensors to provide graphical indication on a display of the tank level). Certainly, a low L.O. Level will ultimately result in loss of L.O. pressure which would ultimately result in a turbine trip.

If you look at the L.O. Schematic Piping Diagram, you should see that the suction of the main L.O. pump is above that of the emergency L.O. pump, such that if the tank level dropped low enough to result in a trip because of low L.O. pressure while the main L.O. pump was running, there should be sufficient volume for the emergency L.O. pump to provide lubrication during the coast-down period, to hopefully save the bearings and bearing journals.

GE is a fairly conservative company, which is another way of saying, "If it ain't broke they don't fix it". Decades ago, it was determined that the risk of low L.O. tank level was minimal, and back then, when instrumentation and wiring was much more expensive than it is now, the decision was made to not trip on low-low L.O. level. (GE even goes so far as to enclose L.O. supply piping inside the drain piping so that if the supply piping springs a leak, it's contained in the drain piping.) You can see the older machines used a single input which changed state on high- or low level; newer units used a separate contact for high level and another for low level (which only requires one additional wire and one additional contact input).

They are also under extreme pressure to keep the cost of their units down, so adding instrumentation is not in line with either way of thinking (if it ain't broke don't fix it, and keep the cost down). Because GE uses redundant sensors for any trip condition, it would mean adding more than one sensor to the L.O. tank level sensing scheme, to prevent nuisance trips caused by a single point of failure. And, while that doesn't seem very difficult, consider all the mechanical drawings which will have to be modified, all the manufacturing processes which will have to be modified, all the electrical drawings which will have to be modified, the new I/O points which will have to be added to the control system, the new logic which will have to be added to the control system, and the new displays and display modifications which will have to be made to the operator interfaces. It really is no small task.

Also, if care wasn't used in adding the instrumentation to the system (say, adding additional tubing outside the tank, which would mean exposing the tubing to possible damage, and more tubing means more tubing fittings that can leak, etc.) that tubing and its fittings could be the source of any leak and also cause nuisance trips if it did leak. They are pretty good about doing everything they can to minimize nuisance trips, forced outages, and keeping unit reliability very, very high. So, a lot of design review and analysis would have to be done to ensure that all these criteria were met.

Another initiative that GE is trying to implement is trip reduction. They are doing everything they can to try to reduce the number of nuisance trips on their units, and adding another possible trip condition doesn't fit that philosophy, either.

Now, as owners of the turbines you are free to implement whatever type of protection you deem appropriate and safe. So, if you want to trip the units on low-low L.O. level, you're free to do so. I would just recommend that you consider all of the above when designing the systems and adding the instrumentation to your units to try to keep nuisance tripping to a minimum and unit reliability high. I have seen companies replace multiple L.O. pressure switches (used to alarm and trip on low-low L.O. pressure) with a single pressure transmitter, and have just about every problem imaginable leading to lots of forced outages.

One owner added a single pressure transmitter because the operators wanted the L.O. pressure displayed on their screens so they wouldn't have to go out and read a gauge, and management said, we'll just eliminate those redundant switches and reduce our spare parts requirements, calibration requirements and I/O count. Well, the transmitter was a "low-cost" unit, and the diaphragms ruptured more than once. The tubing connections vibrated loose more than once. They reused the wires which were connected to the pressure switches, which weren't twisted, shielded pairs, and electrical noise caused problems with the readings. You name it; they had problems and it took them a couple of years to get around to fixing them all. They still have just one transmitter, but it's a higher quality (higher cost) device, and they have fixed their tubing problems and wiring problems. But, at a cost to reliability and availability while sorting through all the issues. And, it's still a single point of failure!
Dear Infoseeker,

We have 2 7EAs MKVI TMR. We are configured for alarms for low and high lube oil level, not trips. My understanding of GE's method for choosing alarm VS trip is this: Protect the unit, but maximize generation time. While a low oil tank level is definitely dangerous, if the bearing lube oil pressure is ok, then let the unit run. Given the nature of any single unit switch, like the lube oil tank level, I would hate to trip a unit offline for a bogus reading.

A point to note: we adjusted both switches on our tanks to go into alarm with the smallest change in level. As they were set from commissioning, it would have taken a loss of several inches of oil to go into alarm. If we ever have a leak it will be a mess, but hopefully less than it would have been if the switches were left where they were.

Hopefully this helps!!
Dear Sirs:
We have Frame 5 GTG but we got another issue the lube oil tank. We have high pressure alarm and trip, and in past month, our GTG tripped several time by high lube oil tank pressure. Does any one has met the same problem issue, what is the possible reason/

Dear Gary,

for frame-V machines, frankly speaking, I have never heard of anything like high pressure trip. I understand your machine would be having an oil demister system and with the fans in a lead/lag configuration, there would not be a condition related to high pressure since the hot mist generated in the tank is being sucked and released to atmosphere. High level is understood (though that also would be only for indicative purpuse rather than protection) but high pressure is unheard of. At the most, check L63QV1.
I, too, have never heard of a high L.O. tank pressure trip. That's not to say such a thing was never instituted, but I would really like to understand why it would be necessary. Must be something very unique, some kind of hazardous zone classification or something, I imagine.

Having said that, we don't know how the high pressure is being detected, or when this problem began. Such as after a maintenance outage when devices were calibrated, or ????

What have you done to try to troubleshoot the problem?

Also, is it a pressure switch (a single (?!?!?!) pressure switch?) that's tripping the unit? Or a pressure transducer (a single (?!?!?!) pressure transducer?)?

Has the calibration of the device(s) or instrument(s) initiating the trip been verified? Has the wiring been verified?

As MOB says, there must be some kind of "vent" on the L.O. tank, usually through some kind of demister (to remove L.O. vapors and prevent them from being discharged into the atmosphere), though older units didn't even have demisters of any type. But, I've never heard of a "sealed" L.O. tank; it must be vented somehow one would imagine whether it's through a demister or not.

Demisters with fan(s) (I've never seen a demister with redundant fans, either, but that doesn't mean there aren't such set-ups) usually are set to maintain a slight negative pressure at Base Load (and the pressure would be even more negative at less than Base Load).

But tripping on high pressure seems extreme. Most L.O. Tanks do have rupture disks to prevent over-pressurization of the tanks, but I've never seen a unit with any kind of L.O. tank pressure "monitor" connected to the turbine control system.

In any case, you need to find out how the high pressure is being detected (pressure switch(es) or pressure transducer(s); determine if the device(s) or instrument(s) (because it's not common to use a single device/instrument to initiate a turbine trip) are calibrated and working correctly; determine if the wiring from the device(s)/instrument(s) is intact (usually, a pressure switch is wired such that an open circuit will initiate a trip, so if there's a loose connection or wire crimp or something similar it will appear as an open circuit); and, lastly, you need to determine if the input circuitry to the turbine control system (you didn't say what kind of turbine control system is being used on the unit) is working correctly.

When you write for help, you should tell us as much about the condition you are experiencing as possible. When the condition started, what "changed" just prior to the condition starting, what you've done to troubleshoot the problem, important information about the control system in use on the turbine (manufacturer/supplier; SIMPLEX or TMR or ?; generation (Mark IV, Mark V, etc., if the control system is a GE Speedtronic; etc.). The more information you can provide, the better and more concise the response you will receive.

Personally, I'm really curious about the site and what the turbine is doing (driving a generator or a compressor or ????) that would necessitate a high L.O. tank pressure trip. Please, at least, let us know something about the turbine/application at your site.