We have a GE Frame 5 M combustion turbine that is having lube oil problems. When on the A/C lube oil pump the pump discharge pressure is 22psi. We are loosing 6psi across the heat exchanger. We loose another 1psi across the filters. That leaves our header pressure at 13.5 psi. We have a single cooler setup. Currently we are operating 1.5 psi above alarm on startup. When the gear pump kicks in the lube oil system is fine. Is it ok to continue to start the unit at 13.5 psi?

 

I'm pretty sure the Main (Access. Gear-driven) L.O. Pump is a gear-type positive displacement pump, and the pressure is usually set using a relief valve (yes; a relief valve) but I don't remember the device number of the relief valve.

The Auxiliary (AC motor-driven) L.O. Pump is usually a centrifugal pump, and it's pressure is what it is. The L.O. Header pressure is set by a pressure regulating valve (and, sorry, I forget the device number for that one, too; I don't have any older Piping Schematic drawings or Device Summarys, just some damned F-class junk).

I believe the discharge of both pumps goes through the cooler then the filter (yes, GE filters "cold" L.O., because they've had debris collect in the cooler downstream of filters and when it dislodged into the bearing header it caused wiped bearings, so they filter cold L.O. to try to prevent debris from getting into the cooler) then the pressure gets regulated by the bearing header pressure regulating valve.

If you're getting a big pressure drop across the cooler then I'd say something's wrong, but I'd expect the pressure drop to be virtually the same regardless of which pump is providing the flow/pressure. Are you actually measuring the pressure drop across the cooler with an inlet- and outlet pressure gauge or is there a dp gauge across the cooler?

A 1 psi drop across the filter isn't bad, but ... I've seen filters opened that are found to have been ruptured and the debris and filter pieces traveled downstream into the cooler, fortunately no bearings were damaged. (The L.O. Filter dp had reportedly been at or near zero psid for years (this on a peaker unit) and so they just opened the filter canister "on a whim". Good thing, too, eh?)

You should be able to find an Aux. L.O. Pump curve in the Maintenance Manual, or, from the pump manufacturer. GE used Buffalo Pumps for many years, and they were always most helpful when called for assistance. Just have the nameplate data ready when you call.

As far as "safety", if you need the power, then you gotta run it, right? And, if it trips on low L.O. pressure, then it trips. You just gotta hope the Emer. L.O. Pump is working properly if needed.

But, something really doesn't seem quite right with this. It sure seems the Aux. L.O. Pump isn't working up to snuff. Have you used a clamp-on on the three phases of the Aux. L.O. Pump Motor? How close to rated is the current draw? Are the phase currents balanced? If the current draw is pretty low, then one has to believe the pump isn't moving much oil. Because if the L.O. Bearing Header Pressure is good when the unit is running on the Main L.O. Pump (which, admittedly, is a positive displacement pump) then it would sure seem somethings wrong with the Aux. L.O. Pump. Again, because, I believe, the flow from both pumps goes through the cooler and the filter then the Bearing Header Pressure Regulator.

Lastly, the Main L.O. Pump being a positive displacement pump, usually builds pressure pretty quickly during acceleration, so it should be helping to raise the Bearing Header Pressure pretty quickly after break-away from zero speed (say around 50-100 RPM). The pressure might not be really high, but there certainly is flow (again, because it's a positive displacement pump).

Have a look at your L.O. Piping Schematic drawing and your Device Summary drawing, and check the Aux. L.O. Pump Motor current draw against the motor's nameplate (and the pump nameplate), but it sure seems something not right with the Aux. L.O. Pump output.

Please write back and let us know what you find. Feedback is what makes the threads so useful on control.com. When others can see if the information was useful, or a complete miss, it helps everyone!

 

We sent the AC motor for reconditioning. The key lube oil system pressure test was performed with the AC motor-pump operation only with the GT unit not operating. The test was conducted with the lube oil/cooling water heat exchanger U tube bundle installed and also another test was conducted with the bundle removed. The Delta P across the heat exchanger varied from one psig without the bundle to six with the bundle installed. Several OEM organizations and vendors have communicated to us that the six psig Delta P is excessive for this heat exchanger. Most estimates are in the one to one and a half range. The bundle has been inspected without evidence of fouling and the exchanger was vat cleaned.

Due to the system design with the twenty psig discharge of the AC motor pump, a six psig Delta P across the exchanger results in a mininum header pressure of 13.5 psig with newly installed lube oil filters. This header pressure is only 1.5 psig above alarm and the pressure switch is unable to reset (reset pressure - 16 psig plus/minus one psig). We are measuring the pressure at the inlet to the filter assembly and the internal exchanger oil pressure at the inlet end.

Please provide any help on similar exchanger situations. Our data and system design identifies the bundle as our most problem cause. Merci beaucoup for your help.

Greengator in SE Louisiana

 

This must be an unusual pump, because I'm more accustomed to seeing pump discharge pressures in the range of 40-70 psig, usually around 60 psig or so.

The motor might be perfect, but it sounds like there might be something wrong with the pump.

It does sound like the differential across the heat exchanger is somewhat excessive, but you say you've examined the tube bundle and it seems okay.

So, one has to wonder what the differential across the heat exchanger is with only the Main L.O. Pump running (at FSNL, for example).

Again, we don't have the Piping Schematic from your site, so my description of the system might not be correct.

As I recall, there are check valves in the discharge of both pumps, and each check valve has a small hole drilled in it to keep it "pressurized" and "primed" (from the discharge of the other pump). One has to wonder if there's something wrong with the check valve in the discharge of the main pump that's allowing leakage of Aux. L.O. pump discharge flow/pressure to flow through "unchecked".

Or, if there's some other piping problem or obstruction. The description does seem to point to the dp across the heat exchanger tube bundle.

Do you know what the Aux. L.O. Pump discharge pressure was before the low pressure problem was noted?

Again, the Aux. pump is a centrifugal pump (usually!), and the internal clearances might be worn. It just might be flowing enough oil to keep the pressure up with any elevated dp across the heat exchanger, which could make the dp appear even worse.

You didn't answer the question about phase currents versus pump nameplate rating or even motor nameplate rating. Low currents would indicate low flow.

I haven't seen a problem like this. But, one thing I've learned troubleshooting problems over the years: It often turns out to be more than one problem, and sometimes symptoms can be very misleading.

 

CSA... "Low currents would indicate low flow."

Only if supply voltage is nominal or lower!

Regards, Phil

 

well CSA is right. your pump will not be 22 PSI. please check your stats again. the normal lube oil header pressure is around 1.6-1.8 kg/cm2 for all GE machines (22-26 psi). are you sure you are confusing the lube oil header pressure with the pump pressure. normal AOP/MOP pressure is about 4.6kg/cm2 (66psi).

well your 1psi drop across the filter is very nominal/low value. across the cooler, it is slightly higher, but the problem can be easily solved.

There is a bearing header pressure regulating valve (tag no - VPR2, you will find this in your lube oil axillary PID) which can be adjusted to set the desired lube oil pressure in your circuit.

this will most probably solve your problem.

 

The originator did say that this was a MS5001M, so it is an old machine and maybe has a "piggy-back" AC & DC Lube pump. Can this be confirmed? Can we see a copy of the L.O. P&ID ? If it is a piggy-back system, it does work a bit differently from the later systems with a full size AC Pump.

 

Bob,

i have not worked on older machines but i have a small doubt here. Why i said with surety that the pump discharge pressure will be more than 22 PSI, is because the AOP/EOP/MOP also supplies to the trip oil circuit and the trip oil circuit has a nominal pressure of around 4-4.5 kg/cm2. Thus to supply this pressure the pump would have to have a higher pressure rating. Does old machines have a trip oil system which has a low pressuring rating? has anyone here seen such a system?

 

Anyone,

This is a really old post but this is reflecting what's happening with our system. Frame 5 with a a piggyback AOP/EOP running about 22 PSI. we're getting about 12 at the bearing header which isn't enough to reset the low lube oil alarm (16 PSI) until the MOP starts running. Did sending off the pump help?

 

Hello

How were the things working before ...that Lube oil header or pump discharge fallen out ...?

Can we see LO P&ID/Schematic/drawing of the unit?

What you mean by" sending off the pump" do you want tp force /unforce an Alarm(s) OR LOGIC Signal name..

Thank you for clarifying these points so that we can provide a better assistance from Here ( and it is for free of charges !)

 

Hello

How were the things working before ...that Lube oil header or pump discharge fallen out ...?

Can we see LO P&ID/Schematic/drawing of the unit?

What you mean by" sending off the pump" do you want tp force /unforce an Alarm(s) OR LOGIC Signal name..

Thank you for clarifying these points so that we can provide a better assistance from Here ( and it is for free of charges !)

No one knows how it was working before since it didn't get much run time. The usual practice is to start it twice because it'd trip on low oil pressure the first attempt. This unit is old and didn't get much use until the last few summers.

Reading 22PSI at pump discharge located next to motor on the diagram. Reading 12 at the PT located by the 63QN,63QL and 63QA switches. Changed filters yesterday as were were only getting 8.5 PSI at the PT located near the previously mentioned switches. 63QA is the switch for low oil pressure. It alarms when the turbine is over 1020RPMs and won't reset until the Oil pressure is over 16PSI per specifications in the book. Will set at 12 PSI decreasing. It's in until we get on the main shaft pump.

When ramping up to speed there is a drop of 1-1.5 psi at the exciter end bearing between 0 and 500 RPM. Then it comes back to 8.9 PSI. This is a transmitter that was installed a few years ago. This is not the same pressure that is driving the 63QA that's in question. This is just another oddity that happens on this machine.

"sending off the pump" was in reference to an earlier post in this thread by JJR. Who was experiencing similar problems and sent off their AC pump for reconditioning. We are trying to get away from forcing/ignoring/cheesing the alarms to make it through startups.

Thanks

 

Thank you for this updating !

Well the file that you shared looks more like a draft than P&ID ( old fashion way ) but can be a good beginning for troubleshooting that issue..

I will review this Draft and come back soon with some quotes..

 

I don't think I ever recall seeing a 63QN.... Can you tell us what it does? What are the setpoints for 63QN?

What control system is in use on the machine, please?

The P&ID says the discharge of the piggy-backed motor-driven L.O. Pump is 20 psi.

Was this machine a black-start machine (meaning does it have a diesel engine starting means--or a diesel to supply power to an electric motor starting means)?

I will go so far as to say the reason the unit trips on the first start attempt is because the L.O. is "cold" (at least it's colder than the second start attempt--just do simply to circulating the oil it warms up a few degrees). Myself, and other posters to the forum in the past, have always believed the temperature setpoint for starting the L.O. Tank immersion heater was way too low. The typical start setpoint was about 50 deg F (if I recall correctly--and it's been a long time) and the deadband on the switch meant it picked up at around 62 deg F (a 12 deg F deadband). I've seen some switches set at 60 deg F drop-out and pick-up and around 72 deg F. BUT, for a typical 30W turbine oil it's not uncommon for it to contribute to high vibrations if the oil temp is below around 80 deg F during a start (on many older Frame 5s, anyway)--which meant that in winter months the oil NEVER got warm enough to prevent the intermittent trip on high vibration during start-up (or the terrifying high vibration alarm during start-up). I just wonder if the oil is cold that it takes so much energy to pump it that the pressure is lower when the oil is cold, and then when the oil gets a little warmer it doesn't take so much oomph from the pump and there's a little more energy left over to raise the pressure a psi or more....?

So, the point I'm trying to make with this point is that it hurts NOTHING to keep the L.O. warmer to prevent nuisance problems. It will cause a little more electricity consumption but what's that worth when it's important the unit gets started and on-line in a timely manner with few issues?

Anyway, it hurts nothing to make sure the pressure regulating valves (VPR-1 and -2) are working correctly. The orifice at the top left of the P&ID is also pretty critical--and could be gummed up or worn.

AC motors do deteriorate over time, and their ability to produce nameplate-rated power can be reduced.

Pumps also can get worn over time....

It bears repeating--while often these kinds of issues are the result of one single factor, many times they are the result of myltiplc factors contributing to what looks like one cause is responsible. Older machines like this, with marginal components (pumps which produce slightly less than required pressure/flow) can be difficult to troubleshoot. Old components and gummed-up 1/4-inch tubing lines, valves and fittings can also be problems.

It would be great if we could hear back from you on your progress in resolving the problem.



I have to ask if the L.O. pressure switches have recently had their calibrations verified--and more importantly, if they've been replaced. I've seen gummed up oil sensing lines and pressure switches cause all manner of problems--especially on older, infrequently operated machines. Flushing the lines in the Access Gage Cabinet can be labor-intensive and messy (if people are in too much of a hurry and don't use proper techniques for tightening and loosening compression fittings. I've also seen improper value line-up on older gage cabinets cause problems, as well as sticking (sticking open) Dragon valves.

Has anyone checked the speed of the AC motor--and the DC motor?

Has anyone checked the coupling between the motors and the pump shaft? Loose set-screws or worn coupling inserts can also cause problems (some of those older pumps used nylon inserts in the coupling which would fail after 20 or 30 or 40 years...), or rubber coupling which can deteriorate the rubber.

I'm also not familiar with the two pressure regulating valve set-up shown in the L.O. P&ID. There's a check

 

I don't think I ever recall seeing a 63QN.... Can you tell us what it does? What are the setpoints for 63QN?

What control system is in use on the machine, please?

The P&ID says the discharge of the piggy-backed motor-driven L.O. Pump is 20 psi.

Was this machine a black-start machine (meaning does it have a diesel engine starting means--or a diesel to supply power to an electric motor starting means)?

I will go so far as to say the reason the unit trips on the first start attempt is because the L.O. is "cold" (at least it's colder than the second start attempt--just do simply to circulating the oil it warms up a few degrees). Myself, and other posters to the forum in the past, have always believed the temperature setpoint for starting the L.O. Tank immersion heater was way too low. The typical start setpoint was about 50 deg F (if I recall correctly--and it's been a long time) and the deadband on the switch meant it picked up at around 62 deg F (a 12 deg F deadband). I've seen some switches set at 60 deg F drop-out and pick-up and around 72 deg F. BUT, for a typical 30W turbine oil it's not uncommon for it to contribute to high vibrations if the oil temp is below around 80 deg F during a start (on many older Frame 5s, anyway)--which meant that in winter months the oil NEVER got warm enough to prevent the intermittent trip on high vibration during start-up (or the terrifying high vibration alarm during start-up). I just wonder if the oil is cold that it takes so much energy to pump it that the pressure is lower when the oil is cold, and then when the oil gets a little warmer it doesn't take so much oomph from the pump and there's a little more energy left over to raise the pressure a psi or more....?

So, the point I'm trying to make with this point is that it hurts NOTHING to keep the L.O. warmer to prevent nuisance problems. It will cause a little more electricity consumption but what's that worth when it's important the unit gets started and on-line in a timely manner with few issues?

Anyway, it hurts nothing to make sure the pressure regulating valves (VPR-1 and -2) are working correctly. The orifice at the top left of the P&ID is also pretty critical--and could be gummed up or worn.

AC motors do deteriorate over time, and their ability to produce nameplate-rated power can be reduced.

Pumps also can get worn over time....

It bears repeating--while often these kinds of issues are the result of one single factor, many times they are the result of myltiplc factors contributing to what looks like one cause is responsible. Older machines like this, with marginal components (pumps which produce slightly less than required pressure/flow) can be difficult to troubleshoot. Old components and gummed-up 1/4-inch tubing lines, valves and fittings can also be problems.

It would be great if we could hear back from you on your progress in resolving the problem.



I have to ask if the L.O. pressure switches have recently had their calibrations verified--and more importantly, if they've been replaced. I've seen gummed up oil sensing lines and pressure switches cause all manner of problems--especially on older, infrequently operated machines. Flushing the lines in the Access Gage Cabinet can be labor-intensive and messy (if people are in too much of a hurry and don't use proper techniques for tightening and loosening compression fittings. I've also seen improper value line-up on older gage cabinets cause problems, as well as sticking (sticking open) Dragon valves.

Has anyone checked the speed of the AC motor--and the DC motor?

Has anyone checked the coupling between the motors and the pump shaft? Loose set-screws or worn coupling inserts can also cause problems (some of those older pumps used nylon inserts in the coupling which would fail after 20 or 30 or 40 years...), or rubber coupling which can deteriorate the rubber.

I'm also not familiar with the two pressure regulating valve set-up shown in the L.O. P&ID. There's a check

63QN setpoint is 18 rising and and 14 decreasing. This switch is supposed to turn off the Emergency Lube Oil if it turns on with 63QA. We use 63QA to start and stop the DCLOP. Oddly enough our system doesn't use this switch for this purpose. I have my doubts it's even logically connected to anything at this point.

Siemens T3000 (Function Block Control) is used for the control system.

Yes the Lube oil says 20 PSI and produces 22 PSI. It pulled about 110% of FLA when the Lube oil was 60 degrees. Amperage will probably drop when it gets up to temp.

I know why the unit trips. It's getting less than 5lbs at the last bearing. During summer when it's 110 and winter when it's 40 degrees. We've seemingly fixed this by changing out the oil and the filters. My concern is we have 22 PSI at the discharge of the pump and 12 psi after the coolers and the filters. Also it's not a 30MW it's a 19.5 MW

VPR-1 maintains at 66 PSI and is working correctly, VPR-2 maintains the oil at 25 PSI and operates correctly.

The orifice at the top left is in the lube oil tank. The company doesn't want to pull oil out to check it.

All switches get checked annually and were verified numerous times through the operation. Gauges while not calibrated annually were calibrated during the troubleshooting that took place recently.

The AC motor is running at 1770 RPM and the DC motor is running at 1800. I don't know the current of the DC motor.

We've replaced the faulty gauges that we found and calibrated the ones that were capable of being calibrated against a known and calibrated standard. I'm pretty confident in the accuracy of the gauges. Should we be dropping 10 PSI across the cooler and the filters? We were dropping 14 PSI across everything but changed the filters and got it to a 10 PSI differential. But that isn't enough to turn off my low lube oil alarm.

 

The chances of finding someone who knows much about or has any experience with a unit this old are somewhere between slim and zero. It's been a long time since that unit had a Speedtronic on it, so no one knows how "well" it worked with a Speedtronic on it (there were LOTS of units from that era which had non-standard settings and configurations after field experience showed that the factory settings just weren't working reliably). The fact that 63QN isn't even used in control and protection says something about prior issues--OR the programming and configuration of the T3000.

Along this line of thought, newer GE Frame 5 heavy duty gas turbines usually only have VR-1 to regulate the Main (Accessory Gear-driven) L.O. pump discharge pressure, and VPR-1 to regulate bearing header pressure. The Aux. L.O. pump is centrifugal and doesn't require any pressure regulation on its discharge. Again, GE controls philosophies changed over time and with experience, and this unit is very old.

Personally, I would say that a 10 psi drop across (clean) filters and the L.O. cooler seems high by today's standards. But, that's a small cooler by today's standards. One of the reasons GE cools L.O. before filtering it is that coolers are KNOWN to be collectors of dirt and debris and if a large chunk of it is loosened for some reason and gets into the bearing header (if the filter is BEFORE the cooler instead of after it) that can problems, ugly problems. I think the PT designations on that drawing mean Pressure Tap--so that means one could install gauges to check for actual pressures, and it looks like there's one on the inlet to the cooler and one on the inlet to the filter (pressure taps, that is).

You say the generator collector-end pressure switch(es), 63QT, only see(s) less than 4 psi when the AC motor is driving the centrifugal L.O. pump. What is the pressure at the collector end when the Main (Accessory Gear-driven) L.O. pump is supplying pressure and flow to the system? The Main (Accessory Gear-driven) L.O. pump is a positive displacement pump (hence the reason for a relief valve to regulate the pressure coming out of the Main L.O. pump), and usually it doesn't take very many RPM of the turbine shaft and Access. Gear to produce good pressure flow from the outlet of the Main L.O. pump....

If the unit is tripping because the L.O. pressure at the collector end of the generator is less than 4 psi when the bearing header pressure is 22 or 20 psi that kind of points to some issue with the supply to the generator bearings OR the sensing lines to the 63QT pressure switch(es). I don't recall if units that old had redundant pressure switches at the collector end of the generator or not. And, just because the 63QT switch(es) work properly when being tested doesn't mean it(they) are sensing the actual pressure at the bearing--again, lots of the tubing used for sensing lines on GE heavy duty gas turbines is 1/4-inch schedule 80 tubing, so the ID is pretty small. Have you tried putting additional gauges on the L.O. supply piping to the generator bearings? (That's probably easiest on the collector-end bearing.) And, as shown at the top left of the L.O. P&ID there was often an orifice in the sensing lines upstream of pressure switches.

You say you know why the unit is tripping--but it's not clear WHEN it's tripping.... I presume it's probably tripping BEFORE the Main L.O. pump flow/pressure gets high enough to "overcome" the centrifugal L.O. pump discharge pressure. Is the logic not allowing the START command to be accepted because the L.O. pressure at the collector end of the generator isn't high enough (that's how it works on MOST GE-design heavy duty gas turbines with Speedtronic turbine control systems)? Is the turbine being tripped when the shaft breaks away from zero speed because the L.O. pressure at the collector end of the generator isn't high enough? When, exactly, is the tripping occurring in the START sequence?

What happens when the unit is on Cooldown to the L.O. pressure at the collector end of the generator? (Is Cooldown on a machine that old that some kind of ratchet mechanism, or what?)

Because the unit has a T3000 control system it's difficult to say for sure how it's programmed to operate. As I wrote, a START sequence is not permitted on a "newer" GE turbine control system UNLESS there is sufficient pressure at the furthest point away from the L.O. pumps--the collector end of the generator. The Master Protective logic (L4) will not pick up until the 63QTs are satisifed and actuated. There is no TRIP indication when the 63QTs aren't actuated prior to a START because a START can be initiated before the Aux. L.O. pump is started; if the START is initiated before the L.O. pump is started, the Master Protective won't pick up until the 63QTs are actuated. Then the START sequence is allowed to continue (presuming all the other permissives are satisfied).

On that machine, you may need to get a little "creative" with the L.O. pressure switches. I suggest you get some data of what the L.O. pressure does at the collector end of the machine during a START after the shaft starts turning (meaning the Main Accessory Gear-driven L.O. pump is putting out flow and pressure). If you REALLY only have less than 5 psi at the collector end of the generator prior to the shaft turning/breaking away during a START (and also during Cooldown) that's probably not a good thing. Sounds like you've checked everything other than the L.O. cooler.... If you've only got 12 psi after the filter/cooler then I don't think VPR-2 and the flow orifice are working correctly. Something is either restricting flow/pressure when only the AOP/EOP is running (which is disconcerting if you're really only getting less than 5 psi at the collector end of the generator--ESPECIALLY at low RPMs when the oil flow is needed most) or there's a measurement issue at the collector end of the generator. This just seems to be too low--AND, this isn't really a controls issue, because unless the T3000 is REALLY SPECIAL it doesn't control L.O. pump discharge or bearing header pressure.

I can't really tell from the small snippet of L.O. P&ID but it's possible there's an excessive flow condition downstream of the bit that's shown in the snippet (such as what feeds the Hyd. Pump) or the Control Oil system--that is not such a serious problem when the Main L.O. pump is putting out flow/pressure...? If the L.O. pressure at the collector end of the generator builds up as turbine shaft speed increases and is normal after a few hundred or a thousand PRM then something is amiss. And, if there's only less than 5 psi at the collector end of the generator bearing when the EOP is operating, that says there is marginally sufficient flow if the unit should trip from rated speed/load AND the Main L.O. pump suffered some kind of failure and couldn't provide oil flow/pressure as the unit coasted down, and the EOP might not either as the unit approaches zero speed.

It would be great if you could share what you learn as you resolve this problem. I remember hearing about this from a colleague a few years back over a steak dinner and some beers; and if it's never been properly resolved it's not trending in the right direction. Again, IF there is sufficient flow/pressure at the collector end of the generator during Cooldown AND when breaking away from zero speed and accelerating until the Main L.O. pump starts providing really good flow/pressure then it might need some custom programming/configuration. But, it really sounds like the place to start is to confirm precisely what pressure IS at the collector end of the generator during Cooldown and during breakaway and initial acceleration. It doesn't sound like management wants to pull the centrifugal L.O. pump, and they don't want to get inside the L.O. tank--but sometimes that's what is ultimately required. Again, until you know for sure precisely what the L.O. pressure is at zero speed and during initial breakaway and acceleration up to firing speed then it sure seems like someone is living on the edge.

Best of luck.

 

Thank you, your description of the cooler will give me enough information to have the company look into cleaning the cooler within the next few years.

 

I was once forced to perform a test of the Emergency L.O. pump system by shutting off the Aux. L.O. pump on a Frame 9E (large, five-bearing, 3000 RPM, 120 MW machine) while it was coasting down from about 1000 RPM. The Main (Accessory Gear-driven L.O. pump was still putting out sufficient pressure and flow until the unit reached about 35 RPM at which point the Main L.O. pump output was too low and the 63QTs started the Emer. L.O. pump. So, while that was a brand new machine, it is very similar in many respects to the Frame 5 at your site in that it has an Accessory Gear-driven Main. L.O. pump and uses a relief valve to limit the Main L.O. pump output pressure to protect the L.O. system.

The point I'm trying to make here is that once the shaft starts spinning the output of the Accessory Gear-driven Main L.O. pump should be overcoming the Auxiliary L.O. pump output pressure--and flow--and be providing sufficient pressure to the 63QA, 63QN, 63QL and 63QT(s) at the collector end of the generator.

What we don't know is when the unit is tripping--we don't know how the pressure switches are used in the programming of the T3000 to control and protect the turbine.

One would think--logically--that the Auxiliary L.O. pump should be capable of actuating the pressure switches when the unit is at zero speed and before the turbine shaft and Accessory Gear start turning the positive displacement Main L.O. pump. It sure seem that a 10-14 psi drop across the L.O. cooler and heat exchanger is excessive. You have said the VPRs (VPR-1 and VPR-2) are working correctly. I presume you make that statement based on the operation of the Main L.O. pump when the unit is at rated speed. It would be very helpful to know what the pressure drop across the filter and cooler is when the Main L.O. pump is the only pump providing L.O. pressure and flow to the system.

Again, I'm not familiar with the two VPR arrangement so I can't really comment on how it should operate at the "low" output pressure of the Aux. L.O. pump--but, still, logically one would presume all things being equal that it was designed to work similarly to systems of today. Unfortunately, the original Speedtronic elementaries for that machine are probably long gone, and we're not sure what Siemens used for that logic (it seems you have said that 63QN isn't even used for any control/protection logic).

There are all kinds of various video cameras (some with some very unusual names) which could be used to look into the L.O. tank and look at various piping and tubing fittings and see if it's possible to see any usual leaks which the output of the Aux. L.O. pump can't overcome by itself, but which the higher flow-rate of the Main L.O. pump can.

The thing you don't want to do is allow the unit to start with really low flows to the bearings furthest from the L.O. pumps (Accessory Base) during break-away and acceleration (at least until the Main L.O. pump can "take over"). AND, allow the Emergency L.O. pump to be blocked from starting on a coast-down on a black-out condition (meaning no Aux. L.O. pump) and wipe the bearings. It's kind of a tough situation.

There is one company you might contact which might have someone who can recall how those Model Ms worked and were originally configured--including the two VPR arrangement in the L.O. system. PAL Turbine Services, formerly Pond & Lucier Engineering (I think was the original name many, MANY years ago) in upstate New York. I see they are on Facebook and LinkedIn. If anyone can possibly help you, it's probably them--or they would know who might could help you.

At this point, I can't offer anything else. I would sincerely like to know how this turns out. But, it's really just because of that time I ran into my old colleague in Northwest Louisiana at a Texas Roadhouse and he asked me about this unit.

 

Changing the filters got us another few PSI at the header which put us at 9 PSI at the last bearing. All 3 Frame 5's we have Main shaft pump don't start overcoming the aux pump until 2100-2400 RPMs. The VPR's are set to operate at 25PSI and 66PSI with a relief set for 67PSI. The trip has been fixed. The lingering question is "Is this how the correct system pressures?"

I think the problem lies with the age of the gas turbine. Lack on knowledge and alot of hands in the logic through the years. Every start up a low lube oil alarm comes in. This comes in when the unit is above 1020 and not over 16 PSI. This alarm isn't present in the original 1969 control scheme and won't go away until around 2200 RPMS. Ten years of having alarms come in during startups, shutdowns, purge credits created a negative stigma about the lube oil system. This is just an alarm. It does nothing. The switch it is tied to turns on the AC pump on the decrease, that's it's purpose. I believe that no alarms should come in during startups/shutdowns. If they do they need to be fixed. I have concerns about the cooler and will push for cleaning in the future, so we can establish a baseline.

Siemens took whatever logic they found on the old system and put it in the new system. I'm reviewing it right now and there is a lot of code bloat and dead code.

Thanks for the information on PAL services. We made a lot of money this summer on these units and are under pressure to increase reliability.

 

Unless the Main L.O. pump is significantly different from newer machines (and GE isn't really like that--to change things up very much (one of the many good things about GE-design heavy duty gas turbines)) it certainly seems that the output of the Main L.O. pump is low--or there's a problem with VR-1 or VPR-1. Or there's a pretty significant leak somewhere downstream of the Main L.O. pump.

Here's where data is a good thing. It would be really good to know what the Main L.O. pump discharge pressure from break-away and all the way to FSNL. The 1020 RPM thing seems like it might be some kind of program setpoint, and it seems kind of arbitrary--without any other information. It would also be good to have that same information from the collector end of the generator. That would tell us a lot about what's happening and where. AND the same information from upstream of the L.O. cooler and downstream of the L.O. filter. And any visuals on VPR-1 and VPR-2 during a start to FSNL would also be a great thing.

PAL rebuilds--or used to rebuild--some pretty sophisticated governor oil pumps, and they probably have some information about that Main L.O. pump.